Research Handbook on Soft Law 183910192X, 9781839101922

Citation preview

RESEARCH HANDBOOK ON SOFT LAW

RESEARCH HANDBOOKS IN LAW AND POLITICS Research Handbooks within this series provide state-of-the-art analysis of discrete areas of research that sit at the intersection of law and political science. Through exploring the legal aspects of political science and the impact of politics on the formation and application of the law this series acts as a platform for cutting-edge interdisciplinary research. The Research Handbooks explore legal and political science approaches to key topics and domains, such as: the actions of legal institutions, the application of the law, judicial politics, constitutions, global governance, law and political theory, democracy, diplomacy, and federalism, among others. Edited by leading scholars in their respective fields, volumes bring together authors with diverse expertise to provide rounded analysis of the substantive issue at hand. Due to the interdisciplinary nature of the series, the volumes within will not only act as valuable resources to those researching the nexus of law and politics but also provide new insights for those whose area of focus sits within either of the respective fields. The research presented will make significant new contributions to existing debates within the broadening study of law and political science whilst also acting as a starting point for fresh analysis. For a full list of Edward Elgar published titles, including the titles in this series, visit our website at www​.e​-elgar​.com.

Research Handbook on Soft Law Edited by

Mariolina Eliantonio Professor of European and Comparative Administrative Law and Procedure, Faculty of Law, Maastricht University, the Netherlands

Emilia Korkea-aho Professor of European Law and Legislative Studies, University of Eastern Finland, Finland

Ulrika Mörth Professor of Political Science, Department of Political Science, Stockholm University, Sweden

RESEARCH HANDBOOKS IN LAW AND POLITICS

Cheltenham, UK • Northampton, MA, USA

© The Editors and Contributors Severally 2023

With the exception of any material published open access under a Creative Commons licence (see www.elgaronline.com), all rights are reserved and no part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher.

Chapter 21 is available for free as Open Access from the individual product page at www. elgaronline.com under a Creative Commons Attribution NonCommercial-NoDerivatives 4.0 International (https://creativecommons.org/licenses/by-nc-nd/4.0/) license. Published by Edward Elgar Publishing Limited The Lypiatts 15 Lansdown Road Cheltenham Glos GL50 2JA UK Edward Elgar Publishing, Inc. William Pratt House 9 Dewey Court Northampton Massachusetts 01060 USA A catalogue record for this book is available from the British Library Library of Congress Control Number: 2023945118 This book is available electronically in the Law subject collection http://dx.doi.org/10.4337/9781839101939

ISBN 978 1 83910 192 2 (cased) ISBN 978 1 83910 193 9 (eBook)

EEP BoX

Contents

List of tablesviii List of contributorsix Acknowledgementsxii Introduction to Research Handbook on Soft Law PART I

1

DEFINITIONS, HISTORY AND CONTEXT

1

Bamboo, or governance through soft law: hybridity, legitimacy, and sustainability  Francis Snyder

2

Soft law: an historical introduction Nils Jansen

31

3

The definition of soft law Fabien Terpan

43

4

An economic analysis of soft law as a regulatory tool Michael Faure and Niels Philipsen

56

5

Anthropology and soft law Filippo M. Zerilli

71

6

Democracy and soft law Ulrika Mörth

87

PART II

10

DISCIPLINES, REGIMES AND AUTHORITY

7

Soft law in European public law Bruno de Witte

101

8

Better regulation as soft law Claudio M. Radaelli and Gaia Taffoni

116

9

Rules and responsibilities: business and social norms in transnational governance132 Boris Holzer

10

Soft law, technical standards and European private law Hans-W. Micklitz

v

145

vi  Research handbook on soft law 11

Soft authority in global governance Jan Klabbers

162

12

International standards and the dilution of responsibility Ingrid Gustafsson Nordin and Kristina Tamm Hallström

177

PART III ACTORS, INSTITUTIONS AND MAKING OF SOFT LAW 13

Soft law and courts: saviours or saboteurs of the rule of (soft) law? Mariolina Eliantonio and Emilia Korkea-aho

191

14

EU economic governance, agencies and soft law: an accountability challenge for the courts? Jacint Jordana and Joan Solanes Mullor

208

15

Fictions and fuzziness: soft law rule-making among the EU’s decentralized agencies Steven Vaughan

223

16

Administrative guidance in the United States: the moral and political stakes of non-binding law Blake Emerson

237

17

Soft law making at the European Commission: not much of a one-institution show Oana Ştefan

253

18

Beyond norm entrepreneurs: civil society and the framing of the ‘legal’ through soft law Rene Urueña and Rafael Tamayo-Álvarez

272

19

The Open Method of Coordination (OMC): a hybrid tool of political leverage in the making Minna van Gerven and Sabina Stiller

288

20

Studying the EU soft law cycle: the role of domestic factors Anne Ausfelder, Adam Eick and Miriam Hartlapp

304

PART IV CRISES, CHALLENGES AND CHANCES 21

Fighting a hard battle with a soft weapon: is international climate change law softening? Kati Kulovesi and María Eugenia Recio

22

Soft law in city regulation and governance Astrid Voorwinden and Sofia Ranchordás

337

23

Soft law and citizenship regimes Timothy Jacob-Owens and Jo Shaw

353

320

Contents  vii 24

Soft law: booster or brake for the promotion of gender equality in the EU? Birte Böök and Linda Senden

368

25

Soft law and the rule of law crisis Joelle Grogan and Clara van Dam

391

26

The role of soft law in the context of the financial crisis Alexander H. Türk

407

27

Soft law governance in the field of AI: a European perspective Frederik Schade and Mikkel Flyverbom

423

Index438

Tables

6.1

Three interpretations of liberal representative democracy

8.1

Hard and soft legal basis for IA and consultation

89 123

8.2 Consultation

125

8.3

Impact assessment 

127

21.1

Comparing mitigation-related provisions in the UNFCCC, Kyoto Protocol and Paris Agreement

331

23.1

Dimensions and sites of soft law

362

viii

Contributors

Anne Ausfelder is Policy Officer at the German Federal Cyberagentur (Agentur für innovation in der Cybersicherheit) Birte Böök is Assistant Professor, RENFORCE at Utrecht University Clara van Dam is a legal advisor at the Council of State of the Netherlands (Raad van State) and guest staff member at the Department of Constitutional and Administrative Law at Leiden University Adam Eick is Research Associate at the Geschwister-Scholl-Institute for Political Science at the Ludwig-Maximilians-Universität Munich Mariolina Eliantonio is Professor of European and Comparative Administrative Law and Procedure at Maastricht University Blake Emerson is Professor of Law at the University of California, Los Angeles School of Law Michael Faure is Professor of Comparative and International Environmental Law at the Law Faculty of Maastricht University and Professor of Comparative Private Law and Economics at Erasmus School of Law in Rotterdam Mikkel Flyverbom is Professor of Communications and Digital Transformations at Copenhagen Business School Minna van Gerven is Professor of Social Policy, Faculty of Social Sciences, University of Helsinki Joelle Grogan is Senior Lecturer in Law at Middlesex University London and Research Fellow at CEU (Democracy Institute Budapest) Ingrid Gustafsson Nordin has a PhD in Public Administration and is an affiliated researcher at the Stockholm Centre for Organizational Research (SCORE) Miriam Hartlapp is Professor of Comparative Politics of Germany and France at the Otto Suhr Institute of Political Science, Freie Universität Berlin Boris Holzer is Professor in Sociology at University of Konstanz Timothy Jacob-Owens is an Early Career Fellow in Citizenship Law and Policy at the University of Edinburgh and a Senior Research Associate with the Global Citizenship Observatory based at the European University Institute Nils Jansen is Professor of Law and Legal History at University of Münster Jacint Jordana is Professor of Political Science at the Universait Pompeu Fabra and at the Institut Barcelona d’Estudis Internacionals (IBEI) ix

x  Research handbook on soft law Jan Klabbers is Professor of International Law at University of Helsinki Emilia Korkea-aho is Professor of European Law and Legislative Studies in the Law School at the University of Eastern Finland Kati Kulovesi is Professor of International Law at the Center for Climate Change, Energy and Environmental Law (CCEEL) Law School at the University of Eastern Finland Hans-W. Micklitz is Professor at the European University Institute (EUI) Ulrika Mörth is Professor in Political Science at Stockholm University Niels Philipsen is Professor of Shifts in Private and Public Regulation at Erasmus School of Law, Rotterdam and Associate Professor in Law and Economics at Maastricht University Claudio M. Radaelli is Professor in the School of Transnational Governance at the European University Institute (EUI) and University College London (UCL) Sofia Ranchordás is Professor of Administrative Law at Tilburg University María Eugenia Recio is Postdoctoral Researcher in CCEEL, Law School at the University of Eastern Finland Frederik Schade is PhD Fellow at Copenhagen Business School Linda Senden is Professor, RENFORCE at Utrecht University Jo Shaw holds the Salvesen Chair of European Institutions and is Head of Edinburgh Law School (2022–). She is also a co-Director of the Global Citizenship Observatory based at the European University Institute Francis Snyder is Starr Professor of Law, EU Jean Monnet Chair ad personam, and Director of the Centre for Research on Transnational Law at Peking University School of Transnational Law, Peking University and Shenzhen Graduate School, College of Europe, Bruges, Emeritus Professor, CERIC at Aix-Marseille University and Adjunct Professor at the University of Macau Joan Solanes Mullor is Tenure Track Professor of Constitutional Law at Pompeu Fabra University Oana Ştefan is Reader in Law at King’s College London Sabina Stiller is Guest Researcher at the Faculty of Management Sciences, Radboud University Nijmegen Gaia Taffoni is Research Fellow and Teaching Associate in the School of Transnational Governance at the European University Institute (EUI) Rafael Tamayo-Álvarez is Professor of International Law in the Faculty of Law at Universidad del Rosario, Bogotá Kristina Tamm Hallström is Associate Professor and Lecturer in Management at the Stockholm School of Economics and Researcher at the Stockholm Centre for Organizational Research (SCORE)

Contributors  xi Fabien Terpan is Associate Professor of Public Law and European Studies, Jean Monnet Chair, Sciences Po, Grenoble, UGA and Deputy Director of CESICE Alexander H. Türk is Professor of Law at the Dickson Poon School of Law at King’s College London Rene Urueña is Professor in International Law at the Universidad de los Andes in Bogota Steven Vaughan is Professor of Law and Professional Ethics and Vice-Dean at University College London (UCL) Astrid Voorwinden is Researcher and PhD candidate at the University of Groningen Bruno de Witte is Emeritus Professor at Maastricht University Filippo M. Zerilli is Professor of Social Anthropology at the University of Cagliari

Acknowledgements

The editing of this volume has raised many interesting questions on regulation and governance. We are very grateful to our esteemed group of authors who agreed to participate in the quest to define and problematize soft law from various academic disciplines and empirical perspectives. As editors, we have been entrusted with the important task of deciding on how to organize this handbook, a task we have taken seriously and with great joy. The book is a result of a collective endeavour, but the mistakes remaining are only ours. We would also like to thank Laura Mann at Edward Elgar who encouraged us to take on this book project and who guided us patiently. In the final steps, Amber Watts took the reins and helped us finish the project that had grown bigger than anyone of us imagined in the beginning. Thanks are also due to Berfin Nur Osso and Markus Sairanen at UEF Law School who efficiently and with great care helped us edit the chapters. In Brussels, Helsinki and Stockholm, Mariolina Eliantonio, Emilia Korkea-aho and Ulrika Mörth

xii

Introduction to Research Handbook on Soft Law 1.

SOFT LAW: AN OMNIPRESENT CONCEPT IN CONTEMPORARY REGULATION

Soft law plays an important role in national, international, and transnational politics as well as the everyday life of public administration at all levels of government. Its role in these contexts is, however, difficult to appraise, as soft law comes in a multiplicity of forms. Guidelines, recommendations, codes of conducts, standards, policy guidance, technical notices – to mention just a few – are all captured by the notion of soft law. The dynamics of soft law in global and European politics suggest that the use of soft law in complex and sovereign-sensitive issues, such as migration and climate change, is a function of political necessity. The non-willingness by nation states to delegate decision-making authority to the EU and other international organizations often explains the utilization of soft law. Global agreements on climate change, migration, public health, gender policy goals, financial crime prevention and poverty reduction may entail a ‘false’ sense of how important issues are being substantially addressed and implemented when the agreements are merely reduced to scoreboards on global performance indicators. From a democratic point of view, such agreements that lack a clear chain of accountability raise questions about the possibilities for realizing democratic accountability. In the EU context, the use of soft law – especially if it is used instead of legislation – can undermine the role of national or EU parliaments and neglect the views of the public. Soft law does, however, have some clear advantages too. Soft law may be seen as more effective than hard law and formal legal sanctions which come with it. Regulators may opt to use soft law given that it is quick to adopt and amend to reflect changing contexts and technological advances. Soft law can also serve to open the door for the participation of (socially responsible) global companies and other stakeholders during periods of heightened need for global governance. For better or worse, soft law is an intrinsic part of regulation and governance at all levels of governance. The concept of soft law has existed for decades and has been tackled by different academic disciplines. Consequently, there are various understandings of the concept, reflecting the fact that the term soft law is used by scholars of international law, European law, international relations, and European politics. All of them approach it from their own scholarly traditions. Lawyers are interested in the role that soft law plays in adjudication and administrative decision-making, and whether legal actors differentiate between hard law and soft law. Social scientists study the design of institutions, public decision-making and how the use of soft law may undermine the role of parliaments and empower private actors. International relations scholars examine the force or ‘bite’ of soft law. Of course, many of the issues are common to different disciplines, further complicating the landscape in soft law research. A common feature in this research is how it problematizes and sometimes also challenges traditional images and perceptions within various disciplines on the dichotomy between law and non-law, public and private actors and between national, European and global levels of regulation and governance. From a conceptual standpoint, soft law functions as an ‘irritant’, encouraging 1

2  Research handbook on soft law researchers to venture into unchartered territories and pose questions that challenge the status quo.

2.

THE AIMS AND STRUCTURE OF THE RESEARCH HANDBOOK

The aim of this Research Handbook on Soft Law is to provide a scholarly, state-of-the-art overview of the research and scope of current thinking in the field, attempting also to envisage the future of soft law research. The chapters in the handbook show the empirical, theoretical, and analytical breadth and depth of the research on soft law. The handbook consists of four broad themes. The first theme covers the conceptual history and development of the notion of soft law, that is, the different concepts, meanings and historical narratives we associate with soft law as well as the different contexts in which soft law can be analysed. What does the notion of soft law entail and how has it changed over the years? Or has it changed at all? Can we talk of an all-encompassing concept of soft law, or should we rather accept the diversity of concepts that are linked to the empirical phenomenon of voluntary and non-binding rules/norms? What can certain methodological approaches, such as the economical and anthropological analysis of law, teach us about the concept of soft law? And how is soft law linked to other concepts commonly used to assess rules and norms and their legitimacy and democratic pedigree? The main finding here is that the development, history and uses of soft law in legal and social sciences research are diverse and complex. The chapters in the first part show how the concept and development of soft law is closely linked to the changing scientific dynamics and contexts in law, politics and history. In the first chapter of the handbook, Francis Snyder, reflecting on the past research and the needs of the (sustainable) future, boldly proposes renaming soft law as ‘bamboo’. In his words, bamboo ‘is part of human and natural life that is strong, fast-growing, highly visible, easily recognizable, resilient, and flexible’. The resilient nature of soft law is manifest from the fact that soft law is not a modern-day folly, something we have only come to know as the product of contemporary complex societies. As Jansen describes, ‘pre-modern jurists’ well understood that laws, while being a source of authority, had to be reconciled with other authoritative sources. In other words, these pre-modern jurists had a very modern idea of the law, which suggests that while soft law as a concept may not have changed, our views of it have. The other chapters of the first part illustrate that soft law is both a multifaceted empirical phenomenon as well as an important concept in the legal and social sciences to understand and explain regulation and governance. A clear definition of soft law has escaped generations of scholars, and Fabien Terpan fearlessly tackles this important question, suggesting that soft law should be approached from a perspective that both recognizes its position in between hard law and non-law as well as its functions with regard to hard law: pre-law, post-law and para-law. Michael Faure and Niels Philipsen also take up the definitional challenge, looking at soft law from an economic perspective. They demonstrate that law and economics research has used the public and private interest perspective to explain the emergence of soft law in particular contexts. They see more research to be done on the policy question of when soft law may bring about better results than hard law. Filippo M. Zerilli’s chapter adds an important legal anthropological perspective. But rather than using legal anthropology as a way to define

Introduction  3 soft law, he suggests that soft law is a refreshingly new field or a standpoint from which to approach the emerging transnational legal order and particularly the relations among state, supra-state, and non-state (private) forms of regulation. The chapter also briefly touches upon the question of how soft law can be examined through the adaptation of ethnographic methods. The conceptualization of soft law is the focus of the final chapter of the first part. Ulrika Mörth concludes that the democratic status of non-binding rules depends on whether and how these rules are ‘coupled to the liberal representative system of democracy’. Like Zerilli, she also wonders what happens to soft law beyond the nation state and in the era of global normative and legal pluralism. What emerges from the first part is indeed that the diversity of concepts of soft law linked to the empirical phenomenon of non-legally binding rules and norms seems an unescapable feature of the research on soft law. The second theme concerns the different disciplinary understandings of soft law, that is, how different academic fields understand and investigate soft law, as well as how these fields characterize and delimit various soft law regimes and their authority. What features characterize legal, political science and administration, international relations or organizations research into soft law? Do lawyers approach soft law from the perspective of actors (courts, regulators, standardization bodies) and social scientists from the perspective of institutions? Do private lawyers see the authority of certain soft law regimes, such as technical standards, differently than public lawyers? Does the EU legal and political scholarship understand the authority and effects of soft law differently than the scholarship interested in transnational and international law? One important finding here is that soft law can either be interpreted and analyzed as one instrument in the regulatory toolbox from which rational decision-makers can choose, or soft law can be seen as providing a forum for deliberation and learning. In the first interpretation, the strategic choices with respect to the pros and cons of soft law by decision-makers are highlighted, whereas the second interpretation emphasizes how deliberative social processes form new modes of governance and regulation. The chapters show how both interpretations are important in understanding how different disciplines analyze soft law and that soft law can be regarded as a continuum of rules rather a static concept. De Witte’s chapter examines the phenomenon of EU soft law from the general perspective of EU public law, by showing that the judicial and institutional practice of the EU departs from a categorical distinction between binding law and soft law. While soft law is a pervasive phenomenon in EU regulation, de Witte reminds that the constitutional principles of conferral and institutional balance need to be respected when soft law is used. Claudio M. Radaelli and Gaia Taffoni focus on the notion of ‘better regulation’ as an instance of meta-regulation, which typically – in the EU – comes in the form of soft guidelines, communications and peer-review processes. They show that the argument for the use of soft law in the field of better regulation adheres to both of the ideas of soft law mentioned above. ‘Better regulation’, as a form of soft regulation, attaches great importance to deliberation and discussion, but it also matches the general virtues of flexibility and respect for Member States’ sovereignty commonly attached to soft law. Focusing on soft law in the form of corporate social responsibility policies as well as methods of social and environmental reporting and certification schemes, the chapter by Boris Holzer shows how interactions between transnational corporations and transnational activists result in the elaboration of a transnational normative framework for corporate decision-making. Here soft law provides a framework for deliberation of socially acceptable norms, informing the public’s appraisal of business practices. The public in turn participates

4  Research handbook on soft law in the deliberative processes through the scrutiny of corporate behaviour during public debates about the rights and duties of corporate actors. Adhering instead to an idea of soft law as a regulatory tool, Hans-W. Micklitz’s chapter discusses soft law in the form of technical standards and shows how these function as a substitute for the failed attempt to find political support for the development of a European Civil Code. Two chapters complete this part by providing critical views on soft law. A reading, which rejects the idea of soft law providing a forum for deliberative social processes, is offered by Jan Klabbers. In his chapter, he argues that soft law has become a sort of elite governance, responsible for the populist backlashes which we witness everywhere in the world, and suggests replacing the notion of soft law altogether with that of ‘epistemic governance’ or ‘epistemic authority’. This would do justice to the idea of soft law being linked to knowledge and expertise. Finally, the chapter by Ingrid Gustafsson Nordin and Kristina Tamm Hallström uses organization theory to show how standards as a form of soft law - tend to dilute responsibility because they have a tendency to generate more organizational structures instead of clarifying or concentrating responsibility. The third theme covers the public and private actors as well as institutions adopting and engaging with soft law, and the process of soft law-making and its challenges. What is the role of national administrations and courts with respect to soft law? Does this role show peculiar features in a multi-level context such as that of the EU or in a federal system such as the USA? Have regulatory agencies become the main soft law regulators and why? How is soft law adopted and is this process accessible to societal stakeholders, social partners and civil society in general? What lessons can we learn from the use of the Open Method of Coordination (OMC), something once hailed as the prime soft law instrument in the EU? One important finding here is the soft law scene is populated with actors. Courts are probably the most contested actors in the scene. By systematically reviewing the literature on courts and soft law, Mariolina Eliantonio and Emilia Korkea-aho argue that there are compelling reasons for the courts to both engage with soft law as well as not to engage with it, leaving the courts in a bind with no clear avenue of escape. Of course, the institutional framework influences the courts’ actions with regard to soft law, and Jacint Jordana and Joan Solanes Mullor, focusing on EU economic governance, show how differently legal accountability is constructed with relation to the European Economic and Monetary Union and the European System of Financial Supervision. The next four chapters focus on actors that could be for good reasons considered as the most hands-on actors. Steven Vaughan offers a fascinating depiction of a series of ‘fictions’ and ‘fuzziness’ that are deployed to think about EU agencies and their normative powers (for instance, ‘EU agencies have no powers’ or that ‘agency guidance is not binding’). In his view, EU agencies’ soft law rule-making is ‘a complex governance situation characterized by various forms of fuzziness’, making them a ‘messy’ but worthwhile area of study. Moving across the Atlantic, Blake Emerson also looks at guidance but in the US context. He reminds us how the concept and use of guidance is hotly contested in the USA, which leads him to raise several serious questions about the impact of these documents on private parties’ moral reasoning as well as considering the ‘fraught role’ that guidance can play ‘in mediating intense political disputes’. His chapter is a welcome reminder to scholars on this side of the Atlantic not to look at soft law only as a neutral, technical instrument of governance. Part of seeing soft law as an instrument of politicized governance is to look at the processes of its creation. Noting that the procedures of soft law-making by the European Commission are famously obscure,

Introduction  5 Oana Ştefan asks how the Commission – the key soft law regulator in the EU – makes soft law. She shows that unlike the common perception, the Commission actively engages with public and private actors when issuing its soft law. While moving away from the specific context of the EU, Rene Urueña and Rafael Tamayo-Álvarez continue with this idea of collaboration. Rather than acting as norm entrepreneurs, civil society organizations play a significant role in the international law-making process by contributing to the ‘framing of the legal’ and shaping legal consciousness and ultimately, perceptions of what is legally possible. The final two chapters of the third part revisit the important question of soft law-making: Is soft law considered the second-best option (to hard law) or is it a new governance opening in its own right? The chapters show that both perceptions of soft law are important in explaining the dynamics behind soft law. By zooming in on the OMC in two different policy domains, social policy and energy and climate policy, Minna van Gerven and Sabina Stiller show that the OMC has undergone gradual changes following the EU developments. What is important is that the OMC is not simply a policy instrument. It provides a mode of governance that fosters creative appropriation and policy leverage for both national and EU level actors. The OMC (and by analogy, soft law) is whatever actors want it to be. In the final chapter of the third part, Anne Ausfelder, Adam Eick and Miriam Hartlapp, take up a neglected aspect of soft law, its use at the national level. They show how the way in which soft law is used at the national level is an integral part of the policy cycle of soft law, and by focusing only on the EU level, we not only miss important insights into soft law but also into the operation of the EU’s multi-level system more broadly. The fourth theme addresses the role of soft law in tackling some of the current global societal challenges. What role – if any – does soft law have to play in new regulatory fields such as artificial intelligence or in terms of emerging regulatory actors such as cities? Is soft law fit to tackle global and recurrent crises such as migration, climate change and financial instability? Can soft law serve as a suitable regulatory tool for politically sensitive problems, such as gender inequality, citizenship, or the backsliding of the rule of law? The main finding here is that soft law continues to have a major role in regulating the most urgent issues that our global societies face today. In fact, it may be the most effective and viable regulation mode in transnational and multistakeholder governance. The technologies of power in these regulatory processes raise fundamental questions about the relationship between normative ideals and regulatory practices. This is certainly the case with the question of how to reconcile liberal representative democratic ideals with times of crisis management and transnational governance. Another challenge is how to secure human rights while also ensuring a high degree of effectiveness of public policy. When it comes to the prevalence of soft law in the efforts to tackle global issues, the chapter of Kati Kulovesi and María Eugenia Recio shows, in the context of the pressing issue of climate change, that while an important regulatory tool in the field, soft law has not yet surpassed hard law when it comes to guiding countries’ behaviours vis-à-vis climate mitigation actions. From the global to the local level, a different picture is depicted by Astrid Voorwinden and Sofia Ranchordás’ chapter. They show that local governments have embraced soft law (in the form of technical standards, memoranda of understanding, charters of ethics, etc.) when using digital technologies, since these new technologies pose challenges unaddressed by traditional regulation. The use of soft law, in turn, foster the creation of transnational networks of cities, which the latter use to tackle global challenges. Along similar lines, the chapter by Timothy Jacob-Owens and Jo Shaw conclude that soft law is already a significant instrument

6  Research handbook on soft law in contemporary citizenship regimes, by contributing to the emergence of international norms shaping and constraining domestic citizenship practices in relation to the right to a nationality, the modes of citizenship acquisition, and multicultural citizenship. They also posit that soft law may come to play an even more substantial role in the development of an international citizenship regime. With respect to the capacity of soft law to generate policy change, the chapter by Birte Böök and Linda Senden shows that, in the field of gender equality, the effectiveness of soft law has been generally rather limited. Soft law has indeed not been sufficient to bring about the desired ‘classic’ legal implementation, nor the behavioural, organizational and social policy changes. Similarly, the chapter by Joelle Grogan and Clara van Dam, which considers the soft law mechanisms that have been adopted to enhance the rule of law in the EU, conclude that soft law tools do not necessarily guarantee an effective outcome in remedying the rule of law crisis. A number of chapters show how soft law has been the ‘go-to’ mechanism to tackle pressing societal issues. However, its use needs to be constantly monitored for compliance with democratic and fundamental rights credentials. In his chapter, Alexander Türk tackles the use of soft law in the financial crisis. He shows how the use of soft law, in particular by Union committees and agencies, as a means of crisis management raised significant concerns about the legal status and legitimacy of soft law instruments in this context as well as the danger of circumventing Treaty constraints. Finally, the chapter of Frederik Schade and Mikkel Flyverbom examines the multitude of soft law initiatives adopted at the EU level aimed at governing artificial intelligence (AI) technologies and their perceived societal risks. They show that significant ambiguities and uncertainties surrounding AI governance still remain, a situation which will need to be closely examined by researchers in the future because of important fundamental rights issues stemming from AI and its many uses.

3.

CONCLUSIONS: IS SOFT LAW (RESEARCH) FUTURE-PROOF?

Will we still be talking about soft law in 2030 or will the concept have become obsolete? If we asked some of our contributors, in the future we should be talking about ‘bamboo’ (Snyder) or ‘epistemic authority’ (Klabbers) instead of soft law. But if we accept that soft law is still part of our vocabulary in the years to come, what should we be talking about exactly? What are the avenues for further research? When putting together the handbook, we realized that soft law research has increasingly adopted a form of case study research. It was difficult to find scholars ready to write about soft law from a broader perspective, and many whom we invited to contribute to the present handbook agreed to do so only if they could submit a case study chapter. We acknowledge that this tendency towards case studies is because of the diversity of soft law instruments and their multiple applications across policy areas. While it may have been possible to write about soft law in general terms in the early 2000s, now 20 years later a soft law scholar feels compelled to characterize soft law as environmental soft law, soft law adopted by the three supervisory authorities in the area of financial market, and so on. And so, the research on soft law is an increasingly specialized field, with scholars of soft law at a growing rate speaking past each other and not engaging with each other’s research. One future avenue and challenge ahead for a broader soft law community is perhaps to zoom out

Introduction  7 a little and to focus on soft law as an instrument that is used across policy fields, recognizing that many advantages and disadvantages of using soft law are common to multiple policy fields. But while distances between researchers of ‘x’ soft law and those of ‘y’ soft law have increased in legal research and political science, there has been increasing rapprochement and cross-contamination between the researchers of ‘x’ soft law in legal scholarship and political science. For instance, researchers of environmental soft law and governance are increasingly cooperating and collaborating, a trend also seen in other sectors. In this sense, soft law has created interdisciplinary bridges and a sense of bonding, something that future research should further encourage. More research is also needed with linguistics scholars, psychologists, or economists, because there is more to know about soft law than just lawyers and political scientists can unveil. Are there any themes that we anticipate to be important in the future? One such theme is populism and the emergence of right-wing politics across Europe and beyond. Has the rise of populist movements changed societies’ use of soft law? Populism involves the rejection of expert-based politics, and soft law, which is often seen as a crystallization of state-of-the-art technical and expert knowledge on a certain matter (just think of agency soft law both in the EU and the USA), would then be rejected too. The extent to which this holds true should be studied (see also Klabbers, this volume). Relatedly, future research should take far more seriously than it has so far done the hidden values and distributional, moral and ethical effects of soft law on divided societies, a theme explored in Emerson’s chapter. While establishing the legal effects of soft law is certainly an important theme, we should also aim to establish the effects soft law has on minorities, for instance. Is using soft law a problem for minorities or from a gender perspective? Are enforceable rules necessary to ensure that governments do not take advantage of softness and fail to respect their obligations? Urueña and Tamayo-Álvarez’s chapter suggests that soft law may also shape the perceptions of what is legally possible, thus supporting civil society in the push for a better, fairer and more inclusive world. Horizons must be broadened also in terms of methods and theoretical frameworks. One obvious new horizon relates to computational text methods in the research on soft law. In terms of theoretical frameworks, it is surprising how little used Michel Foucault is in the research on soft law and governance. For instance, in this handbook, only one chapter (Schade and Flyverbom) discussed Foucault’s governmentality that could, however, open up several new directions for research. Further dialogue is needed between theories of governance and its empirical reality. Take for instance regulatory hybridity, which is a practice almost everywhere we look, but as a notion and an empirical practice it has very little support in theories of democracy (Mörth, this volume). Democratic theories are very much based upon static dichotomies. Should regulation adapt to impossible ideals, or should theories of democracies change and be modified? Is there a case for democratic hybridity that can match transnational regulatory hybridity? These research themes and questions require further interdisciplinary collaboration. The handbook shows how the diversity of the concept of soft law brings researchers together. Indeed, the richness of soft law in empirical, analytical and theoretical research constitutes a goldmine for interdisciplinary research.

8  Research handbook on soft law The practices of governance and regulation and the ensuing fundamental normative questions about power, legitimacy and democracy will certainly continue to be posed in the years to come.

PART I DEFINITIONS, HISTORY AND CONTEXT

1. Bamboo, or governance through soft law: hybridity, legitimacy, and sustainability Francis Snyder

1. INTRODUCTION1 Can ‘soft law’ really govern? Is the ‘rule of law’ dead? Does the world today have ‘rule of law, or law of rules’ (Djelic 2011)? Are standard setters ‘the new global rulers’ (Büthe and Mattli 2011)? Are we governed by an unruly pluralism of transnational legal orders (Halliday and Shaffer 2015; Zumbansen 2020) or sites of governance and transnational normative repertoires (Snyder 2001; Wu 2013; Snyder and Kim 2018), intimately intertwined with regional, national, or local laws and other norms? What ‘institutions, processes, tools, and techniques’ (Snyder 1993) do we need for a post-Covid-19 sustainable future? This chapter reflects on these themes, presents recent research, and rethinks certain received ideas. The chapter, with substantial bibliography as a guide for students, advances three main arguments. Soft law is a key part of legal pluralism in the European Union (EU) and elsewhere. It contrasts strikingly with two other normative forms: state law and executive management. In seeking legitimate norms to govern a sustainable future, priority should be given to normative hybridity. Due to space limits, the chapter deals primarily with Europe.

2.

SOFT LAW

In 1989 Wellens and Borchardt (1989) described the use of non-legally binding measures in the EU, emphasized their international law sources, and identified the importance of formally instituted legislative procedures as a key distinction between ‘soft law’ and legally binding measures (‘hard law’). In 1993 (Snyder 1993, p. 32), supplemented in 2007 (see Ştefan et al. 2019, p. 10), I defined soft law as ‘rules of conduct which, in principle, have no legally binding force but which, nevertheless, may have practical effects’, ‘and also legal effects’. This definition of ‘soft law’ was elaborated in an analysis of European Commission strategies for ensuring the effectiveness of European Economic Community (EEC) law, when the European normative universe appeared to be monopolized by the so-called ‘Community method’, especially in areas of long-standing exclusive EU competence such as the common agricultural policy and international trade (Snyder 1985, 1998), and within specific political and legal relations between EEC and Member States (Weiler 1991); the context is important to avoid being ‘lost in translation’ (Dezalay and Garth 2014). Since then, the definition, and its associated distinction between legal bindingness and legal effects, has been widely adopted. Here it provides a logical starting point for a comparison of the different normative forms and serves as a useful springboard for rethinking our ideas about soft law. Soft law is conceived as part of a dichotomy: it is explicitly counterposed to ‘hard law’, which refers to state law. ‘Hard law’ is legally binding: it creates legal obligations that can be 10

Bamboo, or governance through soft law  11 enforced, usually by courts (Komesar 2001, p. 3). In contrast, ‘soft law’, unless transformed into hard law (Snyder 2010, pp. 264–284), does not create legal obligations, at least for third parties, though it may create an obligation for its author(s) to respect their own creation, and its effectiveness usually relies on social practices and institutions other than courts. This contrast stems unavoidably from my 1993 definition, sometimes to the detriment of our understanding of the range of norms which are embedded in any society, which some scholars have addressed by means of the concept of legalization (Abbott et al. 2000). Nevertheless, it corresponds to ‘the current dominant paradigm for law’ (Tamanaha 1997, p. 99). The contraposition has two implications. First, only the state can be the author of hard law, whereas soft law can be produced by the state or non-state actors. These actors belong to different, though often overlapping, social fields (Lewin 1951, p. 140; Moore 1973; Bourdieu 1987; Trubek et al. 1994; Fligstein and McAdam 2012; Snyder 2016). Either state or non-state actors may produce standards, which sometimes may be legally binding, and sometimes may be converted into norms with legal effects or even legal force. Second, hard law often is assumed to refer to the legitimate use of power and authoritative decision-making in enacting legally binding state norms, following well-known, ideally transparent procedures defined in advance, whereas soft law sometimes has been deemed to be (merely) politics in the sense of competition for power and legitimacy. This contrast is extremely misleading, however, because both soft law and hard law (and executive management) involve politics, even though the normative outcomes may differ in the way and extent to which they impose obligations that can be sanctioned by the state. Soft law is an accepted term for part of the general sociological and anthropological category of social control (Ross [1901] 2017; Elster 1989; Chriss 2013) or ‘the mechanisms to co-ordinate normative expectations’ (Tamanaha 1997, p. 113). It is crucial to take seriously the warning that ‘Le fondamentalisme occidental est dépassé’ (Lepenies 1996), both historically and geographically. Norms other than law, whether written or unwritten, are universal and essential to life (Roberts 2013; Heimbuch 2020). Such norms are part of all societies; they are not limited to state-centred legal systems, to ‘successful’ states (Dixit 2007), to normative systems based on writing (Snyder 1981; Baum 1999; Merry 2000); or even to human beings (Goodall 1971, 1986; Hölldobler and Wilson 1991; Wulf 2016; Ackerman 2017; Balcombe 2017). Soft law is not limited to societies or legal systems derived mainly from Roman law or English common law; a widely accepted but now contested example is the so-called ‘law merchant’ (Kadens 2012, 2015). This argument is supported by recent research on Chinese legal history (Peerenboom 1993) and on China today (Snyder 2021a), with an institutional structure different from the ‘Western’ separation of powers and where soft law may also be created by courts (Finder 2020), as well as the Chinese Communist Party. Soft law, derived in the EU ostensibly from international law, has long been a feature of domestic law, for example in the standard normative administrative law toolkit, even though recognized in EU law only relatively recently. Soft law is an essential part of legal pluralism (Niglia 2013; Avbelj 2018; Davies and Avbelj 2018), meaning ‘the presence in a social field of more than one legal order’ (Ehrlich [1936] 2002); for present purposes, we replace ‘legal’ by ‘normative’ and abandon the civil law system concept of ‘a [legal] order’ (Merryman and Pérez-Perdomo 2019). Soft law is a major theme in transnational governance, whether as standards (Grindley 1995; Brunsson and Jacobsson 2000; Murphy and Yates 2009; Delimatsis 2015; Eliantonio and Cauffman 2020; Snyder 2021c, 2021d); indicators (Davis et al. 2012; Merry et al. 2015; Merry

12  Research handbook on soft law 2016; Nelken et al. 2020); or soft institutions such as the numerous sectoral dialogues between the EU and China (Snyder 2009). Standards are ubiquitous; they translate general objectives into specific rules and are an indispensable, albeit often controversial, aspect of regulation (Snyder 2021c). Soft law standards may be adopted by numerous actors: public, private, or mixed, and local, national, regional, international, or transnational (Schepel 2005). Such actors in global financial and product markets have been called ‘the new global rulers’ (Büthe and Mattli 2011). This shift has been ascribed to five interconnected, mutually reinforcing institutional forces: ‘scientifization, marketization, organizing (including standards and socialization), rationalization, and democracy’ (Djelic 2011), which together ‘foster a governance culture that heavily relies on soft rules’ (Djelic and Sahlin-Andersson 2006, p. 23). These forces are spread unevenly within countries and around the world: witness China’s Belt and Road Initiative (BRI) (Fang and Nolan 2019), which so far relies almost entirely on soft law memoranda of understanding (MoU) and standards, except for very technical measures, for example in the railway sector (Snyder 2021b, 2021c; Wang 2021). Soft law standards may be given legal effect by courts or other dispute procedures, or by ‘triangulation’ via certification or accreditation. For example, the CJEU in James Elliott Construction (C-613/14) concluded that harmonized standards formed part of EU law that the Court could interpret in the context of a preliminary ruling. The WTO Appellate Body (AB) in EC-Hormones (WTO 1998) converted non-binding Codex Alimentarius food safety standards into standards that have important legal effects such that they are binding de facto. Similarly, the AB in US-Gambling (WTO 2005) addressed standards of ‘right’ and ‘wrong’ in analyzing the ‘public morals’ exception in Article XX(a) GATT. Chinese national food safety standards have legal force; other food safety standards, such as non-legally-binding organic food standards are ‘expressed in China in legal form by certification and accreditation measures’ (Snyder 2020, p. 19). Soft law occupies a substantial place in EU legal scholarship; only a few major influences on this chapter are mentioned here. The first book-length treatment (Senden 2004) introduced the concept of soft law in EC law, dissected its categories and functions, and discussed several legal issues of EC soft law, such as legally binding force, competence to adopt soft law, and indirect legal effects for legislature, the judiciary, and the executive. It concluded that soft law was only a limited alternative to legislation; the principle of conferred powers (competence) should apply except in its function of legality; and that the European Parliament (EP) should play a greater role in its adoption. It recommended that soft law instruments should be limited, classified, and used more consistently; that their legal effects should be clarified; and that procedural changes were necessary to increase their legitimacy and transparency. Mörth’s (2004) edited collection studied soft law from law, political science, sociology, and organizational studies perspectives. Contrasting ‘government’ with ‘governance’ and ‘hard law’ with ‘soft law’, following ‘the turn to governance’ (Joerges and Everson 2005, pp. 161–162) initiated by Williamson (1985) in institutional economics, and Rosenau and Czempiel (1992) in international relations scholarship (Joerges and Weimer 2014, p. 304), it defined ‘governance’ as ‘rest[ing] on multiple authorities that are not necessarily public and sharing’. It discussed ‘soft law’, which ‘lacks the possibility for legal sanctions’, from the perspective of legal doctrine, governance through soft regulation, soft law and democracy, and the role of meta-organizations (EU, OECD). It accepted soft law as a useful analytical category but emphasized its diversity and the need to rethink the dichotomous ‘hard law’ and ‘soft law’ distinction on a case-by-case basis.

Bamboo, or governance through soft law  13 Soft law in the EU has been manifested most publicly in the Open Method of Coordination (OMC). The European Commission’s White Paper for the 2001 Lisbon Summit (European Commission 2001) is widely recognized as having formally oriented the EU toward new governance and soft law in the form of the OMC (Trubek and Trubek 2005; Craig 2011). Scott and Trubek’s pathbreaking article (2002) distinguished OMC and other soft law from the classic ‘Community method’ (see also Armstrong and Kilpatrick 2007; Walker and de Búrca 2007). Soft law was not used in isolation, however, and new governance represented a move from only state-centred government to more diverse, diffuse forms of governance, which might or might not be organized hierarchically, but did not always depend on the state, as demonstrated in examples of experimentalist governance (Sabel and Zeitlin 2008; Korkea-aho 2015), except in the general sense that capitalist economies are framed and often structured by the state. In fact, state and non-state forms of government/governance are often closely intertwined (Fligstein and McAdam 2012). Scholars of regulation (e.g., Black 1997; Braithwaite and Drahos 2000) and globalization (e.g., Snyder 1999, 2004; Levi-Faur 2005) oriented the soft law discussion toward regulatory governance. Colin Scott’s (2003) edited volume proposed new concepts of regulatory law, relations between state and market, the character of regulatory rules, the use of standards, and the role of different types of organizations, and the effects of globalization. His theory of post-regulatory governance (Scott 2004) showed how different normative forms, adopted by a variety of authors, including companies, trade unions, and consumers, and focusing on a host of regulatory targets, amounted to post-modern regulation. The combined use of soft law and hard law in the same policy area resulted in calls for a theory of hybridity (Trubek and Trubek 2005; Trubek et al. 2005). A landmark in contemporary studies of soft law in Europe was Ştefan’s (2013) study of soft law in competition and state aid cases in the EU courts. Building on previous scholarship and analyzing all relevant soft law cases in the EU courts, it developed an analytical framework consisting of three parts: (a) hybridity, showing the coexistence and engagement of law and new governance, sometimes created by courts; (b) judicialization, meaning ‘the construction of judicial power’ and ‘the capacity of a court to authoritatively determine the content of a community’s normative structure’ (citing Stone Sweet 1999); and (c) constitutionalization, in which the ECJ (now CJEU) located soft law on a constitutionalist background ‘through the intermediary of general principles of law’ (Ştefan 2013, p. 3). The conjunction of a broad theoretical framework and a thorough analysis of judicial decisions opened new perspectives for the study of EU soft law. The focus of soft law research shifted from the EU to Member States. The next step was a major multinational empirical research project (SoLaR: funded by the European Commission Jean Monnet Programme) on the use of EU soft law by national courts and administrations; the present author served on the Steering Committee. It focused on selected EU Member States (Finland, France, Germany, Italy, The Netherlands, Slovenia, the United Kingdom) and on selected policy areas (environmental law, social policy, competition and State aid law, financial regulation), with a comparative look at China. A first working paper provided a survey of the literature of EU soft law in the EU legal order (Ştefan et al., 2019). The project recommended that the European Commission improve accessibility of soft law for example by creating a publicly available and easily searchable website, provide translations of guidance documents and encourage implementation, and adopt guidelines for the adoption of soft law (Eliantonio et al. 2021). A follow-up international workshop in China (STL News

14  Research handbook on soft law 2019) compared the use of soft law by national courts and public authorities in the EU and China (Snyder et al. (forthcoming)).

3.

HARD LAW

Many writers (Marx 1848, and later: see Weber 1954; Picciotto 1979; Elster 1985; Meckstroth 2000; Tamanaha 2004; Ebner 2011; Capra and Mattei 2015; Jessop 2019; Wolff and Leopold 2021) associate the state and its legally binding rules (hard law) with the development of capitalism, which may be defined as ‘a system in which the production and distribution of goods is entrusted primarily to the market mechanism, based on private ownership of property, and on exchange between legally free individuals’ (Muller 2002, p. xvii). Capitalism may take many forms, including state capitalism (Liebman and Milhaupt 2016). The rise of the market economy required predictability, particularly concerning contracts and property rights, which arguably could be provided by only centralized government (but see Appelbaum et al. 2001; Dixit 2007; Clarke et al. 2008), which in theory had a monopoly of force; the result, as in Weber (1954, p. 146), is the ‘factual autonomy’ of ‘the propertied classes’. An important part of this process of creating ‘a field of experience’ (Cotterrell 1995, p. 3) was the ‘politics of definition’ (de Sousa Santos 2002, p. 91; see also Foucault 1980, pp. 81–108; Bourdieu 1991). The state acquired a monopoly of the term ‘law’, in the sense of measures that create enforceable obligations. This process of marginalization of non-state normative forms occurred in different contexts (Snyder 1982; Arthurs 1985; Djelic 2011, p. 14; Fligstein and McAdam 2012, pp. 67–74; Roberts 2015, pp. 275–279). Generally, ‘the conversion of the public place into the exclusive site of law and politics performed a crucial legitimation function in that it convincingly obscured the fact that the law and politics of the capitalist state could only operate as part of a broader political and legal configuration in which other contrasting forms of law and politics were included’ (de Sousa Santos 2003, pp. 82–83). Such are legal pluralism and hybrid normative configurations. This perspective relegates non-state norms to the category of ‘soft law’. As so defined, ‘soft law’ has no legally binding force. It does not create rights and obligations that can be enforced by (government) sanctions. There does not exist a ‘coercive apparatus’ in Max Weber’s (1954, p. 13) sense of ‘one or more persons whose special task it is to hold themselves ready to apply specially provided means of coercion (legal coercion) for the purpose of norm enforcement’. This perspective neglects other types of sanctions, such as ‘naming and shaming’ (Felstiner et al. 1980; Saiman 2018) or other, mainly nonpecuniary sanctions (but see Trubek and Trubek 2005). These assertions are contested by theories of legal pluralism, which posit that non-state actors may also produce norms that should be called ‘law’ (Griffiths 1986; Merry 1988; MacCormick 1995, 1999; Snyder 1999; Walker 2002; Maduro 2009). While recognizing differences between state and non-state ordering, a recent work (Berman 2012) argues that a focus on legal pluralism means that the distinction between law and non-law is irrelevant, because ‘the key questions involve the normative commitments of a community and the interactions among normative orders that give rise to such commitments, not their formal status’, and that we should ‘treat[…] as law that which people view as law’ (Berman 2012, pp. 54–55, 57). However, this constructivist perspective does not address the fact that most people usually

Bamboo, or governance through soft law  15 consider state law to be ‘law’, but nevertheless may at the same time follow both non-state and state rules unless they are incompatible (or sometimes even if they are), as the Covid-19 pandemic demonstrates. What about the rule of law? Is the rule of law, in one of its types, formal or substantive, and its several forms, for example the thin version or the thick version (see Neumann 1986; Peerenboom 2002), associated, inevitably or contingently, with hard law, the state, market economy, and capitalism? In the case of the thick version, certainly. For a thinner version, this depends (Seppӓnen 2016). A pragmatic statement of the thin version defines the rule of law in terms of ‘government … limited by law’, ‘formal legality’, and ‘the rule of law not man’ (Tamanaha 2004, pp. 119–123); hence it is not ‘apolitical technology’; this gives the impression however that a thick version of the role of law could be separated from politics other than in terms of its own ideology. The rule of law has been associated closely with the market economy, protection of property and contract (Ebner 2011, p. 22). Recent research (Mattei and Nader 2008, pp. 11–12) argues that the rule of law ‘was … born out of [lawyers’] role as guardians of a given, highly unequal, and certainly non-democratic distribution of property in society’ and ‘is a Western cultural artefact, closely connected with the diffusion of Western political domination’ (see also Pistor 2019). It served to legitimate colonialist and imperialist exploitation and plunder, as has been the case of many legal systems of different types characterized in terms of either the thick or thin versions of rule of law. Case C-44/79 Hauer (1979) reminds us that, according to positive law, fundamental rights are not absolute: the exercise of fundamental rights may be restricted in the public interest. But which institutions – non-majoritarian courts or others – should define ‘public interest’? Consider judicial use of the principle of proportionality in attempts to resolve conflicts between fundamental rights and EU market principles, as in cases C-438/05 Viking (2007) and C-341/05 Laval (2007). But why should the EU’s market principles be given priority over fundamental rights? Why should legal principles that emphasize the neoliberal market and the individual take precedence over those that focus on social responsibility, social solidarity, and duties that may be owed to society (Weil 1952; Carty and Nijman 2018; Weiler 2020)? Recent debates on EU executive management have revived basic questions about the rule of law and democracy in the EU. Recall the stricture (Senden 2004) that EU soft law should operate within the limits of EU competence. Concerned with the effects of the CJEU on national parliamentary democracy, Dieter Grimm argues that the so-called ‘constitutionalization’ of the EU Treaties through the concepts of direct effect, supremacy, and effet utile ‘laid the ground for the alienation of the non-political organs of the EU from the democratically legitimized and accountable organs’ ‘[such that] the Court is part of the legitimacy problem of the EU’ (Grimm 2016, p. 259; see also Scharpf 2015). Related is the CJEU’s ‘deregulatory bias’, resulting in Grimm’s view from the transformation of the EU’s four freedoms into fundamental rights (Grimm 2016, p. 245; see the debate in Grimm 1980; Hase et al. 1981; Neumann 1982). In his view ‘[s]overeignty’s most important function today lies in protecting the democratic self-determination of a politically united society with regard to the order that best suits it …. Sovereignty protects democracy’ (Grimm 2015, p. 128). Note however that sovereignty and state law are highly contested concepts (Krasner 1999; Koskenniemi 2004). Current research on transnational governance echoes many of these themes (Braithwaite and Drahos 2000; Hale and Held 2011). Philip Jessup’s (1956) pioneering lectures at Yale introduced transnational law. Studies of international capitalism recognized that transnational institutions may perform state functions and in some cases supplement or even replace the state

16  Research handbook on soft law (Murray 1971; Warren 1971). The resulting normative configurations have been described in different ways: transnational normative repertoires (Snyder 2001, 2016, 2020; Wu 2013), transnational legal orders (Halliday and Shaffer 2015), varieties of transnational law (Calliess and Zumbansen 2012; Zumbansen 2020), or forms of legal pluralism (Berman 2012), embracing both hard law and soft law, and deployed by public actors or private actors (Cutler and Dietz 2017).

4.

EXECUTIVE MANAGEMENT

Karl Llewellyn (1931, pp. 736–737) once wrote about contract, traditionally imagined as an agreement between legally equal parties, that: [T]he major importance of legal contract is to provide a framework for well-nigh every type of group organization and for well-nigh every type of passing or permanent relation between individuals and groups, up to and including states—a framework highly adjustable, a framework which almost never accurately indicates working relations but which affords a rough indication around which relations vary, an occasional guide in cases of doubt and a norm of ultimate appeal when relations cease in fact to work.

Such a framework emerged in relations between EU Member States as a third normative configuration: executive management, also known as governance by contract or ‘contractualization’. In executive management, the practical flexibility of a well-known private law instrument is combined with the flexibility, in both principle and practice, of an MoU. In international law an MoU is a treaty, the legal effects of which depend on the intentions of the parties. In other settings, an MoU may or may not be legally binding, but it indicates a meeting of mind between the parties. Among the myriad contemporary examples is the MoU between Italy and China concerning the Belt and Road Initiative (BRI), which provides, inter alia, that the MoU ‘does not constitute an international agreement which may lead to rights and obligations under international law. No provision of this Memorandum is to be understood and performed as a legal or financial obligation or commitment of the Parties’ (Italy-China 2019, para. VI). Contemporary EU executive management has developed in the framework of the 1992 Maastricht Treaty and in the 1997 Stability and Growth Pact, which began life as a compromise in the form of a non-legally binding Resolution at the Amsterdam Summit (Snyder 2011). The Maastricht Treaty created an unbalanced and perhaps unworkable hybrid Economic and Monetary Union (EMU). Monetary policy was assigned to the EU (Article 3 TFEU), but economic policy remained with the Member States, the EU having only a power of coordination (Article 5 TFEU). Member States were required ‘to regard their economic policies as matters of common concern and to coordinate them within the Council’ (Article 99 TEC, now Article 121 TFEU). The normative framework for this multilateral surveillance included Broad Economic Policy Guidelines (BEPG), the Excessive Deficit Procedure (EDP), both provided in the Treaty, and the Stability and Growth Pact (SGP), now provided in a Protocol to the EDP. In accordance with the Treaty (Article 104 TEC, now Article 126 TFEU), the Protocol (No 12) on the Excessive Deficit Procedure, agreed legally not by the Member States but by the High Contracting Parties to the TFEU, established reference values for excessive deficits: ‘3% of the ratio of the planned or actual government deficit to gross domestic product …’ and ‘60%

Bamboo, or governance through soft law  17 for the ratio of government debt to gross domestic product ….’ (Article 1). The Protocol was legally binding, but there was ‘considerable room for discretion’ (Snyder 2011, p. 698). Subsequently it was encapsulated in two Council regulations (1466/97, 1467/97). As early as 2003, however, France and Germany escaped application of the EDP, contrary to EU law as interpreted by the ECJ (C-27/04 Commission of the European Communities v Council of the European Union) and leading to reform of the EDP that enhanced Member States’ discretion and weakened multilateral surveillance. The European Council adopted Decision 2011/199 that ‘[t]he Member States whose currency is the euro may establish a stability mechanism to be activated if indispensable to safeguard the stability of the euro area as a whole. The granting of any required financial assistance under the mechanism will be made subject to strict conditionality’. This strong intergovernmentalism was accentuated by the adoption by the EU Member States (except the UK) as Contracting Parties, of the European Stability Mechanism (ESM) Treaty outside the EU legal framework, a strategy that was approved by the ECJ in cases C‑370/12 Pringle (2012) and C-62/14 Gauweiler (2015). Whether recognized as the only practical option (Beukers and De Witte 2013) or decried as ‘punitive and cynical politics masquerading as inept economics’ (Schepel 2017, p. 98), it represented a further step toward executive management and normative pluralism in Europe. Long-standing contradictions of EMU became front-page news with the 2009 financial crisis (Scharpf 2011; Chalmers et al. 2016). The EU Commission (‘on behalf of the ESM and with the approval of its Board of Governors) – in liaison with the ECB – with input from the IMF, and the Beneficiary Member State’ (European Commission 2015a, p. 2, para. (D)) concluded MoUs with Cyprus, Greece, Portugal, Ireland, and Spain to specify the conditions for receiving financial ‘bail-outs’ (European Commission 2011, 2012, 2013, 2015a, 2015b). For example, the Greek MoU (European Commission 2015a, pp. 4–6) stated: The Government therefore stands ready to take any measures that may become appropriate for this purpose [the reform agenda programme] as circumstances change. The Government commits to consult and agree with the European Commission, the European Central Bank and the International Monetary Fund on all actions relevant for the achievement of the objectives of the Memorandum of Understanding before these are finalized and legally adopt. … Greece will design and implement a wide range of reforms in labour markets and product markets (including energy) that not only ensure full compliance with EU requirements, but which also aim at achieving European best practices. There will be an ambitious privatization programme, and policies which support investment.

The other MoUs followed the same pattern of imposing dramatic, domestic, undemocratic reforms, which lacked any sound legal basis in the EU Treaties or ex ante in national law. Nevertheless, for Germany, the legal implications were approved in effect by the German Federal Constitutional Court (Joerges and Weimer 2014). MoUs ‘are becoming a normal feature of the EMU governance system’ (Niglia 2020, p. 152). These MoUs operate in a completely different context from traditional international law. Leone Niglia (2020) recently traced their historical antecedents, ideological underpinnings, and legal implications to the context of colonial concessions (Cousin [1899] 2014; Ruedy 1967; Deroche 2004; Nouschi 2012), based on the presumptions of private law, individualism, and a capitalist market economy. In the EU, they amount, in his view, to ‘the regression from integration through law as an anti-hegemonic project of equal membership to a condition in which state orders, under a transformed European Union law, gravitate around unequal rela-

18  Research handbook on soft law tions of subordination’. Inherent legal problems are reflected in the terminology: ‘executive management’, ‘contractualization’, or ‘governance by contract’. The expression ‘executive management’ is unknown in the Treaties. The contractual terminology is deeply misleading, because of the ‘interpretation of the memoranda as an interstate governance technique and as a technique unconstrained by constitutional limitations’ (Niglia 2020, p. 168) that ‘bypasse[s] and neglect[s] the constitutional structure of the then-liberal state’ and is outside the control of the constitutional courts (Niglia 2020, p. 161): ‘what is being sacrificed is the “epicenter” of the idea of Europe as based on law’ (Niglia 2016, p. 133). The use of the term ‘contractual’ for these instruments conveys ‘the downgrading, and ultimately the exclusion, of any form of constitutional control over the memoranda contracts’ (Menéndez 2017, p. 171; Niglia 2020, p. 171): a form of EU ‘democratic default’ (Scharpf 2011; Majone 2014). ‘The consequences of this shift to contract remain to be seen. It will at the very least increase the complexity of the overall regulatory landscape. It will moreover not be easy to ensure the desired legitimacy and accountability’ (Craig 2014, p. 30). Though informed by the ideology of the rule of law, the political shift to contract is not neutral and represents ‘a process of institutional and productive power’ (May 2017, p. 51), even though ‘legalized’ later by EU regulations (472/2013, 473/2013). Executive management amounted in the EU to a new form of governance (Menéndez 2017; Scharpf 2017). Many scholars, often evoking memories of constitutional debates during the short-lived Weimar Republic (Caldwell 1997; Davies 2012), and the legacy of National Socialism and Fascism (Joerges and Ghaleigh 2003), have analyzed in detail the ways in which ‘contractualization’ diverged from the rule of law, democracy, national parliamentary control of legislation, and judicial review (Joerges and Glinski 2014; Chalmers et al. 2016). It stimulated the use of soft law, including MoUs, which ‘camouflaged’ the move from a publicly recognized economic constitution toward ‘bargaining and compromise’ (Adamski 2012; see also Kandil 2005; Chalmers et al. 2016, pp. 9–10). The MoUs may however be considered an implementation of EU law and therefore subject to the EU Convention on Fundamental Rights (Rödl and Callsen 2014, p. 177).

5.

TOWARDS SUSTAINABLE GOVERNANCE

Soft law, hard law, and executive management are typically intertwined, not discrete. In practice, they are often used in concert, as ‘combinations of modes and instruments of governance’ (Armstrong 2013), by ‘conscious design or … because the same objective is being pursued through two routes’ (Trubek et al. 2005, p. 31), with soft law being accompanied or framed typically by hard law or sometimes by executive management. Encounters between different types of norms may take the form of support, effect, or content (Champeil-Desplats 2019). This has been accentuated as soft law has proliferated with globalization, the rise of neo-liberal ideology, an emphasis on the de-regulated or re-regulated market, and new social fields and political actors (Braithwaite and Drahos 2000; Snyder 2004; Brakman et al. 2006). Seen in its social context (Snyder 1990), focusing on more than only the norm in isolation, soft law thus often exemplifies hybridity, expressly or implicitly (Luo and Song 2013). Some examples: the validity of EU recommendations is assessed by the CJEU within the framework of EU law (C-16/16 P Kingdom of Belgium v European Commission); soft law in EU competition law is evaluated according to the general principles of EU law; hard law and the OMC

Bamboo, or governance through soft law  19 are combined in EU employment policy (Kilpatrick 2006); MoU, a classic soft law instrument, is backed by state power; ‘labour lore’ or the ‘law of the shop’ in industrial relations (Arthurs 2019, pp. 33, 71, 74) and cultural norms in football/soccer (Ancelotti et al. 2016, pp. 157–165; Duval and Heerdt 2020) often trump state law and international rules; soft law promoted by market actors depends on the states that create and sustain the market, even in China (Nee and Opper 2014); the evolution of EU fiscal discipline includes executive management, soft law, and hard law (Armstrong 2014); international, national, and local soft law is deeply imbricated with hard law in the platforms characteristic of informational capitalism (Cohen 2019), regulation of toxic pesticides in transnational legal orders (Snyder and Ni 2017; Snyder 2020), and multilateral environment agreements (MEAs) (Heyvaert 2009) for many reasons discussed by Shelton (2000); EU soft law on Covid-19 is dependent for its legitimacy and efficacy on Member States’ law (Eliantonio, Korkea-aho and Vaughan 2021). In recently proposed normative models for pluralism (Krisch 2010; Berman 2012; Micklitz 2013; Niglia 2013; Reich 2013), a crucial issue is how to manage these forms of inter-normativity, in particular how to protect individuals against other individuals, groups, and government. In contrast to the management of inter-legality limited to hard law (Klabbers and Palombella 2019), this cannot and should not be done only by non-majoritarian institutions such as courts. What other institutions (Chalmers and Chaves 2014) are required to ensure participation, legitimacy, and accountability in decision-making? Is the only legitimate answer to this question in our transnational world ‘the emergence of further rules, of further regulators, certifiers, evaluators, and controllers’ (Djelic 2011, p. 9)? Whether this is the ‘only legitimate answer’ remains to be explored: we add that experts must also listen to the people (Zakaria 2020, pp. 89–96). It is essential to keep in mind that ‘legal pluralist regimes’ and ‘non-state legal systems are just as likely as state systems to be dominated and distorted by unequal aggregations of power’ (Arthurs 2019, pp. 129, 136, also pp. 39, 126). Global constitutionalism is a chimera (Krisch 2010; Weiler 2012). Perhaps ‘we have to be content with local coherence’ (Tuori 2014, p. 56; see also Cotterrell 1995, p. 309). Both locally and farther afield, this requires an important turn towards proceduralization, national administrative law and global administrative law (Kingsbury et al. 2005; Cassese 2016; Rose-Ackerman et al. 2017) and a healthy dose of normative imagination, especially about network governance (Ladeur 1997, 2017; Viellechner 2009). Further research is required on algorithmic governance: ‘[t]he introduction of artificial intelligence, predictive analytics, automated decision-making, actuarial risk assessment instruments, and machine learning’ (Sutherland 2021, p. 6) for the purpose of collection, processing, utilization, and feedback and learning concerning big data (Danaher et al. 2017, p. 3; Hassan and De Filippi 2017; Katzenbach and Ulbricht 2019; Latzer and Festic 2019). Compared to soft law, hard law, and executive management, however, algorithmic governance is characterized by even greater lack of transparency, lack of rootedness in localities, and more severe power imbalances (Zarsky 2016; Kellogg et al. 2020; Sutherland 2021). We may expect hybridity to grow in our search in the post-Covid world for the legitimacy of rules and the sustainability of our species. Recently, many writers (Speth 2008; Capra and Mattei 2015; Shiva 2015; Campbell and Campbell 2016; Piketty 2017; Scanlan 2017; Fleurbaey et al. 2018; Schwab and Malleret 2020; Gates 2021; Speth and Courrier 2021) have identified the major challenges: controlling climate change and preserving the environment; respect for nature, the commons, and all species; rethinking the economic system of capitalism; social solidarity, checking inequality, and limiting private accumulation; eliminating

20  Research handbook on soft law hunger and promoting vegetarian and healthy food. Achieving these objectives will require not simply power, experts, management, and technology but also resilience and flexibility. In thinking about sustainable governance, it is time to abandon the label ‘soft law’. Logically, ‘soft law’ is an oxymoron (on oxymorons, see Neuwirth 2018). The word ‘law’ in the expression ‘soft law’ is redundant and misleading, unless we are concerned with maintaining continuity with an established line of legal scholarship, or with increasing semantically the power and legitimacy of norms that formally are not called ‘law’ by the state. We need a different expression for such a normal, well-known, and widely accepted normative configuration. I propose that ‘soft law’ should be renamed ‘bamboo’. Bamboo is part of human and natural life that is strong, fast-growing, highly visible, easily recognizable, resilient, and flexible. Soft law bamboo, hard law oak, the executive management thicket, and the roots of big data are intertwined in so many distinct or overlapping social fields that hybridity is the rule and non-hybridity the exception.

NOTE 1. Thanks to Mariolina Eliantonio, Susan Finder, Danny Friedmann, Emilia Korkea-aho, Lu Qing, Rostam Neuwirth, Leone Niglia, Philip McConnaughay, Nathalie Rubio, Anne-Lise Strahtmann, Oana Ştefan, Yuan Ye, Zhang Hui, and Zhu Mengting for their contributions to this chapter. I am grateful to Peking University School of Transnational Law and Peking University Shenzhen Graduate School for support. Sole responsibility for the chapter remains mine.

REFERENCES Abbott, K. W., Keohane, R. O., Moravcsik, A., Slaughter, A.-M. and Snidal, D., (2000). The concept of legalization. International Organization [online]. 54(3), 401–419. [Viewed 12 October 2021]. Available from: doi: 10.1162/002081800551271. Ackerman, J., (2017). The genius of birds. London: Little, Brown Book Group. Adamski, D., (2012). National power games and structural failures in the European macroeconomic governance. Common Market Law Review. 49(4), 1319–1364. Ancelotti, C., Brady, C. and Forde, M., (2016). Quiet leadership: winning hearts, minds and matches. London: Portfolio Penguin. Appelbaum, R. P., Felstiner, W. L. F. and Gessner, V., eds., (2001). Rules and networks: the legal culture of international business transactions. Oxford: Hart Publishing. Armstrong, K. A., (2013). New governance and the European Union: an empirical and conceptual critique. In: G. de Búrca, C. Kilpatrick and J. Scott, eds. Critical legal perspectives on global governance: liber amicorum David M Trubek. Oxford: Hart Publishing. pp. 249–271. Armstrong, K. A., (2014). Differentiated economic governance and the reshaping of dominium-law. In: M. Adams, F. Fabbrini and P. Larouche, eds. The constitutionalization of European budgetary constraints. Oxford: Hart Publishing. pp. 65–83. Armstrong, K. and Kilpatrick, C., (2007). Law, governance, or new governance: the changing open method of coordination. Columbia Journal of European Law. 13, 649–677. Arthurs, H. W., (1985). ‘Without the law’: administrative justice and legal pluralism in nineteenth-century England. Toronto, ON: University of Toronto Press. Arthurs, H. W., (2019). Connecting the dots: the life of an academic lawyer. London and Chicago: McGill-Queen’s University Press, and Toronto: The Osgoode Society for Canadian Legal History. Avbelj, M., (2018). The European Union under transnational law: a pluralist appraisal. Oxford: Hart Publishing.

Bamboo, or governance through soft law  21 Balcombe, J., (2017). What a fish knows: the inner lives of our underwater cousins. London: Oneworld Publications. Baum, R. M., (1999). Shrines of the slave trade: Diola religion and society in precolonial Senegambia. New York: Oxford University Press. Berman, P. S., (2012). Global legal pluralism: a jurisprudence of law beyond borders. Cambridge: Cambridge University Press. Beukers, T. and De Witte, B., (2013). The Court of Justice approves the creation of the European Stability Mechanism outside the EU legal order: Pringle. Common Market Law Review. 50(3), 805–848. Black, J., (1997). Rules and regulators. Oxford: Clarendon Press. Bourdieu, P., (1987). The force of law: toward a sociology of the juridical field. Translated by R. Terdiman. The Hastings Law Journal, 38(5), 814–854. Bourdieu, P., (1991). Language and symbolic power. J. B. Thompson, ed. Translated by G. Raymond and M. Adamson. Cambridge: Polity Press. Braithwaite, J. and Drahos, P., (2000). Global business regulation. Cambridge: Cambridge University Press. Brakman, S., Garretsen, H., Marrewijk, C. van, and Witteloostuijn, A. van, (2006). Nations and firms in the global economy: an introduction to international economics and business. Cambridge: Cambridge University Press. Brunsson, N. and Jacobsson, B., (2000). A world of standards. Oxford: Oxford University Press. Büthe, T. and Mattli, W., (2011). The new global rulers: the privatization of regulation in the world economy. Princeton, N.J.: Princeton University Press. Caldwell, P. C., (1997). Popular sovereignty and the crisis of German constitutional law: the theory and practice of Weimar constitutionalism. Durham and London: Duke University Press. Calliess, G.-P. and Zumbansen, P., (2012). Rough consensus and running code: a theory of transnational private law. Oxford: Hart Publishing. Campbell, T. C. and Campbell, T. M. II. (2016). The China study. Revised and expanded ed. Dallas, TX: BenBella Books. Capra, F. and Mattei, U., (2015). The ecology of law: toward a legal system in tune with nature and community. Oakland, CA: Berrett-Koehler Publishers. Carty, A. and Nijman, J., eds., (2018). Morality and responsibility of rulers: European and Chinese origins of a rule of law as justice for world order. Oxford: Oxford University Press. Case C-44/79 Liselotte Hauer v Land Rheinland-Pfalz ECLI:EU:C:1979:290. Case C-27/04 Commission of the European Communities v Council of the European Union ECLI:EU:C:2004:436. Case C-341/05 Laval un Partneri Ltd v Svenska Byggnadsarbetareförbundet, Svenska Byggnadsarbetareförbundets avdelning 1, Byggettan and Svenska Elektrikerförbundet ECLI:EU:C:2007:809. Case C-438/05 International Transport Workers’ Federation and Finnish Seamen’s Union v Viking Line ABP and OÜ Viking Line Eesti ECLI:EU:C:2007:772. Case C-370/12 Thomas Pringle v Government of Ireland and Others ECLI:EU:C:2012:756. Case C-62/14 Peter Gauweiler and Others v Deutscher Bundestag ECLI:EU:C:2015:400. Case C-613/14 James Elliott Construction Limited v Irish Asphalt Limited ECLI:EU:C:2016:821. Case C-16/16 P Kingdom of Belgium v European Commission ECLI:EU:C:2018:79. Cassese, S., ed., (2016). Research handbook on global administrative law. Cheltenham, UK: Edward Elgar Publishing. Chalmers, D. and Chaves, M., (2014). EU law-making and the state of European democratic agency. In: O. Cramme and S. B. Hobolt, eds. Democratic politics in a European Union under stress. Oxford: Oxford University Press. pp. 155–179. Chalmers, D., Jachtenfuchs, M. and Joerges, C., (2016). The retransformation of Europe. In: D. Chalmers, M. Jachtenfuchs and C. Joerges, eds. The end of the Eurocrats’ dream: adjusting to European diversity. Cambridge, UK: Cambridge University Press. pp. 1–28. Champeil-Desplats, V., (2019). Law, the plurality of modes of normativity and their interactions. La Revue des Droits de l’Homme [online]. 16. Translated by L. Gilbert. [Viewed 15 April 2021]. Available from: doi: 10.4000/revdh.7108. Chriss, J. J., (2013). Social control: an introduction. 2nd ed. London: Polity Press.

22  Research handbook on soft law Clarke, D. C., Murrell, P. and Whiting, S., (2008). The role of law in China’s economic development. In: L. Brandt and T. G. Rawski, eds. China’s great economic transformation. Cambridge: Cambridge University Press. pp. 375–428. Cohen, J. E., (2019). Between truth and power. the legal construction of informational capitalism. New York, N.Y.: Oxford University Press. Consolidated version of the Treaty on European Union - Protocols - Protocol (No 12) on the excessive deficit procedure (2008) Official Journal C 115, 8 May, pp. 279–280. Cotterrell, R., (1995). Law’s community: legal theory in sociological perspective. Oxford: Clarendon Press. Council regulation (EC) No 1466/97 on the strengthening of the surveillance of budgetary positions and the surveillance of economic policies (1997) Official Journal L209, p. 1. Council regulation (EC) No 1467/97 on speeding up and clarifying the implementation of the excessive deficit procedure (1997) Official Journal L209, p. 6. Council decision 2011/199/EU amending Article 136 of the Treaty on the Functioning of the European Union with regard to a stability mechanism for Member States whose currency is the euro (2011) Official Journal L 91, p. 1. Cousin, A., (1899). Concession coloniale, droits et obligations en résultant. etude sur la concession de la rive gauche de la Casamance. Reprint, Paris: Hachette, 2014. Craig, P., (2011). Institutions, power, and institutional balance. In: P. Craig and G. de Búrca, eds. The evolution of EU law. 2nd ed. Oxford: Oxford University Press. pp. 41–84. Craig, P., (2014). Economic governance and the Euro crisis: constitutional architecture and constitutional implications. In: M. Adams, F. Fabbrini and P. Larouche, eds. The constitutionalization of European budgetary constraints. Oxford: Hart Publishing. pp. 19–40. Cutler, A. C. and Dietz, T., eds., (2017). The politics of private transnational governance by contract. London and New York: Routledge. Danaher, J., Hogan, M. J., Noone, C., Kennedy, R., Behan, A., De Paor, A., Felzmann, H., Haklay, M., Khoo, S.-M., Morison, J., Murphy, M. H., O’Brolchain, N., Schafer, B. and Shankar, K., (2017). Algorithmic governance: developing a research agenda through the power of collective intelligence. Big Data & Society. 4(2), 1–21. [Viewed 15 April 2021]. Available from: doi: 10.1177/2053951717726554. Davies, B., (2012). Resisting the European Court of Justice: West Germany’s confrontation with European law, 1949–1979. Cambridge, UK: Cambridge University Press. Davies, G. and Avbelj, M., (2018). Research handbook on legal pluralism and EU law. Cheltenham, UK: Edward Elgar Publishing. Davis, K. E., Fisher, A., Kingsbury, B. and Merry, S. E., eds., (2012). Governance by indicators: global power through quantification and rankings. Oxford: Oxford University Press. de Sousa Santos, B., (2002). Towards a new legal common sense: law, globalization and emancipation. 2nd ed. London: Butterworths. de Sousa Santos, B., (2003). On modes of production of law and social power. In: C. Scott, ed. Regulation. London: Dartmouth-Ashgate. pp. 74–112. Delimatsis, P., ed., (2015). The law, economics and politics of international standardisation. Cambridge: Cambridge University Press. Deroche, A., (2004). France coloniale et droit de propriété: les concessions en Indochine. Paris: L’Harmattan. Dezalay, Y. and Garth, B. G., (2014). Lost in translation: on the failed encounter between Bourdieu and law and society scholarship and their respective blindnesses. In: G. de Búrca, C. Kilpatrick and J. Scott, eds. Critical legal perspectives on global governance: liber amircorum David M Trubek. Oxford: Hart Publishing. pp. 385–404. Dixit, A. K., (2007). Lawlessness and economics: alternative modes of governance. 2nd ed. Princeton, N.J.: Princeton University Press. Djelic, M.-L., (2011). From the rule of law to the law of rules: the dynamics of transnational governance and their local impact. International Studies of Management & Organisation. 41(1), 35–61. Djelic, M.-L. and Sahlin-Andersson, K. (2006). Introduction: A world of governance: the rise of transnational regulation. In: M.-L. Djelic and K. Sahlin-Andersson, eds. Transnational governance: institutional dynamics of regulation. Cambridge, UK: Cambridge University Press. pp. 1–28.

Bamboo, or governance through soft law  23 Duval A. and Heerdt D., (2020). FIFA and human rights – a research agenda. Tilburg Law Review [online]. 25(1), pp. 1–11. [Viewed 15 November 2021]. Available from: doi: 10.5334/tilr.189. Ebner, A., (2011). Transnational markets and the Polanyi problem. In C. Joerges and J. Falke, eds. Karl Polanyi, globalisation and the potential of law in transnational markets. Oxford: Hart Publishing. pp. 19–40. Ehrlich, E., (1936). Fundamental principles of the sociology of law. Translated by W. L. Moll. Reprint, London: Transaction Publishers, 2002. Eliantonio, M. and Cauffman, C., eds., (2020). The legitimacy of standardisation as a regulatory technique: a cross-disciplinary and multi-level analysis. Cheltenham, UK: Edward Elgar Publishing. Eliantonio, M., Korkea-aho, E. and Ştefan, O., eds., (2021). EU soft law in the Member States: theoretical findings and empirical evidence. Oxford, UK: Hart Publishing. Eliantonio, M., Korkea-aho, E. and Vaughan, S., (2021). EJRR special issue editorial: COVID-19 and soft law: is soft law pandemic-proof?. European Journal of Risk Regulation [online]. 12(1), 1–6. [Viewed 8 November 2021]. Available from: doi: 10.1017/err.2021.1. Elster, J., (1985). Making sense of Marx. Cambridge, UK: Cambridge University Press. Elster, J. (1989). The cement of society: a survey of social order. Cambridge, UK: Cambridge University Press. European Commission., (2001). European governance – a white paper (COM/2001/428 final) Official Journal C287, p. 1. European Commission, (2011). Portugal: memorandum of understanding on specific economic policy conditionality [online]. [Viewed 2 November 2020]. Available from: https://​ec​.europa​.eu/​economy​ _finance/​eu​_borrower/​mou/​2011​-05​-18​-mou​-portugal​_en​.pdf. European Commission, (2012). Spain: memorandum of understanding on financial-sector policy conditionality [online]. [Viewed 2 November 2020]. Available from: https://​ec​.europa​.eu/​economy​ _finance/​eu​_borrower/​mou/​2012​-07​-20​-spain​-mou​_en​.pdf. European Commission, (2013). The economic adjustment programme for Cyprus. European Economy, Occasional Papers 149 [online]. Brussels: European Commission. [Viewed 21 April 2021]. Available from: https://​ec​.europa​.eu/​economy​_finance/​publications/​occasional​_paper/​2013/​pdf/​ocp149​_en​.pdf. European Commission, (2015a). Memorandum of understanding between the European Commission acting on behalf of the European Stability Mechanism and the Hellenic Republic and the Bank of Greece [online]. [Viewed 3 November 2020]. Available from: https://​ec​.europa​.eu/​info/​sites/​info/​ files/​01​_mou​_20150811​_en1​.pdf. European Commission, (2015b). Financial assistance to Ireland [online]. [Viewed 2 November 2020]. Available from: https://​ec​.europa​.eu/​info/​business​-economy​-euro/​economic​-and​-fiscal​ -policy​-coordination/​financial​-assistance​-eu/​which​-eu​-countries​-have​-received​-assistance/​financial​ -assistance​-ireland​_en. European Parliament and Council regulation (EU) no 472/2013 on the strengthening of economic and budgetary surveillance of Member States in the euro area experiencing or threatened with serious difficulties with respect to their financial stability (2013) Official Journal L 140, p. 1. European Parliament and Council regulation (EU) no 473/2013 on common provisions for monitoring and assessing draft budgetary plans and ensuring the correction of excessive deficit of the Member States in the euro area (2013) Official Journal L 140, p. 11. Fang, C. and Nolan, P., eds., (2019). Routledge handbook of the Belt and Road. New York, N.Y.: Routledge. Felstiner, W. L. F., Abel, R. L. and Sarat, A., (1980). The emergence and transformation of disputes: naming, blaming, claiming …. Law & Society Review. 15(3/4), 631–654. Finder, S., (2020). The “soft law” of the Supreme People’s Court [online]. Supreme Court Monitor. [Viewed 21 April 2021]. Available from: https://​suprem​epeoplesco​urtmonitor​.com/​2020/​11/​16/​the​ -soft​-law​-of​-the​-supreme​-peoples​-court/​. Fleurbaey, M., Bouin, O., Salles-Djelic, M.-L., Kanbur, R., Nowotny, H. and Reis, E., (2018). A manifesto for social progress: ideas for a better society. Cambridge, UK: Cambridge University Press. Fligstein, N. and McAdam, D., (2012). A theory of fields. Oxford: Oxford University Press. Foucault, M., (1980). Power/knowledge: selected interviews and other writings, 1972–1977. C. Gordon, ed. New York: Pantheon Books.

24  Research handbook on soft law Gates, B., (2021). How to avoid a climate disaster: the solutions we have and the breakthroughs we need. London: Allen Lane. Goodall, J., (1971). In the shadow of man. London: Collins. Goodall, J., (1986). The chimpanzees of Gombe: patterns of behavior. Cambridge, M.A.: Belknap Press of Harvard University Press. Griffiths, J., (1986). What is legal pluralism? The Journal of Legal Pluralism and Unofficial Law. 18(24), 1–55. Grimm, D., (1980). Reformalisierung des rechtsstaats als demokratiepostulat?. Juristische Schulung. 704–709. Grimm, D., (2015). Sovereignty: the origin and future of a political and legal concept. Translated by B. Cooper. New York: Columbia University Press. Grimm, D., (2016). Europe’s legitimacy problem and the courts. In: D. Chalmers, M. Jachtenfuchs and C. Joerges, eds. The end of the Eurocrats’ dream: adjusting to European diversity. Cambridge: Cambridge University Press. pp. 241–265. Grindley, P., (1995). Standards, strategy, and policy: cases and stories. Oxford: Oxford University Press. Hale, T. and Held, D., eds., (2011). Handbook of transnational governance: institutions and innovations. Cambridge, UK: Polity Press. Halliday, T. C. and Shaffer, G., eds., (2015). Transnational legal orders. Cambridge: Cambridge University Press. Hase, F., Ladeur, K.-H. and Ridder, H., (1981). Nochmals: reformalisierung des rechtsstaats als demokratiepostulat?. Juristische Schulung. 794–798. Hassan, S. and De Filippi, P., (2017). The expansion of algorithmic governance: from code is law to law is code. Field Actions Science Reports [online]. 17, 88–90. Available from: http://​journals​.openedition​ .org/​factsreports/​4518. Heimbuch, J., (2020). 10 surprising facts about Neanderthals [online]. Treehugger. [Viewed 22 February 2020]. Available from: https://​www​.treehugger​.com/​surprising​-facts​-about​-neanderthals​-4864080. Heyvaert, V., (2009). Hybrid norms in international law. LSE Law, Society and Economy Working Papers, 6/2009 [online]. [Viewed 2 March 2021]. Available from: http://​eprints​.lse​.ac​.uk/​23586/​1/​ WPS2009​-06​_Heyvaert​.pdf. Hölldobler, B. and Wilson, E. O., (1991). The ants. Cambridge, M.A.: Belknap Press of Harvard University Press. Italy-China., (2019). Memorandum of understanding between the Government of the Italian Republic and the Government of the People’s Republic of China on cooperation within the framework of the Silk Road Economic Belt and the 21st century maritime Silk Road Initiative [online]. [Viewed 8 May 2020]. Available from: http://​www​.governo​.it/​sites/​governo​.it/​files/​Memorandum​_Italia​-Cina​_EN​ .pdf. Jessop, B., (2019). Critical theory of the state. In: E. Christodoulidis, R. Dukes and M. Goldoni, eds. Research handbook on critical legal theory. Cheltenham, UK: Edward Elgar Publishing. pp. 114–134. Jessup, P. C., (1956). Transnational law. New Haven: Yale University Press. Joerges, C. and Everson, M., (2005). The European turn to governance and unanswered questions of legitimacy: two examples and counter-intuitive suggestions. In: C. Joerges, B. Stråth and P. Wagner, eds. The economy as a polity: the political constitution of contemporary capitalism. London: UCL Press. pp. 159–179. Joerges, C. and Ghaleigh, N. S., eds., (2003). Darker legacies of law in Europe: the shadow of national socialism and fascism over Europe and its legal traditions. Oxford: Hart Publishing. Joerges, C. and Glinski, C., eds., (2014). The European crisis and the transformation of transnational governance: authoritarian managerialism versus democratic governance. Oxford: Hart Publishing. Joerges, C. and Weimer, M., (2014). A crisis of executive management in the EU: no alternative?. In: G. de Búrca, C. Kilpatrick and J. Scott, eds. Critical legal perspectives on global governance: liber amicorum David M Trubek. Oxford: Hart Publishing. pp. 295–322. Kadens, E., (2012). The myth of the customary law merchant. Texas Law Review [online]. 90, 1153–1206. [Viewed 9 November 2021]. Available from: https://​ssrn​.com/​abstract​=​2043550. Kadens, E., (2015). The medieval law merchant: the tyranny of a construct. Journal of Legal Analysis [online]. 7(2), 251–289. [Viewed 9 November 2021]. Available from: doi: 10.1093/jla/lav004.

Bamboo, or governance through soft law  25 Kandil, F., (2005). Economic efficiency and social justice: a prudential approach for public actions. In: C. Joerges, B. Stråth and P. Wagner, eds. The economy as a polity: the political constitution of contemporary capitalism. London: UCL Press. pp. 203–221. Katzenbach, C. and Ulbricht, L., (2019). Algorithmic governance, Internet Policy Review [online]. 8(4), 1–18. [Viewed 15 April 2021]. Available from: doi: 10.14763/2019.4.1424. Kellogg, K. C., Valentine, M. A. and Christin, A., (2020). Algorithms at work: the new contested terrain of control. Academy of Management Annals [online]. 14(1), 366–410. [Viewed 15 April 2021]. Available from: doi: 10.5465/annals.2018.0174. Kilpatrick, C., (2006). New EU employment governance and constitutionalism. In: G. de Búrca and J. Scott, eds. New governance and constitutionalism in Europe and the US. Oxford: Hart Publishing. pp. 121–152. Kingsbury, B., Krisch, N. and Stewart, R. B., (2005). The emergence of global administrative law. Law and Contemporary Problems. 68, 15–61. Klabbers, J. and Palombella, G., eds., (2019). The challenge of inter-legality. Cambridge: Cambridge University Press. Komesar, N. K., (2001). Law’s limits: the rule of law and the supply and demand of rights. Cambridge: Cambridge University Press. Korkea-aho, E., (2015). Adjudicating new governance. deliberative democracy in the European Union. London: Routledge. Koskenniemi, M. A., (2004). International law as political theology: how to read the nomos der erde. Constellations. 11(4), 492–511. Krasner, S. D., (1999). Sovereignty: organized hypocrisy. Princeton, N.J.: Princeton University Press. Krisch, N., (2010). Beyond constitutionalism: the pluralist structure of postnational law. Oxford: Oxford University Press. Ladeur, K.-H., (1997). Towards a legal theory of supranationality – the viability of the network concept. European Law Journal [online]. 3(1), 33–54. [Viewed 9 November 2021]. Available from: doi: 10.1111/1468-0386.00018. Ladeur, K.-H., ed. (2017). Public Governance in the Age of Globalization. London: Routledge. Latzer, M. and Festic, N., (2019). A guideline for understanding and measuring algorithmic governance in daily life. Internet Policy Review [online]. 8(2), 1–19. [Viewed 15 April 2021]. Available from: doi: 10.14763/2019.2.1415. Lepenies, W., (1996). Le fondamentalisme occidentale est dépasse [online]. Le Monde. [Viewed 10 November 2021]. Available from: https://​www​.lemonde​.fr/​archives/​article/​1996/​02/​17/​le​ -fondamentalisme​-occidental​-est​-depasse​_3707063​_1819218​.html. Levi-Faur, D., (2005). The global diffusion of regulatory capitalism. The Annals of the American Academy of Political and Social Science [online]. 598(1), 12–32. [Viewed 19 April 2021]. Available from: doi: 10.1177/0002716204272371. Lewin, K., (1951). Field theory in social science: selected theoretical papers. D. Cartwright ed. New York: Harper & Row. Liebman, B. J. and Milhaupt, C. J., eds., (2016). Regulating the invisible hand? The institutional implications of Chinese state capitalism. New York, N.Y.: Oxford University Press. Llewellyn, K. N., (1931). What price contract?—an essay in perspective. Yale Law Journal. 40(5), 704–751. Luo, H. C. and Song, G. D., (2013). Soft law governance: an integrated approach. Translated by B. Armour and T. Hailong. Buffalo, N.Y.: William S. Hein & Co. MacCormick, N., (1995). The Maastricht-Urteil: sovereignty now. European Law Journal. 1(3), 259–266. MacCormick, N., (1999). Questioning sovereignty: law, state, and nation in the European Commonwealth. Oxford: Oxford University Press. Maduro, M., (2009). Courts and pluralism: essay on a theory of judicial adjudication in the context of legal and constitutional pluralism. In: J. Dunoff and J. P. Trachtman, eds. Ruling the world?: Constitutionalism, international law, and global governance. Cambridge: Cambridge University Press. pp. 356–380. Majone, G., (2014). From the regulatory state to democratic default. Journal of Common Market Studies. 52(6), 1216–1223.

26  Research handbook on soft law Marx, K., (1848). Manifesto of the Communist Party. In: K. Marx and F. Engels. Selected works in one volume. London: Lawrence & Wishart. pp. 35–62. Mattei, U. and Nader, L., (2008). Plunder: when the rule of law is illegal. Oxford: Blackwell Publishing. May, C., (2017). Contract as normative regulation and its relation to the rule of law. In: A. C. Cutler and T. Dietz, eds. The politics of private transnational governance by contract. London: Routledge. pp. 39–56. Meckstroth, T. W., (2000). Marx and the logic of social theory: the capitalist state. Science & Society. 64(1), 55–86. Menéndez, A. J., (2017). The crisis of law and the European crises: from the social and democratic rechtsstaat to the consolidating state of (pseudo-)technocratic governance. Journal of Law and Society. 44(1), 56–78. Merry, S. E., (1988). Legal pluralism. Law & Society Review. 22(5), 869–896. Merry, S. E., (2000). Colonizing Hawai’i: the cultural power of law. Princeton, N.J.: Princeton University Press. Merry, S. E., (2016). The seductions of quantification: measuring human rights, gender violence, and sex trafficking. Chicago: University of Chicago Press. Merry, S. E., Davis, K. E. and Kingsbury, B., (2015). The quiet power of indicators: measuring development, corruption, and the rule of law. Cambridge: Cambridge University Press. Merryman, J. H. and Pérez-Perdomo, R., (2019). The civil law tradition: an introduction to the legal systems of Europe and Latin America. 4th ed. Stanford: Stanford University Press. Micklitz, H.-W., (2013). Monistic ideology versus pluralistic reality – towards a normative design for European private law. In: L. Niglia, ed. Pluralism and European private law. Oxford: Hart Publishing. pp. 29–52. Moore, S. F., (1973). Law and social change: the semi-autonomous social field as an appropriate subject of study. Law & Society Review. 7(4), 719–746. Mörth, U., ed., (2004). Soft law and governance: an interdisciplinary analysis. Cheltenham, UK: Edward Elgar. Muller, J. Z., (2002). The mind and the market: capitalism in Western thought. New York: Anchor Books. Murphy, C. N. and Yates, J., (2009). The International Organization for Standardization (ISO): global governance through voluntary consensus. London: Routledge. Murray, R., (1971). The internationalization of capital and the nation state. New Left Review. 1(67), 84–109. Nee, V. and Opper, S., (2014). Markets and institutional change in China. In: S. Fan, R. Kanbur, S.-J. Wei and X. Zhang, eds. The Oxford companion to the economics of China. Oxford: Oxford University Press. pp. 267–272. Nelken, D., Siems, M., Infantino, M., Genicot, N., Restrepo-Amariles, D. and Harrington, J., (2020). COVID-19 and the social role of indicators: a preliminary assessment. EUI Working Paper, 2020/17 [online]. Fiesole: European University Institute (EUI). [Viewed 24 November 2020]. Available from: https://​cadmus​.eui​.eu/​bitstream/​handle/​1814/​68835/​LAW​_2020​_17​_v2​.pdf. Neumann, F. L., (1986). The rule of law: political theory and the legal system in modern society. Berg: Leamington Spa. Neumann, V., (1982). Rechts- und verfassungstheoretische positionen des staatsrechtlichen linken. Der Staat. 21(4), 551–575. Neuwirth, R. J., (2018). Law in the time of oxymora: a synaesthesia of language, logic and law. London: Routledge. Niglia, L., ed., (2013). Pluralism and European private law. Oxford: Hart Publishing. Niglia, L., (2016). Eclipse of the constitution {Europe nouveau siècle}. European Law Journal. 22(2), 132–156. Niglia, L., (2020). The new transformation of Europe: arcana imperii. The American Journal of Comparative Law [online]. 68(1), 151–185. [Viewed 10 November 2021]. Available from: doi: 10.1093/ajcl/avaa005. Nouschi, A., (2012). La depossession fonciere et la pauperisaton de la paysannerie algerienne. In: A. Bouchène, J.-P. Peyroulou, O. Siari-Tengour and S. Thénault, eds. Histoire de l’Algérie à la période coloniale, 1830-1962. Paris: La Découverte. pp. 189–194.

Bamboo, or governance through soft law  27 Peerenboom, R. P., (1993). Law and morality in ancient China: the silk manuscripts of Huang-Lao. Albany, N.Y.: State University of New York Press. Peerenboom, R. P., (2002). China’s long march toward rule of law. Cambridge, UK: Cambridge University Press. Picciotto, S., (1979). The theory of the state, class struggle and the rule of law. In: B. Fine, R. Kinsey, J. Lea, S. Picciotto and J. Young, eds. Capitalism and the rule of law: from deviancy theory to Marxism. London: Hutchinston. pp. 164–177. Piketty, T., (2017). Capital in the twenty-first century. Translated by A. Goldhammer. London, England: Belknap Press of Harvard University Press. Pistor, K., (2019). The code of capital: how the law creates wealth and inequality. Princeton: Princeton University Press. Reich, N., (2013). Pluralism and private law in the Union: some thoughts on the impact of the non-discrimination principle on private autonomy. In: L. Niglia, ed. Pluralism and European private law. Oxford: Hart Publishing. pp. 73–94. Roberts, A., (2015). Napoleon: a life. London: Penguin Books. Roberts, S., (2013). Order and dispute: an introduction to legal anthropology. 2nd ed. New Orleans, L.A.: Quid Pro Books. Rose-Ackerman, S., Lindseth, P. L. and Emerson, B., (2017). Comparative administrative law. 2nd ed. Cheltenham, UK: Edward Elgar Publishing. Rosenau, J. N. and Czempiel, E.-O., eds., (1992). Governance without government: order and change in world politics. Cambridge: Cambridge University Press. Ross, E. A., (1901). Social control: a survey of the foundations of order. London: Macmillan. Reprint, London: Routledge, 2017. Rödl, F. and Callsen, R., (2014). The struggle for union rights under the euro and the dialectics of social integration. In: C. Joerges and C. Glinski, eds. The European crisis and the transformation of transnational governance: authoritarian managerialism versus democratic governance. Oxford: Hart Publishing. pp. 101–122. Ruedy, J., (1967). Land policy in colonial Algeria: the origins of the rural public domain. Berkeley: University of California Press. Sabel, C. F. and Zeitlin, J., (2008). Learning from difference: the new architecture of experimentalist governance in the EU. European Law Journal. 14(3), 271–327. Saiman, C. N., (2018). Halakhah: the Rabbinic idea of law. Princeton: Princeton University Press. Scanlan, M. K., ed., (2017). Law and policy for a new economy: sustainable, just, and democratic. Cheltenham, UK: Edward Elgar Publishing. Scharpf, F. W., (2011). Monetary Union, fiscal crisis and the preemption of democracy. MPIfG Discussion Paper, 11/11 [online]. Cologne: Max Planck Institute for the Study of Societies. [Viewed 6 May 2022]. Available from: http://​hdl​.handle​.net/​11858/​00​-001M​-0000​-0012​-4022​-0. Scharpf, F. W., (2015). After the crash: a perspective on multilevel European democracy. European Law Journal. 21(3), 384–405. Scharpf, F. W., (2017). De-constitutionalisation and majority rule: a democratic vision for Europe. European Law Journal. 23(5), 315–334. Schepel, H., (2005). The constitution of private governance: product standards in the regulation of integrating markets. Oxford: Hart Publishing. Schepel, H., (2017). The bank, the bond and the bail-out: on the legal construction of market discipline in the eurozone. Journal of Law and Society [online]. 44(1), 79–98. [Viewed 11 November 2021]. Available from: doi: 10.1111/jols.12015. Schwab, K. and Malleret, T., (2020). Covid-19: the great reset. Geneva, Switzerland: World Economic Forum Publishing. Scott, C., ed., (2003). Regulation. London: Dartmouth-Ashgate. Scott, C., (2004). Regulation in the age of governance: the rise of the post-regulatory state. In: J. Jordana and D. Levi-Faur, eds. The politics of regulation: institutions and regulatory reforms for the age of governance. Cheltenham, UK: Edward Elgar Publishing. pp. 145–174. Scott, J. and Trubek, D. M., (2002). Mind the gap: law and new approaches to governance in the European Union. European Law Journal. 8(1), pp. 1–18. Senden, L., (2004). Soft law in European Community law. Oxford: Hart Publishing.

28  Research handbook on soft law Seppӓnen, S., (2016). Ideological conflict and the rule of law in contemporary China: useful paradoxes. Cambridge, UK: Cambridge University Press. Shelton, D., ed. (2000). Commitment and compliance: the role of non-binding norms in the international legal system. Oxford: Oxford University Press. Shiva, V., (2015). Earth democracy: justice, sustainability, and peace. Berkeley, C.A.: North Atlantic Books. Snyder, F. G., (1981). Capitalism and legal change: an African transformation. New York: Academic Press. Snyder, F. G., (1982). Colonialism and legal form: the creation of "customary law" in Senegal. In: C. Sumner, ed. Crime, justice, and underdevelopment. London: Heinemann. pp. 90–121. Snyder, F. G., (1985). Law of the common agricultural policy. London: Sweet and Maxwell. Snyder, F. G., (1990). New directions in European Community law. London: Weidenfeld and Nicolson. Snyder, F. G., (1993). The effectiveness of European Community law: institutions, processes, tools and techniques. The Modern Law Review. 56(1), 19–54. Snyder, F. G., (1998). International trade and customs law of the European Union. London: Butterworths. Snyder, F., (1999). Governing economic globalization: global legal pluralism and European Union law. European Law Journal. 5(4), 334–374. Snyder, F. G., (2001). The origins of the "nonmarket economy": ideas, pluralism and power in EC antidumping law about China. European Law Journal. 7(4), 369–424. Snyder, F., (2004). Economic globalization and the law in the 21st century. In: A. Sarat, ed. The Blackwell companion to law and society. Oxford: Blackwell Publishing. pp. 624–640. Snyder, F. G., ed., (2009). The European Union and China, 1949-2008: basic documents and commentary. Oxford: Hart Publishing. Snyder, F. G., (2010). The EU, the WTO, and China: legal pluralism and international trade regulation. Oxford: Hart Publishing. Snyder, F., (2011). EMU – integration and differentiation: metaphor for European Union. In: P. Craig and G. de Búrca, eds. The evolution of EU law. Oxford: Oxford University Press. pp. 687–716. Snyder, F. G., (2016). Food safety law in China: making transnational law. Leiden: Brill Nijhoff. Snyder, F., (2020). Understanding the regulation of ecological food in China: regulatory intermediation, path dependence and legal pluralism. In: G. Bee Chen and R. Price, eds. Regulatory issues in organic food safety in the Asia Pacific. Singapore: Springer Nature. pp. 11–34. Snyder, F., (2021a). Three worlds of soft law in China. In: M. Eliantonio, E. Korkea-aho and O. Ştefan, eds. EU Soft Law in the Member States: theoretical findings and empirical evidence. Oxford: Hart Publishing. pp. 263–280. Snyder, F., (2021b). The BRI kaleidoscope: European legal pluralism and BRI soft law. In: V. Ntousas and V. Minas, eds. The European Union and China’s Belt and Road: impact, engagement and competition. London: Routledge. Snyder, F., (2021c). EU, China, and technical standards in the Belt and Road Initiative (BRI): extraterritoriality or transnational governance?. In: N. Cunha Rodrigues, ed. Extraterritoriality of EU economic law: the application of EU economic law outside the territory of the EU. Singapore: Springer Nature. pp. 175–195. Snyder, F., (2021d). The EU, China and product standards. In: S. Bersick, K. Brown, A. Cottey, J.-C. Gottwald, W. Shen, eds. Routledge handbook on EU-China relations. London: Routledge. Snyder, F. and Kim, Y. S., (2018). China’s 2015 food safety law: crossing the river but feeling the stones and avoiding low-hanging branches? The Chinese Journal of Comparative Law [online]. 6(1), 1–49. [Viewed 11 November 2021]. Available from: doi: 10.1093/cjcl/cxy004. Snyder, F. and Ni, L., (2017). A tale of eight pesticides: risk regulation and public health in China. European Journal of Risk Regulation. 8(3), 469–505. Snyder, F., Eliantonio, M. and Jin, Z., (Forthcoming). The use of soft law by national courts and public authorities: an EU-China comparison. China-EU Law Journal. Speth, J. G., (2008). The bridge at the end of the world: capitalism, the environment, and crossing from crisis to sustainability. New Haven, C.T.: Yale University Press. Speth, J. G. and Courrier, K., eds., (2021). The new systems reader: alternatives to a failed economy. New York, N.Y.: Routledge.

Bamboo, or governance through soft law  29 Ştefan, O., (2013). Soft law in court: competition law, state aid and the Court of Justice of the European Union. Alphen aan den Rijn: Kluwer Law International. Ştefan, O. A., Avbelj, M., Eliantonio, M., Hartlapp, M., Korkea-aho, E. and Rubio, N., (2019). EU soft law in the EU legal order: a literature review. King's College London Law School Research Paper [online]. [Viewed 11 November 2021]. Available from: doi: 10.2139/ssrn.3346629. STL News., (2019). STL hosts international workshop on soft law [online]. Peking University School of Transnational Law. [Viewed 9 October 2021]. Available from: https://​stl​.pku​.edu​.cn/​stl​-news/​stl​ -hosts​-international​-workshop​-on​-soft​-law. Stone Sweet, A., (1999). Judicialization and the construction of governance. Comparative Political Studies. 32(2), 147–184. Sutherland, V., (2021). The intersection of race and algorithmic tools in the criminal legal system. Maryland Law Review [online]. 80(3), 487–566. [Viewed 11 November 2021]. Available from: https://​ssrn​.com/​abstract​=​3797102. Tamanaha, B. Z., (1997). Realistic socio-legal theory: pragmatism and a social theory of law. Oxford: Oxford University Press. Tamanaha, B. Z., (2004). On the rule of law: history, politics, theory. Cambridge: Cambridge University Press. Trubek, D. M. and Trubek, L. G., (2005). Hard and soft law in the construction of social Europe: the role of the open method of co-ordination. European Law Journal. 11(3), 343–364. Trubek, D. M., Cottrell, P. and Nance, M., (2005). “Soft law,” “hard law,” and European integration: toward a theory of hybridity. Univ. of Wisconsin Legal Studies Research Paper No. 1002 [online]. [Viewed 2 February 2021]. Available from: doi: 10.2139/ssrn.855447. Trubek, D. M., Dezalay, Y., Buchanan, R., and Davis, J. R., (1994). Global restructuring and the law: studies of the internationalization of legal fields and the creation of transnational arenas. Case Western Reserve Law Review. 44(2), 407–498. Tuori, K., (2014). Transnational law: on legal hybrids and perspectivism. In: M. P. Maduro, K. Tuori and S. Sankari, eds. Transnational law: rethinking European law and legal thinking. Cambridge: Cambridge University Press. pp. 11–57. Viellechner, L., (2009). The network of networks: Karl-Heinz Ladeur's theory of law and globalization. German Law Journal [online]. 10(4), 515–536. [Viewed 11 November 2021]. Available from: doi: 10.1017/S207183220000119X. Walker, N., (2002). The idea of constitutional pluralism. The Modern Law Review. 65(3), 317–359. Walker, N. and de Búrca, G., (2007). Reconceiving law & new governance. Columbia Journal of European Law. 13(3), 519–537. Wang, H., (2021). The Belt and Road Initiative agreements: characteristics, rationale, and challenges. World Trade Review. 20(3), 282–305. Warren, B., (1971). The internationalization of capital and the nation state: a comment. New Left Review. 68(1). Weber, M., (1954). Max Weber on law in economy and society. 2nd ed. Trans E. Shils and M. Rheinstein. Cambridge, M.A.: Harvard University Press. Weil, S., (1952). The need for roots: prelude towards a declaration of duties towards mankind. London: Routledge & Kegan Paul. Weiler, J. H. H., (1991). The transformation of Europe. The Yale Law Journal. 100(8), 2403–2483. Weiler, J. H. H., (2012). Prologue: global and pluralist constitutionalism – some doubts. In: G. de Búrca and J. H. H. Weiler, eds. The worlds of European constitutionalism. Cambridge: Cambridge University Press. pp. 8–18. Weiler, J. H. H., (2020). COVID, Europe, and the self-asphyxiation of democracy. In: M. P. Maduro and P. W. Kahn, eds. Democracy in times of pandemic: different futures imagined. Cambridge, UK: Cambridge University Press. pp. 141–152. Wellens, K. C. and Borchardt, G. M., (1989). Soft law in European Community law. European Law Review. 14(5), 267–321. Williamson, O. E., (1985). The economic institutions of capitalism: firms, markets, relational contracting. New York: Free Press.

30  Research handbook on soft law Wolff, J. and Leopold, D., (2021). Karl Marx [online]. Spring 2021 ed. E. N. Zalta ed. The Stanford Encyclopedia of Philosophy. [Viewed 19 April 2021]. Available from: https://​plato​.stanford​.edu/​ archives/​spr2021/​entries/​marx. WTO., (1998). WTO Appellate Body Report: European Communities – measures concerning meat and meat products (Hormones). AB-1997-4, WT/DS26/AB/R, WT/DS48/AB/R (98-0099), 16 January 1998. WTO., (2005). WTO Appellate Body Report: United States – measures affecting the cross-border supply of gambling and betting services. AB-2005-1, WT/DS285/AB/R (05-1426), 7 April 2005. Wu, Q., (2013). Competition laws, globalization and legal pluralism: China’s experience. Oxford: Hart Publishing. Wulf, A., (2016). The invention of nature: Alexander von Humboldt’s new world. New York: Vintage Books. Zakaria, F., (2020). Ten lessons for a post-pandemic world. London: Allen Lane, Penguin Books. Zarsky, T., (2016). The trouble with algorithmic decisions: an analytic road map to examine efficiency and fairness in automated and opaque decision making. Science, Technology, & Human Values. 41(1), 118–132. Zumbansen, P., ed., (2020). The many lives of transnational law: critical engagements with Jessup’s bold proposal. Cambridge, UK: Cambridge University Press.

2. Soft law: an historical introduction Nils Jansen

1. INTRODUCTION For most of its history, European law has been soft law. Indeed, pre-modern jurists could never assume that the law could be found in binding legislation in the same way in which modern civilian jurists derive their law from codifications and other legislation, and in the way in which common lawyers may rely on strictly binding precedents. True, the ideas that the emperor’s will was vested with the force of law, and that legislation was an exclusive prerogative of princes can be traced back to Roman law and to mediaeval legal learning (Ulpian, D. 1,4,1; Justinian, C. 1,14,12; Glossa ordinaria,1 Gl. suo proprio to D. 1,1,9). Yet, before the 19th century, legislation was at best pointillist; it was difficult to find and often forgotten after a few years. Jurists therefore had to rely on, and work with, non-legislative textual authorities, which contemporary jurists might describe as soft law. In those days, ‘soft’ was the normal state of the law, and binding ‘hard’ legislation was the exception. Section 2 of this chapter will first describe the pre-modern legal world, which for the purposes of this chapter was the legal world pre-dating the great codifications such as the French Code civil or the Austrian Allgemeines Bürgerliches Gesetzbuch. In particular, I will characterize some of the leading pre-modern textual authorities before analyzing the modern genealogy of strict law. The story ends with the modern codifications and contemporary legislative law-making. Section 3 of this chapter considers the modern emergence of different forms of soft law, namely the German ‘free-law’ movement, the rise of judge-made law, and the making of non-legislative codifications. Many other forms of modern soft laws cannot be included in this overview; they are covered by the following chapters.

2.

AUTHORITIES IN PRE-MODERN LAW

The histories of modern legal systems begin in the first decades of the 12th century. Indeed, before the establishment of law schools in Bologna and the King’s court in London, both during the 12th century, it would have been very difficult to clearly distinguish legal issues from politics, religion, or other issues. It was only with the law schools in Upper Italy and the English King’s court that professions of lawyers emerged, which were to become the nuclei of legal systems with their specific cultures of legal learning and legal justice (Jansen 2019, pp. 27–77). 2.1

Non-legislative Codifications

None of those jurists could possibly rely on a sovereign’s comprehensive legislation; indeed, there existed very little law in those early days. Therefore, the mediaeval jurists themselves had to produce the texts on which they would base their practices. While English judges began 31

32  Research handbook on soft law to report cases early on and treated them as sources of the law,2 a group of philologically skilled jurists in Bologna3 re-constructed a vulgate of Justinian’s Digest (1080), which was to become the textual basis of Roman legal scholarship (Radding 1988, p. 156; Radding and Ciaralli 2007, pp. 169–210). A few decades later, another group of canon lawyers produced the Decretum Gratiani (concordia discordantium canonum, 1140), a collection of conflicting authorities of the Roman Church’s canon law (Christensen 1993, p. xii; Helmholz 1996, p. 7). Until the 18th century, the Digest, the core part of the Corpus iuris civilis, was treated as the main authority of the learned ius commune all over Europe. From a modern perspective, such authority may appear ‘mysterious’ (Donahue 2008, p. 129), as no Emperor or Prince had declared it binding (Jansen 2010, pp. 20, 28). Indeed, the only reason for the Corpus iuris to become authoritative was that learned lawyers used it as the standard textual reference within their legal discourse. The so-called Lothar legend, according to which the use of the Digest had been prescribed around 1135 by the Emperor Lothar III of Supplinburg, was a political fairy tale invented ex post to legitimize the applicability of Roman law. It did not surface before the 16th century, and it was never taken seriously by the legal profession.4 During the 18th century, jurists would rather explain the application of Roman sources with their reception in legal science, and secondarily with the (alleged) tacit approval of this reception process by governments (Stryk 1713). A similar picture emerges with the Decretum Gratiani, which clearly was no official text either. Rather it was the work of a private group of scholars who worked independently of any official commission (for detail, see Jansen 2010, p. 21). The text brought together canons of church councils and letters and decretals of Popes with excerpts from Roman law, from the Scripture and from the church fathers’ writings. Gratian related all those texts to specific issues of actual canon law and thus integrated them into a single legal text. This book was soon to become a remarkable success: within few years, Gratian’s Decretum standardized canon law. Jurists and theologians treated it as authority in itself. Indeed, it was only this text that made canon law the object of systematic academic study in specialized canon law schools; for the centuries to come it was the main textual basis of positive canon law. A third text worth mentioning in the present context is the Saxon Mirror, which, like other mediaeval and early modern law books such as the French coutumes or the Siete partidas in Castile, was meant to record in writing the traditional customs of a place and to update them in an orderly and authoritative form (Bertelsmeier-Kierst 2008, p. 61; Lück 2013, p. 16). Yet, those texts never simply described the respective customs: the Saxon Mirror clearly exhibits innovative canon law learning and altogether presents a modernized statement of the law. Like the Corpus iuris civilis and the Decretum, this text was never enacted as formal legislation or otherwise put into legal force. It was unofficial law, yet it was nevertheless soon recognized as a leading authority of Saxon law in large parts of the empire. As with the Corpus iuris, a – clearly wrong – legend (according to which Charlemagne had granted the Mirror as a privilege to the Saxon people) was used to legitimize the use of the text (Jansen 2010, pp. 23, 26). Such fairy tales were an expression of political metaphysics. Jurists believed in the imperial authority of a text because they felt obliged to apply it. All those texts functionally resembled modern codifications; yet none of them was based on a sovereign’s command. Even if it might be misleading to describe those texts as private documents, they clearly had a non-legislative character (Jansen 2010, p. 6) and could therefore never be regarded as strictly binding. Jurists rather treated them as legal authorities (Jansen 2010, p. 43): they were accepted by the legal profession as ultimate sources of the law. Jurists

Soft law  33 assumed that they were expected to argue on the basis of these texts, and that no argument was needed to justify their application. Yet, in the legal practice of those days, such texts often collided with other authorities; there could be colliding special statutes and privileges of a more strictly binding character – even if such privileges applied only to specific parts of the population, such as merchants, noblemen, or clerics (Jansen 2012). Hence, it had to be decided which of those texts should be granted priority in legal argument, or how a statute should be interpreted in view of relevant non-legislative authorities. Such decisions depended on the place, on the (secular or ecclesiastical) court, and on many other factors. They gave expression to the relative authorities of the relevant texts, and it was perfectly clear that the weight of such authority could change over time and differ across territories. Like every social institution, legal authority depended on constant processes of recognition.5 Yet, such recognition did not relate to a legal system as a whole, but rather directly pertained to particular texts, and it vested those texts not with formal, but rather with informal authority.6 2.2

Law and Political Domination

In the 17th century, the pluralist state of the law with its many non-legislative authorities became increasingly untenable. On the one hand, conflicts between these texts mutually qualified their respective authority and thus largely undermined the law’s claim to finally determine normative conflicts. Nobody could be sure about what the law actually was; indeed, the many new treatises of learned authors paradoxically contributed to the legal uncertainty, as such treatises were often cited in support of contrary opinions,7 and for every opinion a contrary opinion could be found somewhere (Duve 2003, pp. 243–247). On the other hand, political theorists increasingly attacked the legal practices of their times and their foundation in learned Roman law. Those critics were the contemporaries of fierce religio-political controversies and of brutal civil wars; hence, they were convinced that all power should lie with a strong sovereign ruler. It followed that the administration of the law should be entrusted not to academic elites, but rather to the political government. From the perspective of those writers, it was evident that claims for normative truths had been the source of illegitimate suppression, terror, and violence; this is why the polity (civitas) could not be ruled on the basis of possibly true assertions of theologians and jurists (veritas), but by means of procedurally legitimate law, i.e., the commands of the competent authority representing the people (authoritas): ‘authoritas, non veritas facit legem’ (Hobbes 2012).8 The emerging state was therefore called to control the law (Bodin 1576);9 even the common law, so Hobbes claimed, was not the law of reason, controlled by an elite group of jurists, but rather had to be understood as the creation of the English sovereign (Tuck 1993, p. 343). Natural law theorists such as Grotius and Pufendorf accordingly insisted that the law (lex) was a binding command and thus distinguished it from a merely persuasive authority (consilium).10 German intellectuals like Hermann Conring sought to de-legitimize the learned jurisprudence of their time and called for a national codification of the law (Conring 1665, cap. XXXV; Stintzing 1884, p. 176; Luig 1983, p. 377); the French politicians Colbert and d’Aguesseau began to regulate increasingly large parts of the law by means of ordonnances; and in the 19th century, Jeremy Bentham finally coined the concept of a ‘codification’ (Weiss 2000, p. 448).

34  Research handbook on soft law Many lawyers, of course, were critical of such developments. Their interest was contracts, testaments, taxes, and accidents, and with those issues, it was highly unlikely that governments could help by means of a more comprehensive legislation (Stryk 1713, section 42). Indeed, in early modern times, there were very few examples of successful, convincing codifications. Nevertheless, with the new spirit of the enlightenment, the pre-modern plural state of the law with its many non-legislative authorities became untenable. During the 18th and 19th centuries, increasingly large parts of the law were codified in most (continental) legal systems (Zimmermann 1995, 2012; Jansen 2010, p. 14). Parallel developments can be observed in the common law, where precedents had long been seen as authoritative, though not strictly binding. Here, the theory of stare decisis transformed precedents into formally binding, hard law (Cross and Harris 1991, pp. 97, 108; Vogenauer 2006, p. 64). Everywhere, jurists seemed to believe that the state was the source of all law (see also Ehrlich 1906, p. 425; Reinhard 2002, p. 281; Haferkamp 2008, p. 258; Jansen and Michaels 2008, p. 16).

3.

THE RE-DISCOVERY OF SOFT LAW

It did not take long, however, for jurists to understand that the law could never be clearly and conclusively determined by codifications and other legislation, and that a state’s government could not fully control the law’s development. Even if the concept of ‘soft law’ is today almost exclusively used in the context of international (including European) legal governance (see also Shaffer and Pollack 2010), changing perceptions of national law at the turn from the 19th to the 20th century were to become the decisive turning point in the history of soft law thinking. They made it possible to consider alternatives to the state’s ‘hard’ legislation. Indeed, important modern international soft law instruments were modelled on the American restatements, which were neither international nor instruments of governance. Thus, to understand the history of soft law, we have to look at the developments after 1900 within the legal space of nation-states. Codifications were not usually drafted by government administrators, but rather by the commissions of leading professors and judges, and they were normally meant to restate, rather than to reform, the law. It was obvious that they neither fundamentally changed the established legal rules and principles, nor shifted the power from the legal profession to political elites. Instead, they rationally re-ordered and systematically revised inherited rules, concepts, and doctrines, they clarified formerly disputed issues, and they provided the legal system with new normative foundations (see Zimmermann 1995, p. 98; Jansen 2010, p. 14; 2019, p. 192). Even more importantly, when the codifications had been enacted, it soon turned out that they were not able to determine future legal development. Their relatively few rules did not give much guidance in difficult and new cases. Courts obviously felt free to continue earlier lines of jurisprudence, and a codification never prevented later legal change by academic argument and judicial practice (Bürge 1995, pp. 174, 254; Falk and Mohnhaupt, 2000; Zimmermann 2001, pp. 55, 98; Volkmann 2018, p. 270). The codification process never meant that the law would lose its autonomy vis-à-vis the political system.

Soft law  35 3.1

Free Law

In view of those developments, jurists began critically to reflect the assumption that the law was based on the state’s official legislation. One aspect of this development was the ‘discovery’ of ‘free law’ by German-speaking scholars, including Hermann Kantorowicz and Eugen Ehrlich. With this idea of free law existing besides the state’s legislation, Kantorowicz primarily aimed to attack courts and scholars pretending that they could act as bouches de la loi and could decide cases on the thin basis of the codifications and proper legal method (Gnaeus Flavius [Kantorowicz] 2002). According to Kantorowicz, legal systems grew ‘organically’. They were pluralistic in nature, colourfully complex, often quite incoherent and inconclusive (Gnaeus Flavius [Kantorowicz] 2002, p. 15). Jurists had to open their eyes and minds to the informal processes of reception, interpretation, and recognition in the law’s development (Kantorowicz and Patterson 1928, p. 692; Kantorowicz and Campbell 1958, p. 74), and judges had to assume and reflect their individual responsibility for the decisions they delivered in their daily practice (Gnaeus Flavius [Kantorowicz] 2002, p. 5; see also Kantorowicz 1911, p. 15; Auer 2015, pp. 794, 797). Jurists should stop pretending that legal thinking was a logical practice, or that concepts could be applied mechanically (Kantorowicz 1912, p. 76; Gnaeus Flavius [Kantorowicz] 2002, p. 25). It is important to understand that for Kantorowicz and other members of the ‘free-law’ movement, free law was not an ideal of legal justice or fairness. Rather, it was a real, positive aspect of the law (Gnaeus Flavius [Kantorowicz] 2002, p. 24). It was not necessarily ‘better’ or more just than the state’s legislation. The important point was that it did exist, and that it had an important role to play in the legal system and in judicial decision-making. Of course, Kantorowicz did not speak of ‘soft law’, but his point was that the law could not be equated with the state’s hard legislation. Yet, for the same reason that the law could not be controlled by means of legislation, free law could not be fixed in the form of modern soft law instruments or other normative texts. For Kantorowicz, it would have been pointless to restate free law. Thus, he never used his concept of free law to develop ideas or instruments for international law. Nevertheless, the ‘free-law’ movement was a highly significant step in the history of soft law in that it paved the way intellectually for the re-discovery of law ‘beyond the state’. 3.2

The Rise of the Courts

While the debates on free law and judicial decision-making laid the intellectual foundations for scholarly thinking about legal alternatives to hard, binding legislation, the rise of the courts and case law made it obvious for every lawyer that the law could not be reduced to the state’s official legislation. Jurists had always known that courts, for various and good reasons, normally followed precedent. Thus, wherever courts publish their decisions, they become, in the eyes of jurists, (at least) ‘unofficial’ or ‘informal’ sources of the law (Volkmann 2018, p. 266). Even codified parts of the law thus assume a case-law character (Jansen 2011, p. 35). Indeed, during the 19th and 20th centuries, the codifications in civilian legal systems such as France and Germany have been reshaped and overwritten with thick layers of judge-made law (Albers 2012; Zimmermann 2012, pp. 380, 382; Payandeh 2017). As a result, the authority accorded to the decisions of supreme courts in civilian legal systems has significantly increased during the last decades. Kantorowicz was quite right in insisting on the significance of informal processes of reception, interpretation, and recognition in the law’s development

36  Research handbook on soft law (see above 3.1); today such processes have become characteristic features of judge-made law (Tushnet 1983, p. 813; Duxbury 2008, p. 113). Yet, the role of scholars in those processes is diminishing, as the centre of civilian legal systems has been shifting from the law faculties to supreme courts such as the European Court of Justice, the French Cour de cassation, or the German Bundesverfassungsgericht and Bundesgerichtshof (Jansen 2018, p. 627). Thus, judge-made law has become the main source of the law and of legal doctrine. Now, in most legal systems, judge-made law is soft law: absent a strict rule of stare decisis, precedents are never strictly binding; their authority depends on factors such as the authority of the court and the persuasiveness of its arguments, on conceptually and systematically stabilized doctrinal assumptions, and on informal, often habitual, normative and political expectations of the jurists involved (Jansen 2019, p. 296). Unlike precedents in the common law world, they do not normally apply in a freestanding manner as independent sources of the law. Rather, jurists treat them as authoritative interpretations of the relevant legislation. Civilian precedents thus clarify, determine, and even change the meaning of legislation. For civilian jurists today, the law has thus altogether attained an increasingly soft structure. 3.3

The American Restatements

While non-legislative codifications in pre-modern times were a specific feature of civil law (see above 2.1), the ancestor of modern non-legislative codifications are the American restatements. At the beginning of the 20th century, when it had become clear that attempts to codify the common law had failed (for the United States, Crystal 1979; Cook 1981; for England, Braun 2012), those restatements were designed for similar purposes, yet as a soft law alternative (avant la lettre) to the European codifications. A modern restatement looks quite similar to, but not exactly like, a codification. Like a codification, restatements consist of short provisions, which follow an intellectual order that makes them easily accessible. Unlike codifications, however, the restatements complement those rules with comments and illustrations.11 Moreover, the restatements have not been systematized as rigidly as their European counterparts (American Law Institute (ALI) 1923, p. 28), and they are often more detailed and precise than the provisions in the classical codes. Yet, the most important difference is that the restatements were not adopted by statute (ALI 1923, pp. 19, 23). Indeed, they were never enacted by an official, governmental body. Instead, they are authorized by an informal, non-governmental, private institution, namely the American Law Institute (for all Jansen 2010, p. 50), and the founders of this institute made it quite clear that they did not wish judges to be formally bound to their provisions, but rather wanted to preserve their judicial discretion. Remarkably, however, the restatements were nevertheless designed to become legal authorities. Even if they were to be ‘less than a code’, they should be much ‘more than a treatise’ (Cardozo 1924, pp. 9, 16). Although the restatements should not be ‘formally adopted by legislatures’ and thus were constructed in the form of non-legislative codifications, they should be given ‘such authority as is now accorded a prior decision of the highest court of the jurisdiction’ (ALI 1923, p. 25; see also p. 29). In hindsight, this was an extraordinary claim, not only because a group of private actors (though of highest prestige: Jansen 2010, p. 100)12 pretended to represent the American people (ALI 1923, p. 29; Jansen 2010, pp. 54, 100). The more innovative idea was to draft legal instruments in the form of statutory law and with ‘the mental attitude … of those who desire to express the law in statutory form’ (ALI 1923, p. 20) and nonetheless vest them only with some sort of soft, rather

Soft law  37 than hard authority. Yet, the most remarkable aspect of the story is perhaps the success of the restatement process. The use of the restatements as ultimate sources of the law by the courts soon became pervasive,13 and the restatements have widely been taken as the basic reference texts for the purpose of academic teaching (von Mehren 1988, pp. 21; Frank 1998, p. 638). They have thus become the first soft instruments in modern law. 3.4

Non-legislative Codifications of European and Transnational Law

During the second half of the 20th century, the restatements have become a model also for international private law unification. Today, soft, non-legislative codifications are a characteristic element of both European and transnational private law, the most important examples of such codifications being the Principles of European Contract Law (PECL) (Lando and Beale 2000; Lando et al. 2003) and the UNIDROIT Principles of International Commercial Contracts (PICC) (UNIDROIT 2016). The PECL were an attempt of a group of comparative law scholars in the 1980s and 1990s to show that it would be possible to construct common European rules on the basis of the many different national laws of Europe. Yet, the PECL were modelled on the American restatements not only in that they aim to formulate common principles on the basis of the laws of different jurisdictions. Like the restatements, they consist of systematically arranged rules, complemented with Comments – unlike the restatements, however, the PECL complement the Comments with Notes which contain a comparative overview and thus are meant to show the common basis of the rule (or sometimes the lack of such a basis). This accords with the more academic nature of the PECL (see Jansen 2010, p. 134). Moreover, like the restatements, the PECL were drafted by a ‘private’ group of legal experts (the Lando Commission) claiming to represent the different European nations in their membership (Lando 2000, p. xii), hence their non-legislative character. And finally, again like the restatements, they were constructed as a soft law instrument in that they are ‘intended to be applied as general rules of contract law’ in Europe (Article 1:101, section 1 PECL). Similarly, the UNIDROIT PICC, the most important non-legislative codification of transnational commercial contract law, are modelled after the American restatements particularly as far as their soft law character is concerned (Jansen 2010, p. 66). Like the PECL, the PICC were written by an international group of legal experts, a group which, unlike the Lando Commission, did not however constitute itself, but was established by UNIDROIT, an intergovernmental organization founded in 1926 for the harmonization of private law.14 While the usual working instruments of UNIDROIT are international conventions or model laws, those forms were regarded as inappropriate, as far as the law of commercial contracts was concerned, and preference was given to a soft, non-legislative reference text (Vogenauer 2015, p. 7). The main motives were that only a non-binding instrument would avoid the necessity of being adopted by the individual member states (UNIDROIT 2004, p. xiv) and could hence be formulated without a view to future political negotiations and compromises with national governments. Moreover, such an instrument could be easily revised at a later stage (which, in fact, has regularly been done since their inception; see UNIDROIT 2016). The PICC were thus designed as persuasive authority; they shall not be formally binding (UNIDROIT 2004, p. xv; Bonell 2005, p. 25). Accordingly, they have not been formally approved by the Institute’s council; rather the council informally ‘authorised their publication’ and recommended the

38  Research handbook on soft law publication’s ‘widest possible distribution’ (UNIDROIT 1994, p. 22). Thus, the PICC were officially published under the Institute’s name as if they had been formally approved. Today, soft, non-legislative codifications like the PICC and the PECL have become essential elements in the complex structures of European and transnational contract law. While the PICC are regularly used in international arbitration (Bortolotti 2000, p. 142; Jansen 2010, p. 72; Scherer 2015, p. 27) and have become a general part of transnational contract law complementing other instruments such as the CISG (Michaels 2009, p. 885), both instruments, whose rules do not differ very much from one another, are regarded as authoritative statements of European and transnational contract law, respectively. Legislators and courts have used them to modernize their national laws and to align their laws with the international state of art (Jansen 2010, pp. 62–74; Michaels 2015, p. 143). At the same time, international and national scholars began to regard those principles as general sources of the law. Moreover, those Principles have become models for many, more recently formulated layers of European law, including other non-legislative codifications, such as the Acquis Principles (Acquis Group 2009) or the Principes contractuels communs (Fauvarque-Cosson and Mazeaud 2008), draft codifications, such as the DCFR (von Bar and Clive 2009), proposals for European instruments, such as the Common European Sales Law (CESL)), and even European legislation such as the Consumer Rights Directive15 (Jansen and Zimmermann 2018, p. 9 with detailed further references). Thus, they have grown into an increasingly dense net of transnational contract law.

4. CONCLUSION Of course, the narrative of the history of soft law that I have presented here is radically incomplete. Many jurists would presumably look for soft law primarily in the fields of public international and European law. Yet, my reason for focusing on soft private law is that it allows me to present a broader perspective of soft law, also extending to the pre-modern ius commune. This inclusion of pre-modern legal systems enables us, in turn, to understand that soft law is not a feature unique to modernity. The law has been soft at all times; during the 19th century, the state only complemented, rather than replaced, traditional forms of soft law with a hard structure of codifications, statutes, regulations, directives, treatises, and other international legal instruments. Only during those high times of the nation-state was it thought that all law should be expressed in the form of hard legislation. Today, jurists all over the world again understand the omnipresence of soft law, and they begin to focus on the complex interplay of soft and hard elements of the law.

NOTES 1. The Glossa ordinaria was the authoritative final compilation of mediaeval legal learning by Accursius; until the 17th century, nearly all editions of Roman law were complemented with this gloss. I am using here a large, revised version of 1552/53 (Accursius 1552). 2. On the bureaucratization of the King’s court, see Hudson 2012, pp. 523, 851; for the early reporting of cases, see Baker 2019, p. 188.

Soft law  39 3. Later, it was even told that Irnerius, the Bologna’s school founding father, had been a magister artium, rather than a learned jurist (see Odofredus [1552] 1968, to C. 1,2, auth. Qui res iam [fol. 17 v.]; von Savigny 1850, p. 18). 4. Indeed, the legend was never part of the legal, but rather of the general political discourse; it was told by authors such as Melanchthon, rather than by jurists (see Jansen 2010, p. 34 with further references). 5. For the mechanisms of such processes of social recognition, see Berger and Luckmann 1966, pp. 48, 119; Rehberg 1994, p. 58. 6. For the concept of informal authorities (in the common law), see Schauer 2008, p. 1956. 7. See, as an example, the argumentation of a judge at the Reichskammergericht in Oestmann 2009, p. 567. 8. Hobbes 2012 (vol. II, cap. 26), p. 431: ‘In Civitate constituta, Legum Naturae Interpretatio non à Doctoribus & Scriptoribus Moralis Philosophiae dependet, sed ab Authoritate Civitatis. Doctrinae quidem verae esse possunt; sed Authoritas non Veritas facit legem’. See Tuck 1993, pp. 279–348; Jansen 2019, p. 199. 9. Bodin 1576 (book I, c. VIII), p. 135; cf. Allison 1996, p. 45; van Crefeld 1999, p. 175, both with further references. 10. Grotius 1625, lib. I, cap. I, section IX, n. 1 and cap. III, section VI, n. 1; Pufendorf 1759, lib. I, cap. VI, section 1; Hobbes 1983, cap. XIV, section 1; 2012 (vol. II, cap. 25 and 26), pp. 398, 414. 11. Those comments are not only meant to explain the rule and its meaning more specifically, but also to add authority to the rule (Jansen 2010, p. 126). 12. Thus, when the Institute was founded, the Times described it as ‘probably the most distinguished gathering of the legal profession in the history of the country’. The leading members of the Institute were powerful politicians such as William H. Taft, the former President of the United States, high-ranking diplomats, and administrative lawyers such as George W. Wickersham, a former Attorney General, or Elihu Root, a secretary of the state under President Roosevelt, influential judges (among them Charles E. Hughes, Benjamin Cardozo, Learned Hand, and also the sceptical Oliver Wendell Holmes), well-known academics, such as Roscoe Pound and John H. Wigmore, and leading representatives of the American Bar. 13. Unfortunately, there is no comprehensive analysis; for examples, see Maggs 1998, pp. 512–517; Eisenberg 2002, p. 1251; Zekoll 2008, p. 115. 14. For the history of UNIDROIT, see Vogenauer 2015, p. 14. Currently, 65 states from all parts of the world and with different economic, political, and cultural backgrounds have signed the UNIDROIT agreement and thus are members of the Institute: https://​www​.unidroit​.org/​about​-unidroit/​member​ -states​-2. 15. Directive 2011/83/EU.

REFERENCES Accursius, F., (1552). Digestorum, institutionum, codicis d Iustiniani … libri. Lyon: Hugo à Porta. Albers, M., (2012). Höchstrichterliche Rechtsfindung und Auslegung gerichtlicher Entscheidungen. Veröffentlichungen der Vereinigung der Deutschen Staatsrechtslehrer. 71, 257–295. Allison, J. W. F., (1996). A continental distinction in the common law: a historical and comparative perspective on English public law. Oxford: Oxford University Press. American Law Institute (ALI)., (1923). Report of the Committee on the Establishment of a Permanent Organization for the Improvement of the Law proposing the establishment of an American law institute [online]. Philadelphia, P.A.: Proceedings of the American Law Institute. 1, pp. 1–109. [Viewed 17 November 2021]. Available from: https://​www​.ali​.org/​media/​filer​_public/​bc/​d6/​bcd6e5b4​-682f​-46a4​ -b5f4​-855f7a5ec84c/​report​-establishment​-ali​.pdf. Auer, M., (2015). Der Kampf um die Wissenschaftlichkeit der Rechtswissenschaft – zum 75. Todestag von Hermann Kantorowicz. ZEuP. 23, 773–805. Baker, J., (2019). An introduction to English legal history. 5th ed. Oxford: Oxford University Press.

40  Research handbook on soft law Bar, C. von and Clive, E., eds., (2009). Principles, definitions and model rules of European private law: Draft Common Frame of Reference (DCFR). Full ed. Munich: Sellier. Berger, P. L. and Luckmann, T., (1966). The social construction of reality: a treatise in the sociology of knowledge. New York: Anchor Books. Bertelsmeier-Kierst, C., (2008). Kommunikation und Herrschaft: zum volkssprachlichen Verschriftlichungsprozeß des Rechts im 13. Jahrhundert. Stuttgart: Hirzel. Bodin, J., (1576). Les six livres de la république. Paris: Jacques du Puys. Bonell, M. J., (2005). An international restatement of contract law: the UNIDROIT principles of international commercial contracts. 3rd ed. Ardsley, N.Y.: Transnational Publishers. Bortolotti, F., (2000). The UNIDROIT Principles and the arbitral tribunals. Uniform Law Review - Revue de droit uniforme. 5(1), 141–152. Braun, A., (2012). The English codification debate and the role of jurists in the development of legal doctrines. In: M. Lobban and J. Moses, eds. The impact of ideas on legal development. Cambridge: Cambridge University Press. pp. 204–225. Bürge, A., (1995). Das französische Privatrecht im 19. Jahrhundert: Zwischen Tradition und Pandektenwissenschaft, Liberalismus und Etatismus. 2nd ed. Frankfurt am Main: Klostermann. Cardozo, B. N., (1924). The growth of the law. New Haven: Yale University Press. Christensen, K., (1993). Introduction. In: Gratian, The treatise on laws (Decretum DD. 1-20). Translated by J. Gordley. Washington, D.C.: The Catholic University of America Press. pp. ix–xxvii. Conring, H., (1665). De origine iuris germanici. Helmestadii: Müller. Cook, C. M., (1981). The American codification movement. a study of antebellum legal reform. Westport, C.T.: Greenwood Press. Crefeld, M. van, (1999). The rise and decline of the state. Cambridge: Cambridge University Press. Cross, R. and Harris, J. W., (1991). Precedent in English law. 4th ed. Oxford: Clarendon Press. Crystal, N. M., (1979). Codification and the Rise of the Restatement Movement. Washington Law Review. 54(2), 239–273. Donahue, C., (2008). Private law without the state and during its formation. In: N. Jansen and R. Michaels, eds. Beyond the state: rethinking private law. Tübingen: Mohr Siebeck. pp. 121–143. Duve, T., (2003). Mit der Autorität gegen die Autoritäten? Überlegungen zur heuristischen Kraft des Autoritätsbegriffs für die neuere Privatrechtsgeschichte. In: W. Oesterreicher, G. Regn and W. Schulze, eds. Autorität der Form – Autorisierung – institutionelle Autorität. Münster: Lit. pp. 239–256. Duxbury, N., (2008). The nature and authority of precedent. Cambridge: Cambridge University Press. Ehrlich, E., (1906). Internationales Privatrecht. Deutsche Rundschau. 126, 419–432. Eisenberg, M. A., (2002). The concept of national law and the rule of recognition. Florida State University Law Review. 29(4), 1229–1263. European Parliament and Council directive 2011/83/EU on consumer rights, amending Council Directive 93/13/EEC and Directive 1999/44/EC of the European Parliament and of the Council and repealing Council Directive 85/577/EEC and Directive 97/7/EC of the European Parliament and of the Council (2011) Official Journal L 304, p. 64. Falk, U. and Mohnhaupt, H., eds., (2000). Das bürgerliche Gesetzbuch und seine Richter. Zur Reaktion der Rechtsprechung auf die Kodifikation des deutschen Privatrechts (1896–1914). Frankfurt am Main: Klostermann. Fauvarque-Cosson, F. and Mazeaud, D., (2008). Principes contractuels communs – projet de cadre commun de référence. Association Henri Capitant des Amis de la Culture Juridique Française et Société de Législation Comparée. Paris: Société de Législation Comparée. Frank, J. P., (1998). The American Law Institute, 1923–1998. Hofstra Law Review. 26(3), 615–639. Grotius, H., (1625). De iure belli ac pacis libri tres: in quibus jus naturae et gentium, item juris publici praecipua explicantur. Parisiis: apud Nicolaum Buon. Haferkamp, H.-P., (2008). The science of private law and the state in nineteenth century Germany. In: N. Jansen and R. Michaels, eds. Beyond the state: Rethinking private law. Tübingen: Mohr Siebeck. pp. 245–267. Helmholz, R. H., (1996). The spirit of classical canon law. Athens, G.A.: University of Georgia Press. Hobbes, T., (1983). De cive: The Latin version. H. Warrender ed. Oxford: Clarendon Press. Hobbes, T., (2012). Leviathan. 3 vols. N. Malcolm ed. Oxford: Oxford University Press.

Soft law  41 Hudson, J., (2012). The Oxford history of the laws of England, vol. 2. Oxford: Oxford University Press. pp. 871–1216. Jansen, N., (2010). The making of legal authority: non-legislative codifications in historical and comparative perspective. Oxford: Oxford University Press. Jansen, N., (2011). Methoden, Institutionen, Texte. zur diskursiven Funktion und medialen Präsenz dogmatisierender Ordnungsvorstellungen und Deutungsmuster in Recht und Religion. Zeitschrift der Savigny-Stiftung für Rechtsgeschichte: Germanistische Abteilung. 128(1), 1–71. Jansen, N., (2012). Ius commune. In: J. Basedow, K. Hopt, R Zimmermann and A. Stier, eds. Max Planck encyclopedia of European private law. Oxford: Oxford University Press. pp. 1006–1010. Jansen, N., (2018). Rechtsdogmatik, Rechtswissenschaft und juristische Praxis. zum gesellschaftlichen Ort und wissenschaftlichen Selbstverständnis universitärer Rechtsdogmatik. Archiv des öffentlichen Rechts. 143(4), 623–658. Jansen, N., (2019). Recht und gesellschaftliche Differenzierung: Fünf Studien zur Genese des Rechts und seiner Wissenschaft. Tübingen: Mohr Siebeck. Jansen, N. and Michaels, R., (2008). Private law and the state: comparative perceptions and historical observations. In: N. Jansen and R. Michaels, eds. Beyond the state? Rethinking private law. Tübingen: Mohr Siebeck. pp. 15–67. Jansen, N. and Zimmermann, R., (2018). European contract laws: foundations, commentaries, synthesis. In: N. Jansen and R. Zimmermann, eds. Commentaries on European contract laws. Oxford: Oxford University Press. pp. 1–18. Kantorowicz, H., (1911). Rechtswissenschaft und Soziologie. Tübingen: Mohr. Kantorowicz, H., (1912). Was ist uns Savigny?. In: Recht und Wirtschaft. Berlin: Carl Heymanns. pp. 47–54, 76–79. Kantorowicz, H. (Gnaeus Flavius)., (2002). Der Kampf um die Rechtswissenschaft (1906). K. Muscheler, ed. Baden-Baden: Nomos. Kantorowicz, H. and Campbell, A. H., (1958). The definition of law. Cambridge: Cambridge University Press. Kantorowicz, H. U. and Patterson, E. W., (1928). Legal science – a summary of its methodology. Columbia Law Review. 28(6), 679–707. Lando, O., (2000). Preface. In: O. Lando and H. Beale, eds. Principles of European contract law. Parts I and II. The Hague: Kluwer Law International. pp. xi–xvi. Lando, O. and Beale, H., eds., (2000). Principles of European Contract Law. parts I and II. The Hague: Kluwer Law International. Lando, O., Clive, E., Prüm, A. and Zimmermann, R., eds., (2003). Principles of European contract law. part III. The Hague: Kluwer Law International. Luig, K., (1983). Conring, das deutsche Recht und die Rechtsgeschichte. In: M. Stolleis, ed. Hermann Conring (1606-1681): Beiträge zu Leben und Werk. Berlin: Duncker & Humblot. pp. 355–395. Lück, H., (2013). Über den Sachsenspiegel: Entstehung, Inhalt und Wirkung des Rechtsbuches. 3rd ed. Wettin-Löbejün: Stekovics. Maggs, G. E., (1998). Ipse dixit: the restatements (second) of contracts and the modern development of contract law. George Washington Law Review. 66(3), 508–555. Mehren, A. T. von, (1988). Law in the United States: a general and comparative view. Deventer: Kluwer. Michaels, R., (2009). Umdenken für die UNIDROIT-Prinzipien: Vom Rechtswahlstatut zum allgemeinen Teil des transnationalen Vertragsrechts. Rabels Zeitschrift für ausländisches und internationales Privatrecht. 73(4), 866–888. Michaels, R., (2015). Preamble I: purposes, legal nature, and scope of the PICC; applicability by courts; use of the PICC for the purpose of interpretation and supplementation and as a model. In: S. Vogenauer, ed. Commentary on the UNIDROIT principles of international commercial contracts (PICC). 2nd ed. Oxford, UK: Oxford University Press. pp. 31–109. Odofredus., (1552). In primam codicis partem … praelectiones (quae lecturae appellantur). Lyon: Merchant. Reprint, Bologna: Forni, 1968. Oestmann, P., (2009). Ein Zivilprozeß am Reichskammergericht: Edition einer Gerichtsakte aus dem 18. Jahrhundert. Cologne: Böhlau. Payandeh, M., (2017). Judikative Rechtserzeugung: Theorie, Dogmatik und Methodik der Wirkungen von Präjudizien. Tübingen: Mohr Siebeck.

42  Research handbook on soft law Pufendorf, S. von, (1759). De iure naturae et gentium libri octo. cum integris commentariis … Io. Nic. Hertii, atque Io. Barbeyraci, accedit Eris Scandica. Frankfurt am Main and Leipzig: Ex officina Knochio-Eslingeriana. Radding, C. M., (1988). The origins of medieval jurisprudence. New Haven: Yale University Press. Radding, C. M. and Ciaralli, A., (2007). The Corpus iuris civilis in the Middle Ages. Manuscripts and transmission from the sixth century to the juristic revival. Leiden: Brill. Rehberg, K.-S., (1994). Institutionen als symbolische Ordnungen: Leitfragen und Grundkategorien zur Theorie und Analyse institutioneller Mechanismen. In: G. Göhler, ed. Die Eigenart der Institutionen: zum Profil politischer Institutionentheorie. Baden-Baden: Nomos. pp. 47–84. Reinhard, W., (2002). Geschichte der Staatsgewalt: Eine vergleichende Verfassungsgeschichte Europas von den Anfängen bis zur Gegenwart. 3rd ed. Munich: Beck. Research Group on the Existing EC Private Law (Acquis Group)., ed., (2009). Principles of the existing EC contract law (acquis principles): contract II: General provisions, delivery of goods, package travel and payment services. Munich: Sellier. Savigny, F. C. von, (1850). Geschichte des römischen Rechts im Mittelalter, vol. 4. 2nd ed. Heidelberg: Mohr. Schauer, F., (2008). Authority and authorities. Virginia Law Review. 94(8), 1931–1961. Scherer, M., (2015). Preamble II: The use of the PICC in arbitration. In: S. Vogenauer, ed. Commentary on the UNIDROIT Principles of International Commercial Contracts (PICC). 2nd ed. Oxford, UK: Oxford University Press. pp. 110–149. Shaffer, G. C. and Pollack, M. A., (2010). Hard vs. soft law: Alternatives, complements, and antagonists in international governance. Minnesota Law Review. 94(3), 706–799. Stintzing, R., (1884). Geschichte der deutschen Rechtswissenschaft, vol. 2. Munich and Leipzig: Oldenbourg. Stryk, S., (1713). Discursus praeliminaris de usu et auctoritate juris Romani in foris Germaniae. In: S. Stryk, Specimen usus moderni pandectarum …, vol. 1. 4th ed. Halle: Magdeburgicae Waisenhaus. pp. 1–38. Tuck, R., (1993). Philosophy and government: 1572–1651. Cambridge: Cambridge University Press. Tushnet, M. V., (1983). Following the rules laid down: A critique of interpretivism and neutral principles. Harvard Law Review. 96(4), 781–827. UNIDROIT., (1994). Report of the 73rd session on the Governing Council. UNIDROIT proceedings and papers, C.D. (73) 18. Rome: International Institute for the Unification of Private Law (UNIDROIT). UNIDROIT., (2004). Introduction to the 1994 edition. In: UNIDROIT Principles of International Commercial Contracts. 2nd ed. Rome: International Institute for the Unification of Private Law (UNIDROIT). pp. xiv–xvi. UNIDROIT., (2016). UNIDROIT Principles of International Commercial Contracts. 4th ed. Rome: International Institute for the Unification of Private Law (UNIDROIT). Vogenauer, S., (2006). Zur Geschichte des Präjudizienrechts in England. Zeitschrift für neuere Rechtsgeschichte. 28(1), 48–78. Vogenauer, S., (2015). Introduction. In: S. Vogenauer, ed. Commentary on the UNIDROIT Principles of International Commercial Contracts (PICC). 2nd ed. Oxford, UK: Oxford University Press. pp. 1–30. Volkmann, U., (2018). Verfassungsänderung und Verfassungswandel. Beobachtungen zum Verhältnis von Stabilität und Dynamik im Verfassungsrecht der Bundesrepublik Deutschland. Juristenzeitung. 73(6), 265–271. Weiss, G. A., (2000). The enchantment of codification in the common-law world. Yale Journal of International Law. 25(2), 435–532. Zekoll, J., (2008). Das American Law Institute – ein Vorbild für Europa?. In: R. Zimmermann, U. Blaurock, C. Kirchner and U. Spellenberg, eds. Globalisierung und entstaatlichung des rechts, vol. 2: Nichtstaatliches Privatrecht: Geltung und Genese. Tübingen: Mohr Siebeck. pp. 101–127. Zimmermann, R., (1995). Codification: history and present significance of an idea – à propos the recodification of private law in the Czech Republic. European Review of Private Law. 3(1), 95–120. Zimmermann, R., (2001). Roman law, contemporary law, European law. The civilian tradition today. Oxford: Oxford University Press. Zimmermann, R., (2012). Codification: The civilian experience reconsidered on the eve of a Common European Sales Law. European Review of Contract Law. 8(4), 367–399.

3. The definition of soft law Fabien Terpan

1. INTRODUCTION Soft law is a doctrinal creation, which has no real ground in positive law. Article 38(1) of the Statute of the International Court of Justice makes no reference to soft law as a possible source of international law. Article 288 of the Treaty on the Functioning of the European Union (EU), while mentioning two types of acts deprived of legal force, recommendations and opinions, does not clearly establish soft law as a specific type of secondary law, nor do state constitutions recognize the existence of soft law. However, the concept has been used for a long time by both academics and practitioners, as it helps to encapsulate the complexity of normative action beyond a binary division between law and non-law. It is not surprising to find that soft law has first emerged in the context of international relations. Indeed, since the beginning of the 20th century, states have made several soft international commitments such as gentlemen’s agreements and memorandums of understanding. The paternity of the idea is often attributed to Lord McNair, even though he did not use explicitly the expression ‘soft law’ (McNair 1930). Before becoming a British judge at the International Court of Justice in 1937, and the first president of the European Court of Human Rights in 1959, Lord McNair, as a professor of international law, observed the emergence of international law and concluded that the Treaties are different in terms of their legal character and functions. Some of these Treaties can be seen as instruments with extra-legal binding effect. During the 1970s and 1980s, the concept of soft law further developed as part of a debate on the multifaceted development of international law (Dupuy 1977; Schachter 1977; Seidl-Hohenveldern 1979, pp. 173–177; Baxter 1980), and since then publications on the subject have mushroomed (Hillgenberg 1999; Abbott and Snidal 2000; Guzman and Meyer 2010). Studies on the EU took on board the soft law concept from the international relations / law literature, to catalogue those instruments that have an uncertain legal status (Wellens and Borchardt 1989; Snyder 1993; Senden 2004). The question whether European soft law should be given the same meaning than international soft law was raised (Beveridge and Nott 1998). Recently the emergence of the so-called ‘new modes of governance’ have increased the interest in aspects of soft law (Peters 2011). Also employed at domestic level (Knauff 2010; Gerry-Vernières 2012; Lavergne 2018), the concept is not confined to a particular territorial framework. Nor is it limited to legal scholarship: all disciplines in social sciences make use of the term, sometimes without caring about its precise significance. Soft law is based on the basic idea that some types of acts and norms cannot be seen as proper law although they are not merely politics either. Thus, having a clear view of what soft law is made of, where it starts and where it ends, does matter. This line of thinking is opposed by those who contest the existence of soft law, arguing that the latter is ‘a self-serving quest for new legal materials’ (d’Aspremont 2008), i.e., an opportunity for legal scholars to enlarge the scope of their research. From a legal positivist perspective, law is not a matter of degree, 43

44  Research handbook on soft law and therefore there is either law or non-law (Weil 1982; Blutman 2010). Other scholars, while acknowledging that the normative force of law may vary, rather use the notion of informality (Klabbers 1994; Pauwelyn et al. 2012). Those who use the notion of soft law and reject the binary approach do not necessarily see soft law as a ‘positive’ development in terms of legitimacy, transparency, legal certainty, and respect of the rule of law. They may agree with positivists that the development of a normative category made of soft law is, to some extent, a threat to the rule of law (Klabbers 1998; Duplessis 2007; Eliantonio and Ştefan 2021; Petropoulou Ionescu and Eliantonio 2021; Wessel 2021). Yet, they acknowledge the existence of soft law and pragmatically seek to better understand what it is and how it functions. Defining soft law is thus a difficult task as no real consensus has emerged in the scholarship, neither on the existence of soft law, nor on the criteria that should help to delimit its contours. Following from this, this chapter will account for the diversity of definitions and will discuss the criteria and characteristics that have been used to describe and delineate soft law. The co-existence of several definitions can be seen a source of misunderstanding and an obstacle to a proper analysis of normative action, especially when it comes to comparing soft law in different political and legal systems. Therefore, in the absence of a commonly agreed understanding of soft law, we will try to identify a number of key features that are common to most definitions. These key features relate to two different dimensions. First, soft law is defined in opposition to hard law and non-law. Once soft law has been distinguished from hard law and non-law, it is then possible to add complementary information on the characteristics of soft law by looking at the way it interacts with hard law.

2.

SITUATING SOFT LAW IN BETWEEN HARD LAW AND NON-LAW

Most definitions put the emphasis on the fact that soft law resembles hard law but falls short of it. It can be seen as a ‘lesser’ form of law (Ştefan et al. 2019), a half-way solution, when you cannot or do not want to regulate through hard law, but nevertheless seek to provide some guidance to a number of targeted actors. As Michelle Cini put it in the context of the EU, ‘soft law lies somewhere between general policy statements (and Commission discretion), on the one hand, and legislation, on the other’ (Cini 2001, p. 194). 2.1

Soft Law is Distinct from Hard Law

Soft law is frequently defined as opposed to hard law. In the EU, it is distinguished from both the EU treaties and secondary law adopted on the basis of Article 288 TFEU (regulations, directives, and decisions), and it is often defined in a minimal way by listing several non-legally-binding instruments such as recommendations, codes of conduct, and resolutions, and by pointing out the uncertainty surrounding their legal effects (Ştefan et al. 2019, p. 20). More generally, some elements are missing for soft law to be considered proper (hard) law. Contrary to hard law, soft law will then be seen as ‘weak and fragile’ (Baxter 1980, p. 549), ‘toothless’ (Dupuy 1977), ‘muffled’ (with the words of François Rigaux, see IIL 1984, p. 147) or ‘vague’ (with the words of Pierre Pescatore, see IIL 1984, p.146). Depending on the way

The definition of soft law  45 these weaknesses are seen, two definitions of soft law are possible, a large definition and more restrictive one. The larger definitions of soft law are based on a multidimensional approach where the weaknesses of soft law can be found at different levels. This is based on the idea that normativity depends not only on the normative value of law (whether it is or is not legally binding) but also on other dimensions such as the existence of normative safeguards and the actual scope of the norms (Thibierge 2009). Abbott and Snidal (2000) focused on the concept of ‘legalization’ in order to distinguish between hard and soft law in international relations. Legalization is characterized by three components: obligation, precision, and delegation. Obligation implies that the international actors are legally bound by a certain provision; precision requires unambiguity with respect to the conduct to follow; and delegation means that the implementation, interpretation, application of the rules, and conflict resolution were entrusted to third parties. ‘Soft law’ occurs when one or more of the three dimensions are weakened. Thibierge (2003) offered a similar view of ‘droit souple’ in international relations, while considering the intersections between three facets: content, legally binding force, and enforcement. Law can be defined as soft due to: (1) a lack of precision of its content (in French: droit flou, also called by other authors ‘soft negotium’ or ‘legal soft law’ Chinkin 1989; see also Delmas-Marty 1986); (2) a lack of legally binding force (‘droit doux’); and (3) a lack of enforcement (‘droit mou’). Apart from the sub-division of the concept into three different terms (droit flou, droit doux, droit mou), the added value of this definition, comparing it to Abbott and Snidal’s (2000), is the emphasis put on enforcement instead of delegation. Drawing on Abbott and Snidal (2000) and Thibierge (2003), we will call these three types of soft law: imprecise law, law without obligation, and unenforceable law. Far from being limited to the international law doctrine, multidimensional approaches were later discussed in the context of EU law (Terpan 2015) and domestic law (Lavergne 2018). Terpan (2015) argued that a lack of precision tends to reduce the degree of the obligation, and thus precision is a matter of more or less obligation, which leaves us with different combinations of soft / hard obligations and soft / hard enforcement. Two main critiques can be addressed to these large definitions of soft law. A first critique points at the complexity of the definitions due to multiple combinations of these criteria. Imprecise law, law without obligation, and unenforceable law bear very different types of weaknesses, which makes comparison between them rather difficult. In particular, these definitions do not help to situate soft norms on a normative continuum. Is an imprecise, unenforced but obligatory rule softer or harder than a precise, non-obligatory rule accompanied with some form of enforcement mechanism? However, the main purpose of multidimensional definitions is not to classify norms on a scale of normativity, from the softer norm to the harder one. It is rather to point at the diversity of law and provide the analytical tools that are needed to better account for this diversity. The second argument against large / multidimensional definitions is that they include binding norms in the definition of soft law. Indeed, based on the abovementioned definitions, hard instruments such as international agreements can be considered soft law if there is no enforcement mechanism attached to it (unenforceable law). Similarly, legally binding acts

46  Research handbook on soft law sometimes contain norms that are drafted in conditional terms instead of clearly prohibiting or allowing a particular kind of conduct, which makes them soft rather than hard norms (imprecise law). In the European context, imprecise norms, allowing exemption or derogation from specific standards would be seen as soft law even if they are part of a ‘hard’ EU directive. This is the way Falkner et al. deal with the notion of soft law in their book Compliance with soft law (Falkner et al. 2005, pp. 49–50; see also Duttle et al. 2017). If all norms that are not precise enough and/or not enforceable are considered soft law, then this category would have a very extensive scope, and the distinction between legal commitments and non-legal commitments would be blurred. This is why most scholars choose to differentiate soft and hard law based on one criterion only (the lack of legal obligation) leaving aside the two other dimensions (lack of precision and lack of delegation/enforcement). Restrictive definitions, supported by a large majority of authors, limit the scope of soft law to law without obligation, considering that imprecise law and unenforceable law are in fact hard law. The most cited definitions of soft law insist on the lack of legally binding force as a central tenet of soft law. For Snyder (1993, p. 56), soft law is made of ‘rules of conduct which, in principle, have no legally binding force but which nevertheless may have practical effects’. Similarly, Senden sees soft law as ‘rules of conduct that are laid down in instruments which have not been attributed legally binding force as such, but nevertheless may have certain (indirect) legal effects, and that are aimed at and may produce practical effects’ (Senden 2004, p. 3). Compared to multiple criteria definitions, this has the disadvantage of hiding the weaknesses of some types of hard norms compared to others. The power of description is lower as no emphasis is put on softness deriving from the lack of precision or the lack of enforcement. As soon as a legal obligation is found, this is hard law. But these restrictive definitions are more simple, easier to use, and they allow to clearly distinguish binding norms from non-binding ones. Thus, if the lack of legal binding force is a central aspect of soft norms and acts, a more precise definition of the notion can only be found by differentiating soft law from non-law. 2.2

Soft law is Distinct from Non-law

While soft law is first and foremost defined in opposition to hard law, resemblance with hard law is what makes soft law differ from non-law. Two elements are usually emphasized in the literature: (1) formalization: soft law is framed in a way that makes it close to a hard instrument; (2) effect: soft law aims at influencing, modifying, and directing the behavior of their addressees. Formalization, which appears rather implicitly in a number of definitions, helps to draw a line between soft law and non-law. Due to a lack of formalization, non-binding instruments resorting to ethics, religion, or politics but in no way to the regulation of society through public law are not included in the category of soft law. When a member of the government or any other official publicly argues in favour of gender equality, this is not soft law but merely politics. When standards, targets, indicators aiming at gender equality are included in a document named guidelines or recommendations, then it can be seen as soft law. Snyder (1993) and Senden (2004) both use the expression ‘rules of conduct’ to show that a certain degree of formalization is indeed necessary. However, there is no real attempt in the scholarship to define with precision the degree of formalization that an instrument should attain to enter in the category of soft law (for an exception, see the EfSoLaw project: www​

The definition of soft law  47 .efsolaw​.eu [viewed 9 June 2023]). We know that soft law is a law-like type of rule. The language of soft law can even be so close to hard law that it can be transformed into hard law after only minimal changes. However, there is no consensus on the type and degree of formalization that is needed to clearly distinguish soft law from non-law. Among other things, references are made to the use of statutory language, implementation deadlines, targets, and indicators (Atkinson et al. 2004), and more generally a structuring in articles or any other type of numeration. Instruments such as recommendations, codes of conducts, memorandums of understanding, and guidelines are usually divided in articles including prescriptive indications. It is sometimes noted that soft law is elaborated through a process that is quite similar to the making of hard law. The second element helping to separate soft law from non-law is the fact that soft law aims at producing effects. Soft law goes beyond preparatory acts and advisory opinions; just like hard law, it must have per se the capacity to influence the behaviour of its addressees (Conseil d’Etat 2013, p. 61). These effects remain largely uncertain (Chinkin 1989). As explained by Kenner in the field of environment, ‘while hard laws … create rules that Member States are bound to comply with, soft laws … are essentially methods of Community guidance or rules which create an expectation that the conduct of Member States will be in conformity with them, but without an accompanying legal obligation’ (Kenner 2003, pp. 127–128). To put it differently, compliance with hard law is more likely (although not certain) than compliance with soft law. But soft law aims at having effects and it creates expectations that it will be the case, which contributes to distinguish it from non-law. The effects of soft law have been dealt with by a large literature assessing the effectiveness of soft law in different fields (Chinkin 2000) such as multilateral arms control (Williamson 2003), international environmental norms (Skjærseth et al. 2006), the Basle accord (Ho 2002), employment policy (Ashiagbor 2004; Jacobsson 2004), environment (Korkea-aho et al. 2020), and EU competition law (Cosma and Whish 2003; Blauberger 2009; Berrod 2015). Two types of effects are expected – practical and legal effects – although the distinction between the two is not so easy to establish. Practical effects refer to transformation in the behaviour and practices of the actors targeted by a soft instrument. Soft law leads to policy change such as building a normative framework for future legislation, leading actors to cooperate, interpreting hard law provisions, including (‘subtler changes at the level of discourse, understanding and policy principles’, Ştefan et al. 2019, p. 22). For instance, EU soft law may produce legal effect when national administrations make ‘use’ of them at Member State level (Bérut 2021; Eliantonio et al. 2021; Hartlapp and Hofmann 2021). Practical effects may have a strong sociological character when referring to ‘common discourse; the symbols of a common project; strengthening socialization through repeated meetings; mobilizing the actors and their partnership through networks; and iterative processes’ (Ştefan et al. 2019, p. 22). These effects usually occur when actors feel bound by a soft rule although they are not legally committed. From a constructivist point of view, it can be argued that what matters most is the sense of obligation, i.e., the fact that actors feel obliged to comply (Finnemore and Toope 2001). Legal effects occur when the addressees of a soft law instrument decide to implement it through the enactment of legislation or any other type of legal act. While the adoption of such an act is not an obligation, it is seen as desirable by the addressee of the soft law act. For example, the Declaration on the Rights of Indigenous Peoples, adopted in 2007 by the General Assembly of the United Nations, may have had legal effects not only because it was a first step

48  Research handbook on soft law towards the adoption of legal agreements but also because it was seen as ‘highly legitimate and authoritative’ (Barelli 2009, p. 972). Generally speaking, legal effects of soft law may also result from the existence of (soft) enforcement mechanisms such as evaluations and review processes (Terpan 2015; Ştefan 2017). The question arises as to whether soft law can be given legal value in a court of law, and whether courts can take soft law instruments into consideration when deciding cases (Guinard 2013; Ştefan 2013; Korkea-aho 2015; Eliantonio 2018; Eliantonio and Ştefan 2018). The fact that soft law does not include legal commitment does not mean that it cannot have effects in a court of law. Three different situations must be distinguished. First, soft law may serve as an auxiliary means to interpret hard law provisions. This can be reflected in the case-law of the Court of Justice of the European Union (CJEU), with several rulings where the Court did not use soft law solely but in addition to a hard act, as a complementary means to provide an accurate interpretation of EU law. The CJEU in Grimaldi (C-322/88) also urged national courts to ‘take into consideration’ soft law whenever deciding on cases. However, since then the CJEU has referred to the Grimaldi doctrine only sparingly (Korkea-aho 2015, 2018), and national courts rarely (but possibly increasingly) use European soft law (Eliantonio et al. 2021). Secondly, the consideration of soft law by implementing actors can be seen in a court of law as an obligation under constitutional provisions. For example, the principle of sincere cooperation deriving from Article 4(3) TEU has been presented by the CJEU as a duty to take soft law into consideration, if not considered as an obligation to comply with soft law (C-525/14 Commission v Czech Republic, para. 38). A third type of situation is when soft law provides the basis for judicial review or is annulled by a court. In those cases, soft law is hardened by courts, which can be seen from two different points of view: on the one hand, it is a case of soft law having produced direct legal effects, which blurs the line between binding and non-binding norms and to some extent endangers the law (Bouveresse 2015); on the other hand, it can be seen as a transformation of soft law into hard law, and in that case, the line between soft and hard law is not blurred but rather moved by the court, which decides to change the legal qualification of an act, from soft to hard law. Coming back to the criteria defining soft law, we have seen that soft law is more than non-law thanks to its formalization and to the effects – both practical and legal – it produces. But it is less than hard law because – at least if we stick to a restrictive approach – it does not contain any legal commitment. Additional characteristics can be found by looking at the way soft law interacts with hard law.

3.

DEFINING SOFT LAW THROUGH ITS INTERACTIONS WITH HARD LAW

Another way to define soft law is to look at the function it performs, which is usually done through an analysis of their interactions with hard law (Conseil d’Etat 2013). Soft law may anticipate hard law (pre-law) or follow the adoption of hard law (post-law); it may also be considered as a substitute to hard law (para-law) (Senden 2004).

The definition of soft law  49 3.1 Pre-law Pre-law usually includes preparatory instruments that put forward various proposals for future action. All preparatory acts cannot be considered soft law: White and Green Papers in the EU as well as parliamentary preparatory acts or advisory opinions would not be included, if we follow the idea that soft instruments must contain rules of conduct aiming at producing effects. On the contrary, EU strategies can be seen as pre-law as they foresee the adoption of future legislation. At domestic level, public policies are more and more based on a roadmap including broad objectives, indicators, and the financial means to achieve these objectives. Soft law defined as pre-law must have a programming dimension. International lawyers such as Dupuy and Virally used to present soft law as a form of law in the making, a process of transformation that was reflected in expressions such as ‘droit vert’ (Dupuy 1977, p. 140) and ‘pre-droit’ (with the words of Virally, see IIL 1983, pp. 166–327). This way of defining pre-law is not without ambiguities. It tends to see all forms of soft law as ‘imperfect’ and designed to evolve into hard law. This might be the case in some policy areas where soft law has often been the initial foundation for legislating in the future (Goldstein et al. 2001; Blichner and Molander 2008; and in the EU: Aldson 2014; Saurugger and Terpan 2021; Terpan and Saurugger 2021). Several resolutions of the UN General Assembly have been transformed into conventions (Universal Declaration of the Rights of the Child adopted in 1959 and transformed into a convention in 1989; Standard Rules on the Equalization of Opportunities for Persons with Disabilities adopted in 1993, transformed into a convention in 2006). However, one cannot be sure that eventually soft law will be transformed into hard law. While it is easier to identify pre-law when its transformation into hard law has occurred, it is more difficult when it has not. 3.2 Post-law The category of post-law is made of these instruments that are posterior to the adoption of a hard instrument. The latter may have foreseen the adoption of a soft act. This is the case when legislative acts delegate to agencies the task of adopting recommendations. Another example is given by Article 34 of the French Constitution which allows for the adoption of programming laws as well as resolutions of the parliamentary assemblies. Yet, soft instruments may also be related to hard law without any reference made to the adoption of a soft instrument in any piece of legislation. Generally speaking, post-law may have different functions such as bringing support to the implementation and proper application of hard law, making the implications of hard law more explicit or rendering the application of hard law by administrations more predictable. In EU studies, soft law in the form of post-law has been divided into interpretative and decisional instruments. Interpretative instruments ‘restate or summarize the interpretation that should be given to Community law provisions’. Decisional instruments indicate ‘the way in which a Community institution will apply Community law provisions in an individual case where it has implementing and discretionary powers’ (Senden 2004, p. 140). Beyond the case of the EU, soft law is also frequently used in international relations to complement formal agreements: guiding principles of the High Commissioner for Refugees, comments on the OECD model tax conventions, resolutions of the Parliamentary Assembly

50  Research handbook on soft law and the Committee of Ministers of the Council of Europe precising the scope of the European Convention on Human Rights. 3.3

Soft Law as a Substitute for Hard Law

Soft law can also be meant to be a sustainable substitute for hard law. In this case, soft law is purposively adopted in the place of hard law because the use of hard law is either not desired or not possible in a specific area. Admittedly, the adoption of hard law might always be possible in the future but for the time being soft law is self-sufficient and there is no immediate plan to make use of hard law. Soft law as a substitute for hard law mainly refers to instruments that aim at steering or guiding action in a non-legally binding way, by laying down ‘new rules independently of an existing legal framework’ (Senden 2004, p. 119). This type of soft law is particularly useful at the international level as it is often difficult, and sometimes even impossible, to make legal commitments. Instances of soft law as an alternate to soft law are numerous. In 1975, the Final Act of the Conference on Security and Cooperation in Europe (CSCE) was adopted as a soft law document due to the impossibility for the 35 signatories to adopt a legal agreement in the context of the Cold War (CSCE 1975). Similarly, the International Labour Organization in the 1990s has resorted to recommendations because proper conventions had become more and more difficult to adopt (Blanpain and Colucci 2004). The Paris Club, an informal group of official creditors whose role is to find coordinated and sustainable solutions to the payment difficulties experienced by debtor countries, has reached around 500 soft agreements since its inception in 1956 (Conseil d’Etat 2013, p. 29; www​.clubdeparis​.org [viewed 9 June 2023]). Those soft ‘agreements’ only produce legal effects when implemented through hard law bilateral agreements between the debtor and its creditors. At domestic level, public authorities can more easily adopt rules in the form of hard law. This does not mean that soft law is of no use at all; but it is more likely that soft law be pre-law or post-law. Examples of soft law as a substitute to hard law can be found with regards to the role of agencies dealing with specific fields such as public health, human rights, or internet regulation. In the EU, new modes of governance such as the Open Method of Coordination usually include forms of soft law as ‘an alternative to hard regulation as well as to no regulation at all’ (Streeck 1995 on EU social policy). This governance turn in the 1990s is said to have tremendously developed the use of soft law in areas where the adoption of regulation and directives proved to be difficult (Peters 2011; Korkea-aho 2015). These soft acts, adopted as a substitute to hard law, nevertheless interact with hard law in different ways. Sometimes soft law in a specific policy area can be linked with stricter obligations in another area (soft steering in social policy linked with stricter economic governance under the European Semester, Bekker 2014). What is more, soft and hard law often coexist in a given policy or in a particular instrument, giving birth to the notion of hybridity (Trubek and Trubek 2005; Trubek et al. 2006).

The definition of soft law  51

4. CONCLUSION Soft law can first be defined in relation with both hard law and non-law. It has to be differentiated from hard law in order to be considered ‘law’ at all. A minority view holds a multidimensional definition where law is considered soft when it lacks one of the three following elements: precision of its content (imprecise law), legal obligation (law without obligation) and enforcement (unenforceable law). But most scholars only keep one of the three dimensions – the lack of obligation – as a necessary criterion of soft law. Of course, defining soft law through the criterion of non-bindingness is not enough. Soft law also needs to be distinguished from non-law, which is usually done by looking at two elements: formalization and effect. A soft instrument is usually framed in a way that makes it close to a hard instrument. A certain degree of formalization is necessary, although it is hard to define this element in a very precise manner. In addition, soft law aims at influencing, modifying, and directing the behaviour of their addressees. It is meant to produce legal or practical effects, just as a hard instrument would do. Another way to make the definition of soft law more accurate is to look at the way it interacts with hard law. This allows to distinguish between pre-law, post-law and soft law as a substitute to hard law. Although useful, this approach of soft law is not without its complexity as both types of law intertwine in many different ways, with soft law being adopted in order to implement hard law; soft and hard law being included in hybrid instruments; soft law being hardened in a court of law. Should soft law be more precisely defined? We argue that there is, indeed, room for improvement, and that a more detailed definition could help to better analyze and understand the use of soft law at domestic, EU, and international level. In particular, academic research could benefit from a better knowledge of the characteristics and degrees of formalization, as well as of the types of interaction between soft and hard law. This would facilitate comparative analysis of different cases. Indeed, this is precisely the lack of consensus on the definition of soft law that makes comparative analyzes so hard. The fact that soft law has been used in very different context, by various organizations and to solve different problems, has not helped to build a consensual definition of soft law. On the contrary, the notion, once limited to a rather specific debate over international law, has developed as a catch-all concept, flexible enough to adapt to any situation, whether domestic or international. While soft law diversifies, it is more necessary than ever to refine the categorization of soft law according to criteria such as: the author of the soft instrument, its addressee, the problem it is supposed to solve, the territorial level of its application. A future research agenda should address these issues in order to see whether and to what extent they impact on the definition of soft law. However, improving our approaches to soft law should not lead to further disagreement on the definition of soft law but should rather provide greater clarity. This is why future developments of the notion should be built on existing typologies, and should be situated with regard to the criteria presented above (distinction from both hard law and non-law; type of relation with hard law).

52  Research handbook on soft law

REFERENCES Abbott, K. W. and Snidal, D., (2000). Hard and soft law in international governance. International Organization. 54(3), 421–456. Aldson, F., (2014). The EU’s recommendation on hydraulic fracturing and its implementation in the UK. Environmental Liability. 67(2), 67–69. Ashiagbor, D., (2004). Soft harmonisation: the Open Method of Coordination in the European employment strategy. European Public Law. 10(2), 305–332. Atkinson, T., Marlier, E., and Nolan, B., (2004). Indicators and targets for social inclusion in the European Union. Journal of Common Market Studies. 42(1). 47–75. Barelli, M., (2009). The role of soft law in the international legal system: the case of the United Nations Declaration on the Rights of Indigenous Peoples. International & Comparative Law Quarterly. 58(4), 957–983. Baxter, R. R., (1980). International law in ‘her infinite variety’. International & Comparative Law Quarterly. 29(4), 549–566. Bekker, S., (2014). EU coordination of welfare states after the crisis: further interconnecting soft and hard law. International Review of Public Administration. 19(3), 296–307. Berrod, F., (2015). L’utilisation de la ‘soft law’ comme méthode de conception du droit européen de la concurrence. Revue de l’Union Européenne. 588, 283–290. Bérut, C., (2021). The European Union as an opportunity: structures and uses of European soft law in French, Austrian and Irish eHealth policies. West European Politics. 44(1), 155–175. Beveridge, F. and Nott, S., (1998). A hard look at soft law. In: P. Craig and C. Harlow, eds. Lawmaking in the European Union. The Hague: Kluwer Law International. pp. 285–309. Blanpain, R., and Colucci, M., (2004). The globalization of labour standards: the soft law track. The Hague: Kluwer Law International. Blauberger, M., (2009). Of ‘good’ and ‘bad’ subsidies: European state aid control through soft and hard law. West European Politics. 32(4), 719–737. Blichner, L. C. and Molander, A., (2008). Mapping juridification. European Law Journal. 14(1), 36–54. Blutman, L., (2010). In the trap of a legal metaphor: international soft law. International & Comparative Law Quarterly. 59(3), 605–624. Bouveresse, A., (2015). La portée normative de la ‘soft law’. Revue de l’Union Européenne 588, 291–298. Case C-322/88 Grimaldi v Fonds des maladies professionnelles ECLI:EU:C:1989:646. Case C-525/14 Commission v Czech Republic ECLI:EU:C:2016:714. Chinkin, C. M., (1989). The challenge of soft law: development and change in international law. International & Comparative Law Quarterly. 38(4), 850–866. Chinkin, C., (2000). Normative development in the international legal system. In: D. Shelton, ed. Commitment and compliance: the role of non-binding norms in the international legal system. Oxford: Oxford University Press. pp. 21–42. Cini, M., (2001). The soft law approach: Commission rule-making in the EU’s state aid regime. Journal of European Public Policy. 8(2), 192–207. Commission on the Security and Cooperation in Europe (CSCE)., (1975). The Helsinki Final Act [online]. Helsinki: Commission on the Security and Cooperation in Europe. [Viewed 22 December 2021]. Available from: https://​www​.csce​.gov/​sites/​helsinkicommission​.house​.gov/​files/​Helsinki​ %20Final​%20Act​.pdf [viewed 9 June 2023]. Conseil d’Etat., (2013). Etude annuelle 2013: Le droit souple [online]. Paris: Conseil d’Etat. [Viewed 18 January 2022]. Available from: https://​www​.conseil​-etat​.fr/​ressources/​etudes​-publications/​rapports​ -etudes/​etudes​-annuelles/​etude​-annuelle​-2013​-le​-droit​-souple [viewed 9 June 2023]. Cosma, H. A. and Whish, R., (2003). Soft law in the field of EU competition policy. European Business Law Review. 14(1), 25–56. d’Aspremont, J., (2008). Softness in international law: a self-serving quest for new legal materials. European Journal of International Law. 19(5), 1075–1093. Delmas-Marty, M., (1986). Le flou du droit: du code penal aux droits de l’homme. Paris: Presses Universitaires de France.

The definition of soft law  53 Duplessis, I., (2007). Les vertiges de la soft law: Réactions internationales en droit international. Revue québécoise de droit international (Hors série). 245–268. Dupuy, R. J., (1977). Declaratory law and programmatory law: from revolutionary custom to ‘soft law’. In: R. Akkerman, P. J. van Krieken and C. O. Pannenborg, eds. Declarations of principles. a quest for universal peace. Leyden: Sijthoff. pp. 247–257. Duttle, T., Holzinger, K., Malang, T., Schäubli, T. and Schimmelfennig, F., (2017). Opting out from European Union legislation: the differentiation of secondary law. Journal of European Public Policy. 24(3), 406–428. Eliantonio, M., (2018). Soft law in environmental matters and the role of the European courts: too much or too little of it?. Yearbook of European Law. 37, 496–524. Eliantonio, M. and Ştefan, O., (2018). Soft law before the European courts: discovering a ‘common pattern’?. Yearbook of European law. 37, 457–469. Eliantonio, M. and Ştefan, O., (2021). The elusive legitimacy of EU soft law: an analysis of consultation and participation in the process of adopting COVID-19 soft law in the EU. European Journal of Risk Regulation. 12(1), 159–175. Eliantonio, M., Korkea-aho, E. and Ştefan, O., eds., (2021). EU soft law in the member states: theoretical findings and empirical evidence. Oxford, UK: Hart Publishing. Falkner, G., Treib, O., Hartlapp, M. and Leiber, S., (2005). Complying with Europe? The impact of EU minimum harmonisation and soft law in the member states. New York: Cambridge University Press. Finnemore, M. and Toope, S. J., (2001). Alternatives to ‘Legalization’: richer views of law and politics. International Organization. 55(3), 743–758. Gerry-Vernières, S., (2012). Les ‘petites’ sources du droit: à propos des sources étatiques non contraignantes. Paris: Economica. Goldstein, J. L., Kahler, M., Keohane, R. O. and Slaughter, A.-M., eds., (2001). Legalization and world politics. Cambridge, M.A.: MIT Press. Guinard, D., (2013). A propos du soft law en droit de l’Union européenne: analyses théorique et pratique. Revue de l’Union Européenne. 573, 609–617. Guzman, T. and Meyer, T. L., (2010). International soft law. Journal of Legal Analysis. 2(1), 171–225. Hartlapp, M. and Hofmann, A., (2021). The use of EU soft law by national courts and bureaucrats: how relation to hard law and policy maturity matter. West European Politics. 44(1), 134–154. Hillgenberg, H., (1999). A fresh look at soft law. European Journal of International Law. 10(3), 499–515. Ho, D. E., (2002). Compliance and international soft law: why do countries implement the Basle Accord?. Journal of International Economic Law. 5(3), 647–688. Institute of International Law (IIL)., (1983). Annuaire, session de Cambridge 1983: travaux préparatoires, vol. 60-I [online]. Paris: Institut de Droit International. [Viewed 19 January 2022]. Available from: https://​www​.idi​-iil​.org/​app/​uploads/​2018/​06/​1983​_vol​_60​-I​_Session​_de​_Cambridge​.pdf [viewed 9 June 2023]. Institute of International Law (IIL)., (1984). Annuaire, session de Cambridge 1983: délibérations de l’Institut en séances plénières, vol. 60-II [online]. Paris: Institut de Droit International. [Viewed 18 January 2022]. Available from: https://​www​.idi​-iil​.org/​app/​uploads/​2018/​06/​1984​_vol​_60​-II​_Session​ _de​_Cambridge​.pdf [viewed 9 June 2023]. Jacobsson, K., (2004). Between deliberation and discipline: soft governance in EU employment policy. In: U. Mörth, ed. Soft law in governance and regulation: an interdisciplinary analysis. Cheltenham, UK: Edward Elgar Publishing. pp. 81–101. Kenner, J., (2003). EU employment law: from Rome to Amsterdam and beyond. Oxford: Hart. Klabbers, J., (1994). Informal instruments before the European Court of Justice. Common Market Law Review. 31(5), 997–1023. Klabbers, J., (1998). The undesirability of soft law. Nordic Journal of International Law. 67(4), 381–391. Knauff, M., (2010). Der regelungsverbund: recht und soft law im mehrebenensystem. Tübingen: Mohr Siebeck. Korkea-aho, E., (2015). Adjudicating new governance: deliberative democracy in the European Union. Abingdon: Routledge. Korkea-aho, E., (2018). National courts and European soft law: is Grimaldi still good law?. Yearbook of European Law. 37, 470–495.

54  Research handbook on soft law Korkea-aho, E., Lisi, G., Eliantonio, M., Maljean-Dubois, S., Truilhé-Marengo, E., Hofmann, A., Beijen, B. A., Vatovec, K. and Dobbs, M., (2020). EU environmental soft law in the member states: a comparative overview of Finland, France, Germany, Italy, the Netherlands, Slovenia and the UK [online]. EU Environmental Soft Law. [Viewed 18 January 2022]. Available from: https://​ssrn​.com/​ abstract​=​3656418. Lavergne, B., (2018). Recherche sur la soft law en droit public français. Toulouse: Presses de l’Université Toulouse 1 Capitole. McNair, A. D., (1930). The functions and differing legal character of treaties. British Yearbook of International Law. 11, 100–118. Pauwelyn, J., Wessel, R. A. and Wouters, J., eds., (2012). Informal international lawmaking. Oxford: Oxford University Press. Peters, A., (2011). Soft law as a new mode of governance. In: U. Diedrichs, W. Reiners and W. Wessels, eds. The dynamics of change in EU governance. Cheltenham, UK: Edward Elgar Publishing. pp. 21–51. Petropoulou Ionescu, D. and Eliantonio, M., (2021). Democratic legitimacy and soft law in the EU legal order: a theoretical perspective. Journal of Contemporary European Research. 17(1), 43–65. Saurugger, S. and Terpan, F., (2021). Normative transformations in the European Union: on hardening and softening law. West European Politics. 44(1), 1–20. Schachter, O., (1977). The twilight existence of nonbinding international agreements. American Journal of International Law. 71(2), 296–304. Seidl-Hohenveldern, I., (1979). International economic ‘soft law’ (volume 163). In: Collected courses of the Hague Academy of International Law. Leiden: Brill Nijhoff. Senden, L., (2004). Soft law in European Community law. Oxford: Hart Publishing. Skjærseth, J. B., Stokke, O. S. and Wettestad, J., (2006). Soft law, hard law, and effective implementation of international environmental norms. Global Environmental Politics. 6(3), 104–120. Snyder, F. G., (1993). The effectiveness of European Community law: institutions, processes, tools and techniques. The Modern Law Review. 56(1), 19–54. Ştefan, O., (2013). Soft law in court: competition law, state aid and the Court of Justice of the European Union. Alphen aan den Rijn: Kluwer Law International. Ştefan, O., (2017). Soft law and the enforcement of EU law. In: A. Jakab and D. Kochenov, eds. The enforcement of EU law and values: ensuring member states' compliance. Oxford: Oxford University Press. pp. 200–217. Ştefan, O. A., Avbelj, M., Eliantonio, M., Hartlapp, M., Korkea-Aho, E. and Rubio, N., (2019). EU soft law in the EU legal order: a literature review. King’s College London Law School Research Paper [online]. [Viewed 19 January 2022]. Available from: doi: 10.2139/ssrn.3346629. Streeck, W., (1995). From market making to state building? Reflections on the political economy of European social policy. In: S. Leibfried and P. Pierson, eds. European Social Policy. Between Fragmentation and Integration. Washington, D.C.: Brookings Institution Press. pp. 389–431. Terpan, F., (2015). Soft law in the European Union—the changing nature of EU law. European Law Journal. 21(1), 68–96. Terpan, F. and Saurugger, S., (2021), Soft and hard law in times of crisis: budget monitoring, migration and cybersecurity. West European Politics. 44(1), 21–48. Thibierge, C., (2003). Le droit souple (réflexion sur les textures du droit). Revue trimestrielle de droit civil. 4, 599–628. Thibierge, C., (2009). La force normative. naissance d’un concept. Paris: LGDJ. Trubek, D. M. and Trubek, L. G., (2005). Hard and soft law in the construction of social Europe: the role of the Open Method of Co-ordination. European Law Journal. 11(3), 343–364. Trubek, D. M., Cottrell, P. and Nance, M., (2006). ‘Soft law’, ‘hard law’, and the EU integration. In: G. de Búrca and J. Scott, eds. Law and new governance in the EU and the US. Oxford: Hart Publishing. pp. 65–94. Weil, P., (1982). Vers une normativité relative en droit international?. Revue Générale de Droit International Public. 86(1), 5–47. Wellens, K. C. and Borchardt, G. M., (1989). Soft law in European Community law. European Law Review. 14(5), 267–321.

The definition of soft law  55 Wessel, R. A., (2021). Normative transformations in EU external relations: the phenomenon of ‘soft’ international agreements. West European Politics. 44(1), 72–92. Williamson Jr., R. L., (2003). Hard law, soft law, and non-law in multilateral arms control: some compliance hypotheses. Chicago Journal of International Law. 4(1), 59–82.

4. An economic analysis of soft law as a regulatory tool Michael Faure and Niels Philipsen

1. INTRODUCTION Law and economics is generally interested in instruments that can steer behaviour of parties in order to remedy market failures.1 Formal law, that is, state law with a public enforcement mechanism, is certainly such an instrument. However, there are also other instruments that try to achieve the same goal (compliance with particular norms) without any intervention of the state, neither in setting the standards, nor in the enforcement. Law and economics recognizes a wide variety of instruments other than state law that might steer behaviour of individuals. Sometimes this is referred to as traditional law (Gutmann and Voigt 2020, p. 1), in some other cases as customary law (Parisi and Fon 2009, p. 129). Douglass North (1990) recognized that state law may often fail in situations where informal institutions are neglected. In addition to the one extreme of social norms and the other extreme of state regulation with state enforcement, there is a wide variety of shades of grey where norms emerge from various non-state actors or are enforced by non-state actors. Many of those could probably be qualified as ‘soft law’ in the sense that they either are not drafted by public authorities, or not enforced through public law mechanisms.2 In fact, the law and economics literature with respect to soft law (even though the literature does not use that particular notion) is so broad that our main goal is to show a few important examples of areas where law and economics has indicated that non-state law plays an important role in guiding human behaviour. In this chapter we will focus on four different perspectives. The first area we will focus on is the customary allocation of property rights in close-knit societies. In many Sub-Saharan states in Africa, for example, customary law still today plays an important role (Gutmann and Voigt 2020, p. 1). However, also in Western societies neighbours settle disputes without formal intervention of law, as was famously shown by Robert Ellickson (1991). These are examples of what has been referred to as ‘order without law’. The same is true of customary trade rules, sometimes referred to as a lex mercatoria.3 The equally famous study of Lisa Bernstein (1992) with respect to the customs in the diamond trade powerfully showed how all participants in that trade comply with particular implicit rules, which could probably be qualified as ‘soft law’. We will further discuss these examples in section 2. A second area we will focus on is the wide domain of so-called self-regulation. Self-regulation is a system where traders themselves formulate and enforce rules on, for example, entry into a profession or rules of conduct. Examples can be found in markets for professional services, such as those provided by architects, pharmacists, and lawyers. In practice, the self-regulatory bodies involved may be very important, both in the drafting of norms and in their enforcement, and this type of self-regulation can therefore certainly be considered as ‘law’. Moreover, in many legal systems the government is relying upon these self-regulatory mechanisms, for example by making the membership of a professional organization (and compliance with their 56

An economic analysis of soft law as a regulatory tool  57 rules) mandatory for particular professionals. That already shows that there are in practice often mixes between hard law and soft law. The government may in statutes (hard law) rely on self-regulation (soft law). That is also the case with the ‘sister’ of self-regulation, that is, private regulation. In that case it is not the business community itself that determines the norms to which they will comply, but rather a private organization, a professional third party that will, for example, as a standard-setting entity, determine particular rules. This will be discussed in section 3. A third area where soft law is important is international law.4 Some may argue that the domain of so-called international customary law is largely soft law (Parisi and Fon 2009, p. 131). Law and economics scholars like Goldsmith and Posner (2005) have, from a rational choice perspective, argued that states will only agree to particular treaties if it is in their self-interest. Precisely since the enforcement of international law does not rely on the classic coercive instruments, some have doubted that international law can be qualified as law at all. Many, however, agree that at least international law, more particularly international customary law, can be qualified as law. These issues will be examined in section 4. Finally, increasingly policy guidelines are also used at the EU level via communications, notices and similar instruments that have become an increasingly important tool of EU policy. Since similar questions arise as in the other areas covered by this chapter, also these policy guidelines at EU level merit a brief discussion in section 5.5 The goal of our chapter is to sketch how law and economics scholars view soft law. From a positive perspective, we aim to explain why particular stakeholders in specific cases might prefer to use soft law rather than hard law. Using the well-known law and economics framework (referring inter alia to information asymmetry, transaction costs and administrative costs) we will explain why in particular situations stakeholders may prefer soft law and why in particular domains hard law norms do not emerge. From a normative perspective, we will equally indicate under which conditions soft law may be able to contribute to the central (economic) goal of legal and policy instruments, that is, to correct market failures. In addition to providing these general theoretical insights on soft law from an economic perspective, we will discuss a few of the above-mentioned examples of soft law and evaluate those using the law and economics framework. The central research questions of our chapter are therefore: (1) why do certain stakeholders sometimes prefer to use soft law rather than hard law?; and (2) how can soft law contribute to solving market failures? We will attempt to answer those research questions in section 6 where we evaluate the economic literature. Section 7 provides some concluding remarks.

2.

SOCIAL NORMS AND CUSTOMS: ‘ORDER WITHOUT LAW’

Formally, social norms are not law: they are not formulated by the state or any other public authority, and they lack formal enforcement mechanisms. However, social norms can be a powerful device to steer people’s behaviour, as becomes clear also from the law and economics literature. In this section we will provide a quick scan of that literature.6 However, we begin by repeating the Coase Theorem.

58  Research handbook on soft law 2.1

The Coase Theorem

Perhaps the best-known example of order without government intervention in law and economics theory follows from the classic paper ‘The Problem of Social Cost’ by Ronald Coase (1960). Coase’s paper dealt with externality situations: situations where costs or benefits are imposed on third parties as a side effect of production or consumption. Examples of negative externalities include noise pollution caused to one’s neighbours and environmental harm as a side effect of productive activities.7 Coase showed that legal intervention is not necessary to solve externality problems, if (and only if) property rights are clearly defined and parties can ‘internalize’ the harm caused by the externality through negotiations. Crucial is that the so-called ‘transaction costs’ are low enough to make such bargaining possible. Law and economics literature has shown that these transaction costs consist primarily of the costs of information, the costs of the bargaining process (contracting costs, including the costs of legal aid if needed) and enforcement costs (Cooter and Ulen 2016, p. 88). Coase did not address social norms in his paper, however, and his analysis assumes rational behaviour of the parties involved in the conflict. 2.2

Customary Allocation of Property Rights in Close-knit Societies

In traditional societies, for example in Sub-Saharan Africa, formal state law allocating property rights did not exist. The division of agricultural land and the right to work on the land was often decided by a village chief on the basis of customary law. In many Sub-Saharan states (but also in other countries, like Indonesia and Bangladesh)8 customary law still today plays an important role.9 Law and economics literature has often applied game theory when analyzing customs, regarding them as a non-contractual strategy to change the outcome of inefficient games to efficient games (e.g., Axelrod 1986; Parisi 1995, 2011). Parisi (1995, p. 212), referring to the work of Friedrich Hayek, points to the economic advantages of spontaneous and decentralized processes like social norms and customary law compared to other institutional settings that are more formal in nature. Individuals and social groups will indicate their preferences during the formative process, so that the spontaneously emerging rule will ‘embody the aggregate effects of the independent choice of the various individuals that participate in its formation’.10 According to Parisi (1995), ethical and moral rules are likely to destabilize ‘inefficient’ activities and social arrangements, that is, those that tend to destroy wealth or impede human flourishing. In this way, in a competitive external environment, more efficient norms and customs are likely to prevail over less efficient ones. Nevertheless, the corrective process carried out by spontaneously changing norms is always incremental and is never completed (Parisi 1995, p. 216).11 Robert Ellickson (1991) in his multi-disciplinary book Order Without Law showed how, also in modern society, neighbours settle disputes without formal intervention of law. Instead of relying on a state or another central coordinator, people in the examples studied by Ellickson (1991) prefer to govern themselves through informal rules, that is, social norms. This is possible as long as people are well-informed and similarly endowed and if the cooperative behaviour of these people does not cause negative externalities on other groups in society (see also Posner 1996, p. 1697). The main example provided in the book concerns cattle ranchers and grain growers in Shasta County, California, who settle their cattle-trespassing disputes (i.e., resulting from

An economic analysis of soft law as a regulatory tool  59 damage caused by escaped cattle to third parties) by negotiations; other examples of such relatively small and homogeneous societies covered in the book include whaling grounds and photocopying centres. While Ellickson’s analysis seems to share many similarities with Coase’s (1960) classic study, it also differs in some respects. Notably, Ellickson (1991) focused also on the interaction between social norms and legal rules. Furthermore, he refuted some of the strict assumptions in Coase’s model (rational and self-interested behaviour of actors is replaced by a multi-disciplinary approach that combines law, economics, sociology and anthropology), explaining with the help of game theory how people choose to cooperate when it is to their mutual advantage. Translating Ellickson’s multi-disciplinary results to law and economics, one can conclude that low transaction costs and the presence of informal markets (where reputational effects play a role) remain important indicators to make social norms a successful type of (soft) law. 2.3

Customary Trade Rules

Dispute settlement without legal intervention also occurs in the context of customary trade rules. The lex mercatoria used by merchants in Europe in medieval times is a well-known example of this that dates back many centuries. However, also in today’s society there are examples of business rules which are not ‘hard law’ and which share similarities with the old merchant law. A famous study by Lisa Bernstein (1992) on the customs in the diamond trade demonstrated how all participants in that trade comply with particular implicit rules (reputational bonds, customary business practices, and arbitration proceedings), which could probably be qualified also as ‘soft’ law. Diamond traders have always resisted state law, instead handling disputes among members of the industry via its own institutions and sanctions (Bernstein 1992, p. 115).12 For example, arbitration is preferred over adjudication, as it is more secretive, quicker, and less expensive, thereby limiting reputational damage to the parties to a dispute, as well as limiting the actual damage suffered by a victim of breach (Bernstein 1992, p. 148). More generally, trust and reputation are important values in the diamond industry (Bernstein 1992, p. 157). In a later publication, Bernstein (1999, p. 714) reflected on the origins of unwritten commercial customs, noting that such customs are most likely to arise and endure under specific conditions: (1) transactors interact on a repeat basis; (2) transactors interact over a long period of time; and (3) the transactions are relatively similar.13 Again, we can note here a resemblance to the conditions we mentioned above, that informal law arises where transaction costs are low and where reputational effects are important and informal markets exist. 2.4

Relationship Between Law and Social Norms

Also the question how law should interact with social norms has been addressed by several scholars, most prominently by Eric Posner (e.g., Posner, 1996, 2002). Central in Posner’s analysis is that people are concerned about setting up cooperative relationships, which leads to imitative behavioural patterns. Posner argued that the resulting social norms are sometimes desirable, yet at other times they can be odious. More specifically, norms of closely-knit groups can be inefficient if they fail to enable group members to exploit the full surplus of collective action. Such inefficiencies may be caused by information problems, strategic behav-

60  Research handbook on soft law iour and (differences in) moral values, but also by the presence of negative externalities on non-group members (Posner 1996, pp. 1711–1725). Therefore, the law (i.e., both legislatures and courts) is critical to enhancing good social norms and undermining bad ones. Applying game theory, Posner (2002) examined several areas in which social norms are regulated, such as laws that govern gift-giving and non-profit organizations, family law, and laws governing speech, voting and discrimination. Some legal scholars have argued that particular types of customary law could (or should) become part of state law (and hence formal law). One could for example think of integrating certain foundational values of indigenous laws into the constitutions of African countries. Diala and Kangwa (2019, p. 205) in that respect note that foundational values such as family continuity, preservation of the ancestral home, and the duty of care owed to family members by the family head, have the potential to become constitutional principles. Starting from the notion that lack of trust in state institutions is a pervasive problem in many developing countries, Acemoglu et al. (2020) examined the relationship between formal law and informal law (more specifically: between state actors and non-state actors) in Pakistan. The authors conclude that the reliance on non-state actors is lower when the perceived quality of state-enforced law is higher. In other words, their conclusions suggest that at least to some extent state law and informal law should be considered substitutes. However, Gutmann and Voigt (2020, pp. 458–459) point out that state law and informal law (i.e., traditional law) might also complement each other, for example when they are used to settle different types of conflicts.

3.

SELF-REGULATION AND PRIVATE REGULATION

In some areas of the law, operators themselves determine the rules by which they will comply. Especially in the domain of so-called professional regulation (such as architects, physicians, accountants, or lawyers), the bodies regulating these professions often also formulate the rules. However, self-regulation can be seen in many other domains as an alternative to public regulation (see generally Philipsen 2018). In the law and economics literature self-regulation is addressed from two perspectives: the public interest and the private interest perspective. From a public interest perspective there may be particular advantages of self-regulation in some specific situations (Ogus 1995). A first theoretical advantage is that the private parties to be regulated may have better information than public authorities or can obtain the information (e.g., on the optimal care levels in the profession) at lower costs than the government. A second advantage is that it may be less bureaucratic than public regulation and is more dynamic, that is, more adaptable to changing circumstances than static public regulation (Miller 1985, pp. 897–898). A third argument advanced in favour of self-regulation is that it would be less costly than public regulation. This is not only related to the lower information costs and administrative costs (less bureaucracy), as discussed above, but also to the idea that compliance costs of the regulated community may be better taken into account by self-regulatory bodies. Furthermore, the creation and enforcement of self-regulation is often paid by the regulated community itself and will thus not negatively affect the public budget (Philipsen 2018, pp. 544–546).14 Moreover, the private sector that is responsible for drafting the self-regulation is usually profit-driven and may therefore have greater incentives than public authorities to reduce costs

An economic analysis of soft law as a regulatory tool  61 and improve services (Hart et al. 1997). Those advantages of self-regulation may obviously not all apply to the same extent in all regulated areas. Not surprisingly, self-regulation can therefore be observed for example, in markets for professional services, but less in markets where larger risks (e.g., of health and safety) are involved. The fact that the use of self-regulation is limited to particular areas can also be explained by the second perspective to regulation, which is the public choice or private interest perspective. Public choice theory has warned that industry groups will often either try to capture the regulatory process or change the process to its personal advantage. The result may be that the private interests of the regulated community prevail over the public interest. Such ‘rent-seeking behaviour’ can have two different effects. It could lead to inefficiently low standards of regulation, which would make it very easy for members to comply with these standards. However, the influence of private interests could also have the opposite result of creating inefficiently high standards. That would be the case when frontrunners in the industry (who enjoy first-mover advantages and informational advantages) may lobby for very high standards of regulation to create barriers to market entry for competitors (Vogel 1997). The dangers of private interests prevailing are especially serious when there is no countervailing power against lobbying by industry, for example in the case of pure self-regulation, where only industry determines the contents of the regulation. Olson indicated that private interest groups will especially be successful if transaction costs for the group are low and if the information costs for the public are high (Olson 1971). When the regulation is of a very complicated or technical nature, the information costs for the public will be high. Also, when the group of operators involved is relatively small and share similar interests (i.e., they are single-issue oriented) the likelihood of private interests prevailing over the public interest is high. In that particular case, the self-enforcing entities may also lack incentives to adequately enforce self-regulation. It is precisely because of these potential disadvantages of self-regulation that in practice ‘conditioned self-regulation’ is sometimes used. This refers to the situation where there is public supervision on self-regulatory agencies (Philipsen 2018, p. 546). Another way to reduce the risk of rent-seeking behaviour while keeping the advantages of self-regulation is ‘co-regulation’, which refers to joint regulation by a public regulator and private groups. Both conditioned self-regulation and co-regulation attempt to achieve the best of both worlds: having the information advantage of self-regulation while reducing the possibility that the private interests of self-regulatory agencies are served rather than the public interest. However, this could come at the expense of losing some of the advantages of flexibility and lower costs (Philipsen 2018). Another type of soft law comparable to self-regulation is private regulation. In that particular case the business entities that are subject to regulation do not themselves formulate and enforce the regulation, but other private parties. Again, this is not state-made law and would therefore qualify as soft law. Private regulation could thus be made by private organizations such as an international standard organization (ISO) and yet other private operators (so-called certifiers) could control compliance with regulation and thus certify that industry complied with the regulation. There are today many instances of such private regulatory approaches in the market, ranging from food labelling to green building to video game rating systems (Faure and Partain 2019, p. 109). Again, the arguments for private regulation are largely the same as those in favour of self-regulation: expected lower costs, higher flexibility, and better information. Also in this case, one can often notice that public regulation (by government) may require

62  Research handbook on soft law certification by private institutions or compliance with privately set standards. Therefore, the line between government and private regulation becomes blurred and hybrid forms of governance emerge. This can for example be seen in the EU Timber Regulation (995/2010). It requires EU traders who place timber products on the EU market to exercise due diligence. Compliance with those due diligence systems is verified by private certifiers.

4.

INTERNATIONAL SOFT LAW

Another example of a situation where there are rules with which many parties comply without a formal external enforcement mechanism can be found in the area of international law. Some consider the system of international law as the most notable illustration of the recognition of social norms by the legal system, as there is no central legislative authority (Parisi and Fon 2009, p. 131). It is striking that in Article 38(1) of the Statute of the International Court of Justice, which states the sources of international law, ‘international custom, as evidence of a general practice accepted as law’ is even recognized as a primary source of law, having the same status as international conventions. It is an example where customary rules are granted a legitimate status and then receive the same effect as other primary sources of law (Parisi and Fon 2009, p. 132). That still raises the question why states, in the absence of a formal enforcement mechanism, consider themselves bound by international custom. The question could even be asked more broadly: why do states accept to conclude treaties and why is there (to a large extent) compliance with treaties? Some law and economics scholars, notably Goldsmith and Posner (2005), are critical concerning the traditional explanation in international law. They are particularly critical of the notion that customary rules derive their force from the occurrence of a uniform practice and of the belief that adherence to them is obligatory (so-called opinio juris). They argue that the explanation of the binding nature of customary law is in fact a circular one as the rules would be binding because states adhere to them (Goldsmith and Posner 2005; Posner and Sykes 2013, pp. 50–63). As states normally act in their self-interest, the question arises why states would agree to customary rules that provide for the common good. The answer is that customary law as a source of international law creates state reliance (Parisi and Fon 2009, pp. 134–135) and that states may enjoy mutual gains via cooperation (Posner and Sykes 2013, p. 52), even though cooperation may often be difficult, given the large heterogeneity between the states. It is this heterogeneity that may also explain why mutually beneficial cooperation does not emerge. The explanation for this can be provided with the help of game theory (Barrett 2007). The fact that mutually beneficial agreements sometimes do not emerge can be illustrated with an example of transboundary environmental pollution (Jacobsen and Brown Weiss 1995). Individual states may all hope that others take the necessary measures to reduce the source of a transboundary pollution problem. Thus, they get a free ride on the potential efforts of other states. Mutual efforts (investments in pollution reduction) would be in the interest of all states, but the fear that one of the states may not take actions may lead the other states to believe that they should not invest in prevention and can thus take a free ride on the efforts of others. Obviously, if all states have the same belief, a prisoner’s dilemma emerges and no action at all will be taken. This example also illustrates that the goal of international law (either via customary law or treaty-making) should be to correct interstate failures that result from this

An economic analysis of soft law as a regulatory tool  63 free-riding behaviour. However, the above raises the question why states would voluntarily conclude treaties, for example to reduce pollution, thereby potentially imposing costs upon their domestic industry. At first blush, there is indeed a paradox: the prisoner’s dilemma would prevent international agreements from being created, whereas in reality there are an amazing number of international conventions.15 There are two, partially competing and partially complementary, views on this. The traditional economic approach (represented in Goldsmith and Posner 2005; Posner and Sykes 2013) is that states will mainly conclude agreements and comply with those when it is in their self-interest (see also Guzman 2002). States are from this perspective considered as rational actors (Barrett 2007, p. 235; Mitchell 2007, p. 902). The rational choice approach relies strongly on the principle of the sovereignty of states. In this approach, states are sovereign actors in the international arena, meaning that they are free to act as they find necessary, unrestricted by any external authority or rules. From this perspective governments only accept those international treaties that are in their own interest, and a breach would be viewed as intentional. Towards the end of the 1990s, this traditional view of compliance problems was increasingly criticized in scholarly writings, criticism that goes hand in hand with the so-called new approach to sovereignty. The proponents of this new approach argue that states should no longer be seen as completely sovereign entities but as entities willing to accept limits on their original sovereign rights for the benefit of the environment, future generations, or the international community as a whole.16 The international community is increasingly organized in ‘regimes’.17 These regimes consist of frameworks with relatively well-developed sets of rules and norms on specific subjects. The development of regimes can be placed between the traditional concept of sovereignty, leaving the states unbound, and a comprehensive world order, placing the states within a new world governance. Examples of important regimes are the climate change regime, constructed around the 1992 United Nations Framework Convention on Climate Change, the 1997 Kyoto Protocol and the 2015 Paris Agreement, and the international trade regime, based on the agreements concluded under the World Trade Organization. With the development of these regimes, ‘sovereignty no longer consists in the freedom of states to act independently, in their perceived self-interest, but in membership in reasonably good standing in the regimes that make up the substance of international life’ (Chayes and Handler Chayes 1995, p. 27). States’ interests are increasingly determined by their membership in, as well as good reputation under, these regimes.18 In the literature evidence for both perspectives is presented. For example, it is argued that participation in an international environmental agreement is often self-selected by countries that can easily comply with the obligations in an agreement because it will require no behavioural change (Barrett 2007, p. 250). Others argue that states conclude international agreements following a particular social identity: the agreement would be based on an assessment of the right thing to do. This alternative approach (to the rational actor model) is more popular among lawyers (Mitchell 2007, p. 902). Even though the starting points of both models are different and to some extent competing, Mitchell (2007, pp. 902–903) points out that the two logics are to some extent also mutually reinforcing. In sum, international law is a good example of the importance of an (international) legal order that functions without a formal enforcement mechanism. Customary international law is recognized as a formal source of law and treaties are concluded even when it may not be in the individual interest of the state, but rather for the common good. It is especially Andrew

64  Research handbook on soft law Guzman (2008) who explained that states’ compliance with international law can be explained by the ‘3 Rs’: reciprocity, retaliation and especially reputation. Reputation is an important explanation for why states cooperate to enjoy better long-term collective benefits at the expense of short-term individual benefits.

5.

POLICY GUIDELINES AT EU AND NATIONAL LEVEL

European and national legislators often make use of instruments like guidelines, communications, frameworks, and notices. These instruments, which have no legal force but may have legal and practical effects, have also been defined in the legal literature as ‘soft law’ (Snyder 1993; Ştefan et al. 2019). Now accounting for a substantial percentage of all EU law, soft law instruments are used more and more at EU level, particularly since the Maastricht Treaty of 1989 (Ştefan et al. 2019).19 In legal domains such as competition law, environmental law, financial regulation, and many others, nowadays a lot of such EU soft law is used, although the ‘softness’ of the legal instrument differs across fields. Particularly in relation to competition law and state aid, where the Commission’s Directorate-General for Competition (DG COMP) is the enforcement agency of competition law, soft law is binding on the Commission itself and indirectly it may also bind national competition authorities. Furthermore, while courts are allowed to deviate from these rules, they may not do so very often, as this may be at the expense of the predictability (i.e., legal certainly) of competition rules. One may therefore conclude that soft law is sometimes not soft at all. While there is not much existing economic literature that directly addresses the question why soft law is so often used by the European Commission (and national regulators) and whether the use of such soft law is actually an efficient way of addressing market failures, the literature we reported in previous sections of this chapter already provided some indications.20 On the one hand, the relative flexibility of such soft law as well as the low administrative costs of drafting it are clear advantages compared to hard law like Regulations and Directives. On the other hand, it may be important to have all relevant stakeholders involved in the drafting of soft law (also to profit from their information advantages), which would require a longer and hence less flexible and more costly drafting process. In other words, there may be a trade-off between costs and flexibility on the one hand and the ‘quality’ of the rules in terms of addressing market failures (which as we explained may require the involvement of different interest groups in the formation of soft law) on the other. While legal scholars often point to the importance of legitimacy in the drafting of EU soft law, in the law and economics literature the topic ‘involvement of stakeholders’ fits in with the private interest (public choice) analysis of interest group behaviour, which we briefly introduced in section 3 above.21

6.

EVALUATION OF THE ECONOMIC LITERATURE

So far we have provided examples of what could be qualified as soft law as they have appeared in the law and economics literature and in practice in four different domains. Note (again) that these regimes are as such not identified as ‘soft law’ in the law and economics literature but appear under different headings such as for example ‘social norms’, ‘customary trade rules’ (both discussed in section 2), ‘private regulation’ (section 3), or ‘international customary law’

An economic analysis of soft law as a regulatory tool  65 (section 4). Soft law hence seems to be a notion that is used more in the legal literature than in the economic literature. Another starting point is that the law and economics literature largely recognizes the importance of non-state law. For example, North (1990) clearly showed the failure of many law reforms in developing countries that were simply based on formal legislation disregarding the existing system of norms and traditions, in other words soft law. The way in which law and economics approaches soft law is therefore primarily from a positive perspective. That means that law and economics recognizes the existence and importance of soft law in many situations and relationships without immediately deriving normative implications from this. But law and economics does stress the importance of recognizing soft law, such as customary law in developing countries and trade customs in particular industries. The existence of soft law may limit the possibility for the formal legislator to intervene and achieve changes. This has for example also been recognized concerning attempts of land rights reform and more particularly introducing formal titling for land. As in Sub-Saharan Africa land rights were often either unregulated or regulated via customary rules, the state-imposed formalization of titling in some cases led to an outright failure or to legal uncertainty and an increase of disputes. That was caused by the combined application of customary and formal law, a ‘legal pluralism’ creating institutional shopping (Platteau 1996; Lavigne-Delville 2000; Deininger and Feder 2009). One implication from this literature is that formal law (e.g., land formalization) may fail (not being able to achieve its goals) if it does not take into account pre-existing customary rules. At a general level one should be very careful with normative conclusions concerning the desirability of soft law, because the assessment very much depends on the specific context within which soft law is used and on the reasons why it was employed in the first place. For example, as far as customary rules in Sub-Saharan Africa are concerned, empirical evidence seems to indicate that the countries where customary law plays an important role, score low on the rule of law index as well as indices related to GDP (Gutmann and Voigt 2020). But that obviously depends on the type of soft law one is considering. It may for example not be the case at all in the communities in California where Ellickson (1991) described the ‘order without law’. There is hence no causation between the use of soft law generally, the rule of law, and economic growth. At the normative level the desirability of soft law may be related to the two questions which we posed at the start of this chapter: (1) why do certain stakeholders sometimes prefer to use soft law rather than hard law; and (2) how can soft law contribute to solving market failures? The question under which circumstances specific actors prefer soft law may strongly depend upon the type of soft law referred to and upon the reasons why actors call on soft law. In some cases, soft law may be used as a second best to fill in for the weaknesses of formal law. That could for example well be the case with traditional law which still is strongly prevalent in many developing countries. For example, in Indonesia users reported to be more satisfied with the performance of the customary system (adat law) and respondents in Somaliland reported higher satisfaction with the customary justice system than with the formal (Islamic) legal system (Gutmann and Voigt 2020, p. 1). In those cases, the preference for customary law could to a large extent be explained by the weakness of the formal legal system (and more particularly the judiciary) which in some countries is tainted by partiality, corruption, and inefficiency. In those cases, the choice for soft law can hence be explained by the dissatisfaction with (or in some cases even absence of) the formal legal system. However, the presence of soft

66  Research handbook on soft law law is certainly not confined to developing countries. We provided examples, both of ‘order without law’ (section 2), but also of self/private regulation (section 3) and the increasing use of policy guidelines (section 5) where the choice for soft law is not necessarily a negative choice against formal state law, but rather a positive choice given the potential advantages of particular soft law arrangements. The literature points at the decentralized norm-making within soft law and the possibility that sustainable, welfare enhancing customs could be created (Parisi and Fon 2009, p. 137). We already mentioned the potential benefits of soft law from the actors’ perspective (answering our first research question): more expertise, better information, flexibility and thus the guarantee for stakeholders that they can create rules perfectly matching their preferences. Moreover, actors can keep compliance mechanisms and enforcement in their own hand, thus encouraging social stability (e.g., also in trade relationships or in close-knit societies), avoiding the social disturbance that could be created with formal law and formal adjudication via the court system.22 Those reasons why stakeholders may prefer soft law could also be conditions under which soft law may efficiently remedy market failures. The better expertise and information advantage of the stakeholders involved could enable the creation of rules at lower costs (for example in the case of private regulation) and could equally stimulate spontaneous compliance (thus reducing compliance costs) given the higher legitimacy that norms would have if they are created by the community of regulatees themselves (under self-regulation). However, one has to remain careful, as soft law can also have substantial disadvantages, not only from society’s, but also from the individuals’ perspective. To the extent that soft law consists of unwritten norms, there may be difficulties in simply obtaining information on the contents of the rules. Soft law, especially traditional law, is not always written and consists sometimes only of oral customs passed on via tradition and providing a power monopoly to chiefs that apply the rules. Customary rules may also confirm existing power imbalances (e.g., the prohibition in adat law on Bali for women to inherit) and are often not adapted to changes in society. Customary rules are also much more difficult to change. The traditional argument in favour of soft law of having more flexibility therefore certainly does not apply to all types of soft law. Recall that especially in developing countries, traditional law is often a negative choice given the inadequacy of state law (and the state adjudication system). But the use of traditional law is especially prevalent in countries with low income and low respect for the rule of law (Gutmann and Voigt 2020, p. 11). In other words, traditional communities may use customary law, not because it is a deliberate choice, but sometimes simply because there is no adequate alternative. Again, in other situations, for example, when it concerns customary trade rules, there may be a deliberate (positive) choice to rely on soft law as stakeholders judge that the advantages of soft law (lower information cost, more flexibility) make them choose the particular soft law arrangement. In some cases, one can understand why stakeholders use soft law, but it is not always clear whether that aligns with society’s interest. Take the use of self-regulation and private regulation. From the stakeholders’ (industrial operators) perspective, the advantages are clear: self-regulation allows them to make the rules themselves, it avoids state regulation with which they might not be able to comply as well as state enforcement. However, self- and private regulation could either result in inefficiently low standards or in the creation of barriers to entry (i.e., when standards are too stringent). That is a clear example of a situation where

An economic analysis of soft law as a regulatory tool  67 private interests explain why stakeholders choose soft law, but where this is not necessarily the optimal tool to cure market failures. In sum, one has to carefully analyze the specific type of soft law and to distinguish what the reasons are why stakeholders would like to use soft law and whether this also corresponds with a society’s interest in using soft law to remedy a market failure. Specific attention needs in that respect to be given to the potential dangers inherent in soft law (e.g., power abuse by chiefs in traditional law or private interests prevailing in private regulation). It is therefore not possible to generally evaluate soft law from an economic perspective, as the desirability of soft law as a regulatory instrument will to an important extent depend upon the specific context within which the particular soft law instrument is applied, also in view of the adequacy of the (state-provided) alternatives.

7.

CONCLUDING REMARKS

From an economic perspective, legal rules are important tools to correct market failures. To the extent that the conditions of the Coase theorem are not met (and market failures cannot be corrected via voluntary contracting) there will be a role for law. However, economists have often stressed that law should not necessarily be limited to formal state law. It is for that reason that economists have paid a lot of attention to alternatives for formal state law, arguing that for example informal laws, traditional laws, customary laws, self-regulation, and trade practices can under certain circumstances also serve the goal of remedying market failures. This already shows that even though the concept of soft law is as such not to be found explicitly in the economic literature, several alternatives to formal state law are discussed in the economic literature from different perspectives. One perspective is related to the question why certain stakeholders sometimes prefer to use soft law rather than hard law. In that respect, economists traditionally make a distinction between a private interest (public choice) and a public interest perspective. From a public interest perspective, that is, focusing on the role of law to remedy market failures and increase social welfare, soft law can suggest particular advantages over hard law with respect to information, transaction, and administrative costs and may have a higher flexibility. Those advantages do, however, not apply to the same extent to all of the examples discussed. For example, higher flexibility and lower information costs may not always apply to some forms of (traditional) customary law, which is not always very flexible and where information costs to find out the contents of the rules may even be rather high. Another perspective is the private interest perspective. One way of looking at this is that some of the (business) partners may prefer soft law, for example, customary trade rules in the diamond industry. Soft law may work better in informal and more secretive markets based on reputation; the application of customary trade rules is also likely to lead to quicker dispute resolution. The same certainly applies to international trade where the importance of reciprocity and reputation are strongly stressed, more particularly in the work of Guzman (2008). In these situations (where private parties prefer soft law, like in customary trade rules) the preference of parties may to a large extent align with the public interest. In other words, many of those solutions chosen by private parties (also customary rules in close-knit societies) may well be welfare maximizing, especially in developing countries where formal state law is often not available, not trustworthy, not enforced or simply too costly and inefficient.

68  Research handbook on soft law However, the other angle of the private interest perspective is that all types of legal rules (including soft law) could also be abused to the personal benefit of particular interest groups (rent-seeking behaviour). That may be a problem if some parties have an information advantage concerning the contents of soft law and do not fully share their private information. For example, particular village chiefs in traditional law systems could abuse their information advantage. Also some forms of self-regulation could merely be to the advantage of the interest group concerned (attempting to self-regulate in order to avoid formal state regulation). In those cases, soft law may clearly be inefficient. For that reason, most of the economic literature has rather focused on the question why it is that in some cases soft law rather than hard law emerges; the normative question of whether under particular circumstances soft law may be more effective to cure market failures than formal state law is much more complicated to answer. This cannot be answered in general terms as, as this chapter has shown that soft law may appear under a wide variety of different types and headings, which each require a separate analysis, also with respect to their effectiveness. So far, the law and economics research has strongly focused on explaining the emergence of soft law in particular contexts. There has so far been less focus on the policy question under which circumstances soft law may (eventually in a particular combination with hard law) be expected to reach better results than formal state law. That would require a delicate empirical analysis into the effectiveness of a particular soft law regime, also in comparison to alternatives. Those are undoubtedly interesting points for further law and economics research.

NOTES 1. Market failure refers to a situation in which the allocation of goods and services by the market mechanism is not efficient. For a further discussion see, for example, Cooter and Ulen (2016, pp. 11–54) and Philipsen (2018). 2. The literature distinguishes between various examples of soft law. For a categorization see inter alia Gutmann and Voigt (2020, p. 5). 3. See also Jansen’s chapter in this volume. 4. See also Klabbers’s chapter in this volume. 5. See more generally on these tools Senden (2013). 6. Due to the limited scope of this contribution, we can only focus on a few well-known publications on this topic. We refer the reader to the bibliographies in these publications for further reading. 7. Note that externalities can also be positive. An example would be the benefits to society generated by private investments in R&D. 8. Gutmann and Voigt (2020, pp. 445–446), provide a brief overview of literature on the use of customary courts and (what they call) ‘traditional law’ in some African and Asian countries, while referring also to the example of Sharia councils, which operate even in Western countries. 9. On the historical development of African customary law and its relation with macro- and micro-economic developments, see Snyder (1984). 10. Note that a similar analytical framework is used in the so-called ‘economics of federalism’ literature, which addresses the question whether a particular legal rule should be designed at local, national, or transnational level. This literature advocates a bottom-up approach. See, e.g., Van den Bergh (2016). 11. Note that the view that efficient social norms prevail has been challenged by several authors, including Eric Posner. See the discussion below. 12. It should be noted, however, that the customs and institutions in the diamond industry did not emerge because of particular shortcomings in the legal system; they developed historically. 13. Bernstein (1999) lists several authors who support this view, including Ellickson (1991).

An economic analysis of soft law as a regulatory tool  69 14. However, some of the administrative costs of self-regulation may still be passed on to consumers via the price mechanism. 15. For example, in the domain of the just mentioned example regarding environmental pollution, there are according to Mitchell (2003) more than 700 multilateral environmental agreements and over 1,000 bilateral. Obviously that amount may even have increased now, 18 years later. 16. This new idea is probably best formulated by Chayes and Handler Chayes (1995), especially Chapter 1. 17. For a review of the early literature on regimes, see Levy et al. (1995) and Young (1999). 18. The role of reputation in complying with international agreements is strongly stressed in the work of Andrew T. Guzman. See Guzman (2005, pp. 579–612) and Guzman (2008, especially Chapter 3). 19. More generally, see the various working papers published by the SoLaR research project on EU soft law. 20. Notably the literature on private regulation: see Philipsen (2018) and references therein. 21. See for the link between legitimacy and public interest theory also Faure and Philipsen (2020). 22. See on the importance of social stability (i.e., the need for quick and secretive dispute resolutions and reputational mechanisms), more particularly our discussion in section 2 of the diamond trade, based on Bernstein (1992).

REFERENCES Acemoglu, D., Cheema, A., Khwaja, A. I. and Robinson, J. A., (2020). Trust in state and non-state actors: evidence from dispute resolution in Pakistan. Journal of Political Economy. 128(8), 3090–3147. Axelrod, R., (1986). An evolutionary approach to norms. The American Political Science Review. 80(4), 1095–1111. Barrett, S., (2007). An economic theory of environmental law. In: D. Bodansky, J. Brunnée and E. Hey, eds. The Oxford handbook of international environmental law. Oxford: Oxford University Press. pp. 232–262. Bergh, R. van den, (2016). Farewell utopia?: why the European Union should take the economics of federalism seriously. Maastricht Journal of European and Comparative Law. 23(6), 937–964. Bernstein, L., (1992). Extra-legal contractual relations in the diamond industry. Journal of Legal Studies. 21(1), 115–157. Bernstein, L., (1999). The questionable empirical basis of Article 2’s incorporation strategy: a preliminary study. University of Chicago Law Review. 66(3), 710–780. Chayes, A. and Handler Chayes, A., (1995). The new sovereignty: compliance with international regulatory agreements. Cambridge, M.A.: Harvard University Press. Coase, R. H., (1960). The problem of social cost. The Journal of Law & Economics. 3, 1–44. Cooter, R. and Ulen, T., (2016). Law & economics. 6th ed. Boston, M.A.: Pearson Education. Deininger, K. and Feder, G., (2009). Land registration, governance, and development: evidence and implications for policy. The World Bank Research Observer. 24(2), 233–266. Diala, A. C. and Kangwa, B., (2019). Rethinking the interface between customary law and constitutionalism in sub-Saharan Africa. De Jure Law Journal. 52(1), 189–206. Ellickson, R. C., (1991). Order without law. How neighbours settle disputes. Cambridge, M.A.: Harvard University Press. European Parliament and Council regulation (EU) no 995/2010 laying down the obligations of operators who place timber and timber products on the market (2010) Official Journal L 295, p. 23. Faure, M. G. and Partain, R. A., (2019). Environmental law and economics. theory and practice. Cambridge, UK: Cambridge University Press. Faure, M. and Philipsen, N., (2020). Standardisation from a law and economics perspective. In: M. Eliantonio and C. Cauffman, eds. The legitimacy of standardisation as a regulatory technique: a crossdisciplinary and multi-level analysis. Cheltenham, UK: Edward Elgar. pp. 156–178. Goldsmith, J. L. and Posner, E. A., (2005). The limits of international law. Oxford: Oxford University Press.

70  Research handbook on soft law Gutmann, J. and Voigt, S., (2020). Traditional law in times of the nation state: why is it so prevalent?. Journal of Institutional Economics. 16(4), 445–461. Guzman, A. T., (2002). A compliance-based theory of international law. California Law Review. 90(6), 1823–1888. Guzman, A. T., (2005). The design of international agreements. European Journal of International Law. 16(4), 579–612. Guzman, A. T., (2008). How international law works: a rational choice theory. Oxford: Oxford University Press. Hart, O., Shleifer, A. and Vishny, R. W., (1997). The proper scope of government: theory and an application to prisons. The Quarterly Journal of Economics. 112(4), 1127–1161. Jacobsen, H. K. and Brown Weiss, E., (1995). Strengthening compliance with international environmental accords: preliminary observations from a collaborative project. Global Governance. 1(2), 119–148. Lavigne-Delville, P., (2000). Harmonizing formal law and customary rights in French-speaking Africa. In: C. Toulmin and J. Quan, eds. Evolving land rights, policy, and tenure in Africa. London: International Institute for Environment and Development (IIED). pp. 97–122. Levy, M. A., Young, O. R. and Zürn, M., (1995). The study of international regimes. European Journal of International Relations. 1(3), 267–330. Miller, J. C., (1985). The FTC and voluntary standards: maximizing the net benefits of self-regulation. Cato Journal. 4(3), 897–903. Mitchell, R. B., (2003). International environmental agreements: a survey of their features, formation, and effects. Annual Review of Environment and Resources [online]. 28, 429–461. [Viewed 9 December 2021]. Available from: doi: 10.1146/annurev.energy.28.050302.105603 Mitchell, R. B., (2007). Compliance, effectiveness, and behaviour change in international environmental law. In: D. Bodansky, J. Brunnée and E. Hey, eds. The Oxford handbook of international environmental law. Oxford: Oxford University Press. pp. 894–922. North, D. C., (1990). Institutions, institutional change and economic performance. Cambridge: Cambridge University Press. Ogus, A. I., (1995). Rethinking self-regulation. Oxford Journal of Legal Studies. 15(1), 97–108. Olson, M., (1971). The logic of collective action: public goods and the theory of groups. Cambridge, M.A.: Harvard University Press. Parisi, F., (1995). Toward a theory of spontaneous law. Constitutional Political Economy. 6(3), 211–231. Parisi, F., ed., (2011). Production of legal rules. Cheltenham: Edward Elgar. Parisi, F. and Fon, V., (2009). The economics of lawmaking. Oxford: Oxford University Press. Philipsen, N. J., (2018). The role of private actors in preventing work-related risks: a law and economics perspective. European Public Law. 24(3), 539–553. Platteau, J.-P., (1996). The evolutionary theory of land rights as applied to Sub-Saharan Africa: a critical assessment. Development and Change. 27(1), 29–86. Posner, E. A., (1996). Law, economics, and inefficient norms. University of Pennsylvania Law Review. 144(5), 1697–1744. Posner, E. A., (2002). Law and social norms. Cambridge, M.A.: Harvard University Press. Posner, E. A. and Sykes, A. O., (2013). Economic foundations of international law. Cambridge, M.A.: Harvard University Press. Senden, L., (2013). Soft post-legislative rulemaking: a time for more stringent control. European Law Journal. 19(1), 57–75. Snyder, F. G., (1984). Customary law and the economy. Journal of African Law. 28(1-2), 34–43. Snyder, F. G., (1993). The effectiveness of European Community law: institutions, processes, tools and techniques. The Modern Law Review. 56(1), 19–54. Ştefan, O. A., Avbelj, M., Eliantonio, M., Hartlapp, M., Korkea-aho, E. and Rubio, N., (2019). EU soft law in the EU legal order: a literature review. King's College London Law School Research Paper [online]. [Viewed 9 December 2021]. Available from: doi: 10.2139/ssrn.3346629 Vogel, D., (1997). Trading up: consumer and environmental regulation in a global economy. Cambridge, M.A.: Harvard University Press. Young, O. R., ed., (1999). The effectiveness of international environmental regimes: causal connections and behavioral mechanisms. Cambridge, M.A.: MIT Press.

5. Anthropology and soft law Filippo M. Zerilli

1.

SOFT LAW AS NON-BINDING COERCIONS

In a ground-breaking essay originally published in 1973, probably one of the most influential scholars in legal anthropology, anthropologist and lawyer Sally Moore introduced the concept of the ‘semi-autonomous social field’ (Moore 1973). This concept opened up innovative ways of considering ‘law’ and legal arrangements in anthropology (and law) and helped generations of researchers go beyond the accepted (legal positivist) idea according to which law is simply a uniform body of regulations and coherent interpretations.1 In fact, any system of legal arrangements (whether it be, as in Moore’s examples, the New York dress industry or the lineage-neighbourhood complex among the Chagga of Kilimanjaro) is not completely autonomous or isolated. According to Moore, all observable social fields are linked to ‘larger settings’ (Moore 2000 [1973], p. 78). Subsequent developments in the anthropology of law, to which Moore herself made tremendous contributions on both the ethnographic and theoretical levels (e.g., Moore 2001, 2002, 2005) have sought to address the question of the extent to which these ‘larger settings’, where norms are produced and negotiated, might be theorized and subjected to ethnographic scrutiny, and in what way this can be done (e.g., Nader 2005; Snyder 2005). From this perspective, the concept of a semi-autonomous social field remains an extremely valuable research tool; however, it is reasonable to ask what types of challenges the globalizing context in which territories and populations are currently organized and governed presents. In other words, it is crucial to explore how specific legal arrangements are produced and enforced within the complex network of national, international, supranational, and non-national sites where norms are established. To what extent can the emerging ‘global legal environment’ (Berman 2012) be explored and framed in ethnographic terms, beyond the framework of links between the small-scale study of a semi-autonomous social field and the larger settings, helping us to imagine and confront unexpected ‘topographies of power’ (Ferguson 2004) as emerging forms of ‘neoliberal governmentality’ (Ferguson and Gupta 2002)? In this chapter, I argue that a number of concerns related to such issues might be fruitfully framed focusing on soft law and its uncertain or ambivalent normative status, exploring how it actually operates in a given social context beyond conventional analytic distinctions such as centre and periphery, local and global, domestic law and international legal regime, and above all public and private. In order to better assess the kind of socio-legal processes I draw attention to, and before I specifically consider some ethnographic analyses regarding soft law, it is worth clarifying what soft law is about from a legal standpoint, when it appeared, and how it has been received and debated among legal scholars and law practitioners. The name itself, soft law, requires at least a few words of clarification to dissipate any possible misunderstanding. In fact, we imagine that the adjective ‘soft’ denotes, at least metaphorically, ‘flexible’ or ‘informal’ legal practices for many readers, but within the field 71

72  Research handbook on soft law of legal scholars and law practitioners, the expression soft law immediately materializes into a specific set of legal instruments. Despite controversies regarding its normative status and its still unclear, fuzzy, or nuanced contours from a technical point of view, soft law is a legal device any student provided with a broad university background in international or European law would easily recognize. Discussing European Community law, Francis Snyder gave a celebrated description according to which soft law is defined as ‘those rules of conduct which, in principle, have no legally binding force but which nevertheless may have practical effect’ (1993, p. 16). Soft law is typically embodied within non-binding legal instruments such as ‘recommendations’ or ‘declarations’, but also within ‘resolutions’, ‘codes of conduct’, ‘guidelines’, and ‘opinions’ (see Somma 2009). Most of these instruments take the form of written documents commonly produced by international organizations such as the Organization for Economic Co-operation and Development, the World Trade Organization, the International Monetary Fund, or the World Bank, just to name a few of the most influential actors in economic globalization, occasionally depicted as ‘informal global legislators’ (Mattei and Nader 2008, p. 45). However, even a simple draft proposal elaborated by groups of international experts could possibly fit into the soft law category. As a matter of fact, soft law instruments are widely used in many fields of law such as trade law, environmental law, administrative law, tort law, human rights law, and even criminal law (e.g., Kirton and Trebilcock 2004; Somma 2009). However, contrary to widespread assumptions and despite their recent proliferation, soft law technologies are not completely new (see also Jansen in this collection). Legal scholar Anna Di Robilant (2006) attributes the interest of many contemporary lawyers and legal scholars in (notably European) soft law to two major sources. The first, which she calls ‘neo-medievalist’, goes as far back as medieval mercantile law, also known as lex mercatoria.2 The second one, which she labels ‘social’, locates soft law mechanisms in the antiformalist reaction of European jurists of the late nineteenth and early twentieth centuries, that is, within the organicist and pluralist tradition of European continental law. Surely, as Di Robilant convincingly argues, both the neo-medievalist and social sources are rhetorical constructions of contemporary scholars and law practitioners participating in the soft law versus hard law debate on European harmonization, according to their different professional and political agendas. In spite of such allegedly remote sources, soft law has acquired considerable popularity under neoliberal globalization.3 In particular, it has developed and been conceptualized within the field of international public law during the 1970s, notably in the United States. Its role has become particularly salient with the adoption, within the European Union, of the Open Method of Coordination (OMC), an inter-governmental means of governance based largely on soft law instruments, elaborated within the Lisbon strategy adopted since 2000 (Radaelli 2003; Trubek and Trubek 2005; see also Katcherian 2010 and van Gerven and Stiller in this collection). Consequently, soft law technologies have increasingly attracted the attention of numerous legal scholars and law practitioners, generating contrasting feedback and reactions. According to Jan Klabbers (2007, p. 326), one of the most perceptive scholars among soft law critics, soft law [...] fits neatly with the disappearance of the public/private distinction. Formal treaties may still be the instrument of choice for the exercise of public authority on the international level, but much regulation takes place through informal channels, by informal means, and in informal instruments which are not treaties but are still supposed to engender normative effects or can be seen to

Anthropology and soft law  73 engender normative effects. Indeed, in a world where the distinction between public and private is no longer clearly demarcated, public and private authority too run into each other; as a result, it may well be that hard law emanating from public authority will become the exception rather than the norm; perhaps, in such a world, soft law is the most plausible form that law can take, precisely because it remains unclear who exercises authority, and on what basis.

Interestingly, compliance in soft law is not enforced by specific personnel or formal institutional bodies such as judges or courts but is made effective through moral suasion, social pressure, and above all self-regulation. However, despite their voluntary (non-binding) character, it would be mistaken to consider such soft regulations lacking a coercive dimension.4 As we could easily demonstrate simply looking at our own social experiences of citizens acting in different social fields, individual as well as institutional actors often perceive what legal scholars would call soft law as hard regulations and behave accordingly. The point to underline here is not the continuum between soft and hard normativity. The important point to stress is that a number of policies prove effective not because they are provided with features lawyers are ready to recognize as hard law (i.e., legally binding provisions), but precisely because a number of enforcing soft mechanisms such as shaming, conformity, persuasion, self-interest, opportunity, alarm or distress are indeed effective. From this perspective, soft or non-binding rules can be as coercive as binding rules and agreements. This is why it seems appropriate to refer to soft law also as ‘non-binding coercions’, apparently an oxymoron that alludes to those specific procedures and mechanisms aimed at obtaining compliance notwithstanding their non-justiciable character and lower degree of formality. Although lawyers would traditionally include such enforcing mechanisms amid ‘atypical’ law sources, the long-standing anthropological interest in social ordering in small-scale societies has produced important insights on legal processes in which shame, social pressure, moral suasion, and imitation play a significant role, as is clearly shown, for example, by Max Gluckman’s seminal article on gossip and scandal (Gluckman 1963). Interestingly, legal anthropologist Sally Merry, reflecting on the enforcing mechanisms of the international human rights regime, has argued that ‘the implementation of international laws bears some similarity to that of customary village law’ (Merry 2007, p. 162). In an earlier essay, Merry (2006a, p. 101) explicitly refers to gossip and scandal as mechanisms of international law: Gossip and scandal are important in fostering compliance internationally as they are in small communities. Social pressure to appear civilized encourages countries to ratify international legal treaties much as social pressure fosters conformity in small communities. Countries urge others to follow the multilateral treaties they ratify, but treaty monitoring depends largely on shame and social pressure.

Stressing similarity between customary village law and international law is not merely a way to recognize the need to de-exoticize the discipline of anthropology (e.g., MacClancy 2002), a slow process still in motion, at least starting from Laura Nader’s well-known incitement for anthropologists to ‘study-up’ (Nader 1972). In fact, it is an invitation to directly deal with transnational legal processes per se (Merry 2006b). Throughout this chapter, I contend that this is a challenge which an ethnographic approach to soft law is in a good position to face. Indeed, I suggest that attention to specific forms and practices of regulations such as soft law might enrich our understanding of the emerging transnational legal space and help us clarify some aspects of the politics of the globalization of law. More specific, I argue that if one wants to understand or trace the current global transformation of law and legality, it is not

74  Research handbook on soft law only analytically useful but a practical necessity to look at particular sites and locations where such non-legally binding coercions originate and take shape through ordinary, mundane social practice.

2.

SOFT LAW AND TRANSNATIONAL LEGAL PROCESSES

Unquestionably, as many scholars dealing with globalization have noted, we are currently facing a wide-ranging economic, political, and cultural transformation, including legal and policy-making processes and dynamics (e.g., Likosky 2002; Tsing 2002; Westbrook 2004; Ong and Collier 2005; Edelman and Haugerud 2005; Sassen 2006). Particularly in Europe, the central role played by the nation state is currently challenged by a model of supranational and multilevel governance embodied by the European Union (Bobbio 2005). Some authors have suggested portraying this change as a shift ‘from government to governance’, thus implying the emergence of new methods in the decision-making process (see, e.g., Held 1995). Obviously, what is at stake is not simply a transformation from one governance model to another; rather it is a complex process in the making, one whose outcome, according to many observers, is uncertain and far from producing a homogeneous transnational legal code. However, it goes almost without saying that an unprecedented institutional framework made up of a set of legal competences in which state law is only one of the multiple public and private actors participating in the regulatory process is currently emerging. In many fields of law, the traditional domestic actors in the law-making process (e.g., policy-makers, politicians, academics, the judicial system) reciprocally interact and eventually compete with an increasing number of actors outside national jurisdiction, creating a complex transnational space where decision-making, political, and legal processes take place. The point is not simply that some of the leading actors involved in regulation are found outside the borders of the nation state; it is rather that the burgeoning of sites in which actors and institutions produce and perceive normativity has broken the monopoly of the nation state over law and policy-making for its citizens. Along with national actors, transnational, supranational, and non-state actors such as NGOs and social movements are strongly interlinked and increasingly contribute to shaping the production of norms and regulations affecting the everyday life of many people within and beyond national borders. The legal outcome of this new ‘institutional alchemy’ (Caruso 2005, p. 3), is far from a coherent body of law doctrine, despite recent theoretical efforts to establish a ‘uniform concept of law’ (Günther 2008). It is a fragmented, displaced, and often contested assemblage of regulations, the producers and beneficiaries of which are closely interlinked, as well as mutually conditioned (e.g., Ong and Collier 2005). Legal pluralism, a notion referring to the coexistence of multiple overlapping norms and legal regimes, which only some 20 years ago was conventionally considered specific to postcolonial settings (see Fuller 1994), is at present widely accepted as the ordinary state of affairs of any socio-legal context in our ‘world society’ (Teubner 1997; Moore 2005). Admittedly, it is difficult to establish whether the globalization of law is an entirely new phenomena, a culturally constructed social fact, or a simple rhetorical effect produced by academic as well as popular discourse, while ‘the life of the law’ (Nader 2005), like any social process, continues to flow and change everywhere. I tend to agree with sociologist Saskia Sassen that yes, the social, political, economic, and legal changes many people are experiencing and witnessing correspond to an ‘epochal transformation’, another phase of capitalism, one mostly based on

Anthropology and soft law  75 neoliberal principles and practices (Sassen 2006; Harvey 2007; Mattei and Nader 2008). New legal configurations emerging under global or possibly more aptly ‘supranational capitalism’ (Westbrook 2004) deserve close inspection by scholars from different disciplinary affiliations using multiple methods and, perhaps, creating new terminology. Francis Snyder’s ‘global legal pluralism’ (2005; see also Berman 2012) seems an appropriate formula to tentatively single out and summarize the current transformations of law and legality. Nevertheless, the point at stake here is not so much to provide a comprehensive description of the globalization of law (see, e.g., Ferrarese 2000; Twining 2000; Likosky 2002; Cassese 2006; Merry 2006b; Halliday and Osinsky 2006; Mattei and Nader 2008; Benda-Beckmann et al. 2009; Berman 2012) as to attract attention to the existing correlation between soft law instruments and mechanisms and the legal transformations occurring at the global level. Interestingly, according to Flood (2002, p. 116, emphasis in the original), ‘global law is soft law made by professionals during the construction of legal deals. Without this softness, global law could hardly come into being: it would be brittle and fragile, under constant disputation. Yet soft law has been remarkably successful in constituting the global legal order’. Similarly, Teubner (1997, p. 16) considers softness as a ‘typical characteristic of global law,’ while in the critical opinion of Klabbers (2007, p. 326), ‘soft law, in the end, is but one emanation of a broader trend: it is the handmaiden of the increasing deformalization of global politics’. There is a paradox here that should not go unnoticed. Although the strength of (state) law is traditionally tied to its power to compel (by way of formal sanctions), the strength of global (transnational) law would reside in its ability to adapt to different circumstances, that is, to its malleability and softness. In this perspective, it is practical to refer to soft law as a specific technology of (global) law production, not in the sense of its capacity to become ‘hard law’ in the future, as a number of law scholars would argue, but rather as a specific way to enforce compliance using different, usually self-regulatory means. This is consistent with the dominant logic of neoliberal governance as described by anthropologist Cris Shore (2008, p. 284, emphasis in the original), discussing the ‘culture of audit’: ‘a key characteristic of neoliberal governance is that it relies on more indirect forms of intervention and control. In particular, it seeks to act on and through the agency, interests, desires and motivations of individuals, encouraging them to see themselves as active subjects responsible for improving their own conduct’. Hence a detailed empirical analysis of specific soft law instruments and mechanisms seems important in order to understand how global law originates and takes shape. More specific, when, where, and how global norms are created, internalized, transmitted, negotiated, and possibly enforced (see Halliday and Osinsky 2006). In other words, in order to grasp how global (legal) governance is locally experienced and exercised under neoliberalism, in a space that is neither domestic nor international (Suh 2002). Fundamentally, I posit soft law as a legal, hence a ‘political technology’ consistent with the current phase of global capitalism and related changes in law and governance as roughly outlined above. If, as I argue, a firm relationship exists between the two, it would be realistic to admit that soft law reflects and at the same time participates in the production of two major trends in the globalization of law (Di Robilant 2006, p. 500). First in the increasing proliferation of law-making procedures and sites and, second, in the privatization of legal regimes, a dual, parallel process that in different terms might be broadly conceived as the retreat of the state from law-making in many policy areas. From this perspective, soft law instruments are called upon to play an important role in the current process of denationalization occurring at the global level (Sassen 2006) and simultaneously within the margin of the nation state,

76  Research handbook on soft law the latter being an organizing principle far from having lost its authority (see Ferguson and Gupta 2002, pp. 989–990 especially; Di Robilant 2006, p. 523; Sharma and Gupta 2006, p. 5 et passim). In fact, as legal anthropologists Benda-Beckmann, Benda-Beckmann, and Eckert have noted, ‘while state sovereignty is increasingly challenged by the international legal system, at the same time state governments assume controlling and surveillance powers, imposing restrictions on the rights of citizens or peoples that are unprecedented in recent legal history’ (2009, p. 7).5 From a theoretical perspective, soft law concepts and practices seem to be a vantage point from which to understand and learn how state and non-state law procedures and products are currently reassembled in unpredicted and novel ways, overcoming scholarly rhetoric, which not infrequently still considers the ‘national’, the ‘global’, and the ‘local’ as mutually exclusive and hierarchically spatialized realms. Of course, this is not to deny a hierarchical organization among different sites of norm production but is, along with Ferguson and Gupta (2002), to invite scholars to engage with images and discourses of state verticality and encompassment as naturalized hierarchical relations between the state (‘the above’) and civil society (‘the grassroots’). As they (2002, p. 983) suggest in a thought-provoking article, ‘the point is not to denounce a false ideology, but to draw attention to the social and imaginative processes through which state verticality is made effective and authoritative’. In this framework, exploring new means of governance such as soft law might be crucial in order to understand how state and non-state actors and practices are currently reorganized. Today, as Moore (2001, p. 107) cogently pointed out, ‘it is clear that much of the debate that surrounds legal pluralism is not just an argument about words, but is often a debate about the state of the state today, one that asks where power actually resides’. Therefore, an ethnographic understanding of soft law could help us to single out changing styles of governance and shed light on emerging forms of global legal arrangements challenging the conventional understanding of law and legality (including extralegality) on the traditional temporal and spatial scale of the state.6

3.

THE FORCE OF SOFT LAW: BEYOND ADVOCATES AND CRITICS

Concerning the reception of soft law among contemporary scholars, Di Robilant (2006, pp. 504–511) has provided a broad picture of the extremely diverse positions represented within the international legal community, ranging from soft law ‘enthusiasts’ to soft law ‘radical critics’. Although it is beyond the scope of this chapter to offer an extensive review of this growing body of scholarly writings, I offer few examples that are emblematic of the two opposite camps. On the one hand, soft law enthusiasts consider that ‘the soft law approach offers many advantages: timely action when governments are stalemated; bottom-up initiatives that bring additional legitimacy, expertise, and other resources for making and enforcing norms and standards; and an effective means for direct civil society participation in global governance’ (Kirton and Trebilcock 2004, p. 5). On the other hand, radical critics argue that ‘soft law is soft with aggressive and opportunistic market actors, which under the shield of soft legality succeed in transferring costs to society, and hard for weak actors’ (Mattei 2003, cited in Di Robilant 2006, p. 508). Hence, for enthusiasts soft law is likely to ‘produce more effective global governance promoting economic openness, environmental enhancement, and

Anthropology and soft law  77 social cohesion’ (Kirton and Trebilcock 2004, p. 7), but to critics it is ‘yet another pattern of reception of American categories poorly fitting the fabric of European law’ (Mattei 2003, cited in Di Robilant 2006, p. 509). Admittedly, I am sympathetic toward the critical camp, whose stance is alerting us to the threat of the privatization of the legal regimes that soft law instruments entail. However, although one might even agree with liquidating soft law (like neoliberalism) as another ‘creative destruction’ (Harvey 2007; see also Gledhill 2004), soft law practices and discourses may possibly represent ‘sites of opportunities’ (e.g., Lorenzo 2010; Sandvik 2011; Merry 2015), where powerless or discriminated individuals and collective actors can rearticulate power relations, enhancing them within specific political and social contexts. That is why I suggest confronting soft law, accepting the challenge to empirically deal with its discourses and ideologies in social practice before emitting a final verdict with regard to its presumed merits or pitfalls. Thus I agree with a number of legal scholars (e.g., Abbott and Snidal 2000; Trubek et al. 2005; Di Robilant 2006, p. 554) when they encourage taking a more ‘pragmatic’ approach toward the heterogeneous array of non-binding instruments. Interestingly, a similar point has been made by anthropologist Anna Tsing vis-à-vis globalization: in order to overcome ‘a misleading portrait of a single global future’, we should be able to look at globalization as ‘a set of projects that require us to imagine space and time in particular ways’ (Tsing 2002, p. 476). In the same way, this chapter is an invitation to go beyond the controversy on soft law versus hard law, also encouraging an understanding of soft law projects and ventures as concrete manifestations of multifaceted, overlapping, and often contradictory circumstantial arrangements of specific legal articulations in/of given social and political settings, beyond the local/global divide (e.g., Maurer 2009; Merry 2015). An undertaking of this kind will not remain unchallenged by a number of assumptions explicitly and more often implicitly posited by mainstream legal reasoning. Reviewing the soft law literature produced by international law practitioners, it is possible to detect two major concerns that deserve critical consideration. They both refer to the ‘means to ends relation’ (Riles 2004, 2005) and are closely interconnected. I distinguish them here for analytical purposes. The first concern is a theoretical one and raises the question: is soft law really a law, or is it rather a non-law? In other words, if according to many legal scholars and law practitioners, law is binding – has the power to attain compliance by way of formal sanctions – how can an instrument of social pressure, influence, or persuasion be considered ‘genuine’ law? Although from a strictly legal point of view, the objection might be consistent, notably for a project that looks for legal concepts of uniform if not universal validity, to many others and especially to anthropologists, it probably sounds surprising, not to say idle. To those who raise doubts concerning the normative value of soft law, the anthropologist is tempted to reply with the words leading international lawyer Richard Falk used in discussing international law, notably discerning ‘the jurisprudential insult that arises when influential persons continue to raise the question as to whether international law is really law at all, or in its more moderate form, is a species of primitive law that deserves only qualified respect’ (2002, p. 355). Obviously, as most scholars would agree, the hard and soft logic of regulations are by no means mutually exclusive and are better seen as tools provided with a different degree of normativity along a continuum, rather than in the binary logic distinctive of legal positivism (see, e.g., Goldmann 2008). Moreover, as a number of studies within the field of legal anthropology have shown, despite a conventional call for neutrality and its universal claims, law (soft or hard) and justice are produced and operate in a social context (see, among many others,

78  Research handbook on soft law Moore 2005; Dembour and Kelly 2007). The same scholarly tradition has greatly contributed to clarifying that if we want to understand or map out how legality works in social settings, we need a more open definition of what law is, including the one conventionally assumed in Euro-American legal thought. International lawyers implicated in the soft law versus hard law debate, ultimately questioning the normative status of the former, seem rather reluctant to assume such socio-legal premises, nowadays taken for granted even by many legal scholars. The question of whether soft law is authentic law should ultimately encourage lawyers and non-lawyers to go beyond a strictly legal (positivist) conception of law, still influential among the representatives of the legal community of scholars, notably within the group Riles (2005) calls the ‘instrumentalist’ law scholars (as opposed to the ‘culturalists’).7 A second and related concern, recurrent within the ‘instrumentalist’ group of scholars, is of a practical nature and asks: is soft law effective? In other words, despite its non-legal binding character, is soft law able to produce norms and regulations of real effectiveness? Once again, this is a perfectly legitimate and certainly important question from a legal and most of all political and policy-making perspective. For culturalist scholars, I imagine the question of efficacy is also relevant, but in a rather different way: it is an invitation to scrutinize the very idea of (soft) law solely as a means to an end. From an anthropological perspective, soft law should not be seen simply as a tool to reach specific stated ends; it should rather be explored as an object, an apparatus, in Foucauldian terminology un dispositif, namely a political technology which creates, enhances, maintains, perverts, and modifies the exercise of power within a given social body. Hence, it is relatively unimportant to assess to what extent soft law is able to meet its deliberate goals. Once its machinery is set in motion, we could hardly say that it goes ahead without producing any social or political effect. However, from the same perspective, it is not important to ask whether the tangible outcome of any soft policy corresponds to the original intentions designated by lawyers and policymakers. It is instead necessary to scrutinize and question the presumed neutrality and apolitical character of such an instrument. To put it bluntly, as James Ferguson (1994, p. 17) does while discussing ‘development’: ‘the anthropologist cannot take planning at its word’. Moreover, beyond the consequences as an unintended product of any regulative means, the question of soft law effectiveness might also be critically explored by looking at the instrument in its own right (see also Katcherian 2010). To better bring this into focus, Riles (2004, p. 790) is again helpful; when drawing on Agamben, Dewey, and Heidegger, she argues that ‘political analysis and critique must begin from within the epistemological boundaries of the instrument – that solutions must be found in the means of technology’.

4.

TOWARD AN ANTHROPOLOGY OF SOFT LAW

Whereas a growing number of legal scholars welcome soft law instruments as the appropriate way to boost global governance, soft law critics raise doubts regarding their real efficacy and/ or draw attention to their ideological assumptions and implications. However, beyond optimistic or sceptical views and reactions, what is surprising, from an anthropological standpoint, is that advocates and soft law critics both often fail to appreciate that, in spite of its apparently technical and/or ideological dimension, soft law – despite its international, transnational, or global range – is always the product of concrete social and historical arrangements taking place by means of individual, collective, and institutional practices in specific locales and

Anthropology and soft law  79 temporalities. Unfortunately, how such arrangements are produced in the course of ‘the immediate, direct, vivid impression of the lives of Peter, Paul, and John, of single, real individuals’ (Antonio Gramsci, cited in Crehan 2002, p. 165) is seldom carefully considered, and consequently how soft law works in concrete social locations is often left vague and opaque. This is why an empirical consideration and ethnographic understanding of soft law also seems a challenging and timely undertaking. Ultimately, despite the multiple ramifications regarding its development and the controversial opinions concerning its elusive normative status and legal character, soft law projects and practices are legal (hence social and cultural) artefacts stimulating thinking about law and legality, including ‘extralegality’ (Smart and Zerilli 2014), and their changing mutual relationships within the current legal transformation occurring at a global level. A number of field research and publications encourage thinking about this particular technology of law production: questioning its apparatuses, working out its actual mechanisms and understanding its workings and logic by means of specific soft law instruments and projects in diverse policy area and context, such as the politics of ‘culture’ within the European Union (Katcherian 2010); the protection of minority language, identity, and culture in post-conflict Northern Ireland (Craig 2010); the introduction of affirmative action for Afro-descendants in the Brazilian workplace (Lorenzo 2010); the refugee’s management at the UNHCR in Kampala, Uganda (Sandvik 2011); and the role of quantitative indicators in translating justice and human rights soft law standards into enforceable obligations and mode of measurement within international and governmental institutions (Merry 2015), to which I now briefly turn. Examining the Civil society platform for intercultural dialogue, an initiative to foster European cultural policy recommendations under the Open Method of Coordination umbrella, Jeff Katcherian (2010) explores how European bureaucrats and civil society members address the tensions generated by preserving the principle of sovereignty and the parallel aspiration to reach suitable European integration with specific reference to the notion of ‘culture’. Katcherian’s work illustrates how, under the pressure of the crisis of the European ideal (best exemplified by a failure on the part of France and the Netherlands to ratify the European Constitution in 2005), the European Union has shifted from promoting cultural programs to coordinating a plan aimed at creating a truly European cultural policy by way of soft law instruments. Interestingly, as Katcherian points out, here soft law should not be seen as a means employed to attain a more stringent degree of normativity in the future, as often suggested by the formula ‘from programs to prescriptions’, recurrent in international law. As Katcherian’s multi-sited ethnography of soft law practices shows, the recourse to soft normativity is not instrumental but an end in itself, notably in the context of a European cultural policy in which openness, negotiation, and (intercultural) dialogue are values to be affirmed and promoted for what they are. Thus, he argues that the soft law ideology pervading the Platform needs to be explored on its own terms, dismissing the rhetoric of effectiveness and suggesting instead to seriously adopt its affective dimension, as a constitutive part of the process of culturally constructing Europe. This is a challenge Katcherian faces, examining softness and the process of softening culture as an aesthetic, as an object provided with its own internal logic, to be comprehended in its ‘form’ rather than its content, as Riles’ ethnographic analysis of the network has also argued (see Riles 2000). Furthermore, dealing with the controversial production of written documents during the Platform’s meetings, Katcherian shows that the enacting of soft regulations became not only a way to foster intercultural dialogue but an opportunity to rethink policy-making, beyond the conventional logic of the nation state.

80  Research handbook on soft law The interplay between state practices and supranational legal arrangements is at the core of another research paper in which legal scholar Elizabeth Craig (2010) explores a process in the making concerning some highly controversial issues, such as the protection of language, culture, and identity under the ‘national minority’ umbrella. Craig typically asks if and how non-legally binding instruments such as international human rights conventions might be translated and enforced in domestic law. Relying on both soft law scholarly literature and recent contributions in the anthropology of human rights, Craig’s research develops a twofold argument with reference to specific soft law instruments and mechanisms. At a general comparative level, she considers the development of different international minority rights instruments over the last two decades and measures their usefulness when confronted with specific norms to be enforced at the domestic level. Observing the elusiveness of some provisions contained in these instruments and noting their failure to clearly identify what either a ‘minority’ or a ‘national minority’ actually is, Craig raises doubts about softness as an appropriate means of enforcing cultural and minority rights. Craig tackles issue relying on her own experience as a legal adviser to the Culture, Identity, and Language Working Group of the Northern Ireland Bill of Rights Forum, a project favoured by the changing political circumstances established by the agreement signed in 2006 by the British and Irish governments and the main political parties. Specifically, her article highlights the problems involved in the incorporation into state law of the Framework Convention for the Protection of National Minorities, a ‘legal soft law’ instrument developed by the Council of Europe in 1995. Describing how the representatives of the main actors involved in the process (political parties, churches, trade unions, NGOs) meet to discuss the issues at stake, Craig shows how law technicalities intersect with the political intricacies of a divided society like Northern Ireland. She then clarifies the difficulties in translating into national legislation specific rights, such as the right to self-identification or receiving instruction in minority languages. Her article also reveals paradoxes and ambiguities concerning the use of several soft law instruments. In fact, from a strictly legal perspective the Bill of Rights drafting process could hardly be considered a successful event (on many issues no substantial agreement was found), but the debates that have taken place at the domestic level might be considered crucial in monitoring and advancing the minority rights agenda in Northern Ireland. Craig concludes that if the Framework Convention has shown its limitations in the drafting process, these are not necessarily due to its internal weaknesses; we should rather discern how such an instrument was used and manipulated in social process by key players, including legal academics and transnational actors. In a significant contribution, Rocío Lorenzo (2010) offers yet another good example of how soft technologies operate in mundane social practices and contexts, specifically focusing on a network of private companies and non-profit organizations in São Paulo that have adopted affirmative action policies to struggle against racial discrimination, a widely debated and controversial issue in Brazilian society. From a theoretical perspective, Lorenzo’s work provides a challenging attempt to complement Moore’s notion of a semi-autonomous social field (Moore 1973) with Bourdieu’s celebrated theorization of the ‘juridical field’ (Bourdieu 1986) in order to disentangle the ‘transnational social’, a space where non-conventional policy-makers (e.g., experts, consultants, managers, and activists) participate in the production of norms, using voluntary means of regulation. Interestingly, while Katcherian and Craig’s works both deal with the creation or adoption of soft regulations primarily by (supranational and/or national) public actors, Lorenzo’s research deals with the introduction of affirmative

Anthropology and soft law  81 action in the private sector, that is ‘outside’ the national legal system. To capture such ‘beyond the law’ or private regime of regulation, she explores affirmative action in the framework of a vast array of non-binding instruments such as corporate social responsibility, managing diversity, sustainability, and other human rights-oriented, typically voluntary forms of regulation. Actually, the adoption of self-regulatory means in the Brazilian workplace proves to be only relatively effective in assuring preferential treatment for Afro-Brazilians, or the ‘black’ population, as specific target actors. As many of Lorenzo’s informants imply, the adoption of such policies responds better to new market imperatives such as business ethics and corporate accountability, rather than to the actual implementation of rights for the targeted population. This hypocritical character of corporate social responsibility has led many critics to argue that binding governmental and international norms, rather than voluntary measures, seem necessary to ensure that companies behave in a socially responsible manner. However, according to Lorenzo’s work, the emerging regime of global regulations influencing both public and private actors also represents an opportunity for disadvantaged groups such as those targeted in the Brazilian affirmative action programs. In a thought-provoking article titled Blurring boundaries, Kristin Sandvik (2011, p. 12) examines ‘how refugee resettlement is managed through plural and overlapping norms and rationalities that emanate from transnational soft law’. Drawing on long-term ethnographic research as well as on her work experience with refugees and humanitarian workers at the UNHCR office in Kampala, Uganda, Sandvik explores a number of formal, informal, and illegal transactions that shape and govern the distribution of third-country resettlement slots. In the examined context, resettlement represents the only legally recognized way for migration to the West, and Sandvik convincingly argues that the struggle over the distribution of resettlement slots dominates the social relationships between the refugees and the legal protection officers. At the same time, her research also reveals important insights into the actual functioning of the UNHCR office and its bureaucratic culture, in close connection with the urban refugee population in Kampala. Using the formal, the informal, and the illegal as analytical frames, she explicitly refers to Moore’s (1973) concept of semi-autonomous social field: ‘although structured to some extent by national and transnational legal orders, these fields have the capacity to generate their own (non-legal) obligatory norms, with which compliance can be induced or coerced’ (Sandvik 2011, p. 12). Interestingly, transnational soft law tools and strategies adopted in name of a global reformist agenda actually concur in making formality, informality and illegality constitutive on the ground. Intended to unify and standardize refugee management, soft law further complicates the relationship between protection officers and urban refugees, perpetuating exploitative exchanges of sex, money, and power for resettlement. Instead of promoting regularization, transnational soft law norms actually contribute to creating a context of legal pluralism where law and its violation are reciprocally constitutive. Sandvik concludes that this is nevertheless ‘intrinsic to soft law governance, as the international community continues to channel resources in an effort to improve local implementation practices and to rectify the deficiencies of international institutions such as UNHCR’ (ibid., p. 26). In her brilliant research paper Firming Up Soft Law, Sally Merry (2015) explores how human rights soft law tools and mechanisms might serve to address a major problem of human rights legal order, a field famously lacking clear sanctioning power, in which a wide range of resolutions, declarations, guidelines and codes of practice complement the relevant treaties and conventions. Merry’s discussion relies on two interesting case studies. The first examines

82  Research handbook on soft law a project of the UN Office of the High Commissioner of Human Rights to develop indicators for human rights treaties, while the second considers an initiative of the US government to use indicators of compliance with a 2000 law against trafficking in persons (Merry 2015, p. 373). Drawing on her experience and observations in these two projects, Merry underlines how international organizations tend to establish quantitative indicators in order to measure human rights performances. In both cases this is possible mainly thanks to a work of ‘translation’: the language of justice and rights are translated into the language and thinking of economics and management. Hence, norms are translated into numbers and rankings. From this perspective quantitative indicators may be apparently useful in hardening soft law in transnational legal orders, clarifying and specifying various forms of obligations and enforceability. At the same time, human rights no longer appear part of a legal discourse with a broad and flexible vision of what justice and rights are, but they participate in a technocratic project consistent with economic thinking. Interestingly, as Merry observes, the practice of measuring performances has also an impact on the way human rights are understood as much as on their role in promoting global justice. If on the one hand performance indicators might serve to harden human rights soft law promoting its institutionalization beyond national borders, on the other hand ‘measurements produce readily comparable measures that permit ranking and shaming but at the cost of a simplification and decontextualization that runs counter to human rights legal thinking’ (Merry 2015, p. 375). As Merry herself concludes, ‘the turn to indicators is a shift from a justice-focused set of norms to a more technical, rational one that crosses more readily into non-legal fields, such as development planning or policymaking’ (ibid., p. 376). A paradox here should not go unnoticed: ‘measuring hardens soft law by increasing its enforceability but in doing so reduces its power to articulate a broad vision of a just society, a fundamental feature of its value as a political resource for social movements’ (ibid., p. 376). Albeit different, all research work reviewed here rely on a direct observation of soft law practices in a variety of social, political, and legal contexts. They all raise issues and ask questions from inside the soft law projects they have researched and lived, sharing experiences and conversations with both the producers and beneficiaries of different non-legally binding means of coercion. However, while Katcherian’s, Lorenzo’s and Sandvik’s contributions reflect their acknowledged methodological commitment to ethnographic methods, Craig’s and Merry’s analysis develops from direct participation in the soft law ventures they explore in their capacity of ‘legal expert.’ Intriguingly, at a time when anthropologists are increasingly interested in how the practice of fieldwork might be refunctionalized by means of doing ethnography among and notably in collaboration with ‘expert subjects’ (e.g., Holmes and Marcus 2005; Boyer and Marcus 2021), it is somehow ironic to learn that the legal experts themselves are also refining their professional agenda via ethnographic sensitivity and anthropological lenses. A reversal worthy of note, which undoubtedly deserves closer inspection from both legal and anthropological perspectives.

ACKNOWLEDGMENTS This chapter is a revised version of the introductory article to Non-binding coercions. Ethnographic perspectives on Soft Law, a theme section appeared in Focaal. Journal of Global and Historical Anthropology (vol. 56, 2010), published by Berghahn Books. I am grateful to both the publisher and the journal which have graciously granted permission for

Anthropology and soft law  83 publication. I would also like to thank Ellen Hertz for her comments on an earlier draft. Usual disclaimers apply. In memory of Sally E. Merry (1944–2020) and Sally F. Moore (1924–2021).

NOTES 1. This is a general – common sense – definition of law based on the binary opposition binding/ non-binding (i.e., legal/non-legal). One that goes beyond the differences between civil law and common law legal traditions, which according to many authors (e.g., Ferrarese 2000) are currently being eclipsed under the emerging ‘global legal canon’. 2. According to Galgano (2001), economic globalization represents regeneration of the medieval lex mercatoria, apparently fulfilling the old dream of a world merchant’s republic, the lords of which would be the members of the international business community. For an assessment of lex mercatoria as soft law, see also Teubner (1997). 3. The connection between soft law and Joseph Nye’s notion of ‘soft power’, introduced in international relations in the early 1990s, might be fruitfully explored, notably considering soft power’s allegedly non-coercive nature (namely its capacity to reach desirable outcomes without employing force). 4. Despite its conventional and widely accepted political meaning (i.e., government by force), the verb ‘to coerce’ (from Latin coercere) etymologically relies on notions of enclosure, confinement, warding off, namely ways of establishing boundaries between the inside and outside (i.e., relationships of inclusion and exclusion). 5. Moreover, in several circumstances, states increasingly delegate these controlling and surveillance powers to non-state actors such as security guards, private associations, and even criminal organizations (see, among others, Penglase 2009). 6. On the specific issue of law spatialization from a legal-anthropological standpoint, see Benda-Beckmann, Benda-Beckmann, and Griffiths 2009. 7. According to Riles (2005, p. 973): ‘the legal academy currently consists of roughly two groups [...] the Culturalists and the Instrumentalists. [...] The culturalists generally treat law as the embodiment of norms, the outcome of political compromise, and the repository of social meanings [...] The instrumentalists, in contrast, view law in primarily pragmatic instrumental terms, as a tool to be judged by its successes or failures in achieving stated ends. For them [...] law is a means to an end [...] and it should be evaluated according to its usefulness in solving actual legal problems.’ Interestingly, the same article urges ‘culturalist’ law scholars (including anthropologists) to take ‘law technicalities’ seriously (ibid., p. 975), obviously including the means-to-end relationship and its reversals (Riles 2004, p. 789–790). Albeit limited, this chapter is also an effort in this direction, going beyond interdisciplinary rhetoric.

REFERENCES Abbott, K. W. and Snidal D., (2000). Hard and soft law in international governance. International Organization. 54, 421–456. Benda-Beckmann, F. von, Benda-Beckmann K. von and Eckert J., (2009). Rules of law and laws of ruling: Law and governance between past and future. In: F. von Benda-Beckmann, K. von Benda-Beckmann, and J. Eckert, eds. Rules of law and laws of ruling. On the governance of law. Aldershot, UK: Ashgate. pp 1–30. Benda-Beckmann, F. von, Benda-Beckmann, K. von, and Griffiths, A., eds., (2009). Spatializing law. An anthropological geography of law in society. Aldershot, UK: Ashgate. Berman, P. S. (2012). Global legal pluralism: A jurisprudence of law beyond borders. Cambridge: Cambridge University Press.

84  Research handbook on soft law Bobbio, L., (2005). Governance multilivello e democrazia. Rivista delle politiche sociali. 2, 51–62. Bourdieu, P., (1986). La force du droit. Éléments pour une sociologie du champ juridique. Actes de la recherche en sciences sociales. 64, 3–19. Boyer, D. and Marcus, G. E., eds., (2021). Collaborative anthropology today: A collection of exceptions. New York: Cornell University Press. Caruso, S. B., (2005). Il diritto del lavoro tra hard law e soft law: nuove funzioni e nuove tecniche normative. Working Papers Centro Studi di Diritto del Lavoro Europeo ‘Massimo D’Antona’. 39, 1–19. Cassese, S., (2006). The globalization of law. Journal of International Law and Politics 37: 973–993. Craig, E., (2010). From soft to hard law? Culture, identity, and language issues within the Northern Ireland Bill of Rights process. Focaal. Journal of Global and Historical Anthropology. 56, 35–48. Crehan, K., (2002). Gramsci, culture and anthropology. London: Pluto Press. Dembour, M.-B. and Kelly, T., eds., (2007). Paths to international justice: Social and legal perspectives. Cambridge: Cambridge University Press. Di Robilant, A., (2006). Genealogies of soft law. American Journal of Comparative Law. 54, 499–554. Edelman, M. and Haugerud, A., eds., (2005). The anthropology of development and globalization: From classical political economy to contemporary neoliberalism. Oxford: Blackwell. Falk, R., (2002). Re-framing the legal agenda of world order in the course of a turbulent century. In: M. Likosky, ed. Transnational legal processes. Globalisation and power disparities. Cambridge: Cambridge University Press. pp. 355–377. Ferguson, J., (1994). The anti-politics machine. ‘Development,’ depoliticization and bureaucratic power in Lesotho. Minneapolis: University of Minnesota Press. Ferguson, J., (2004). Power topographies. In: D. Nugent and J. Vincent, eds. A companion to the anthropology of politics. Oxford: Blackwell. pp. 383–399. Ferguson, J. and Gupta, A., (2002). Spatialising states: Toward an ethnography of neoliberal governmentality. American Ethnologist. 29(4), 981–1002. Ferrarese, M. R., (2000). Le istituzioni della globalizzazione. Diritto e diritti nella società transnazionale. Bologna: Il Mulino. Flood, J., (2002). Capital markets, globalisation and global elites. In: M. Likosky, ed. Transnational legal processes. Globalisation and power disparities. Cambridge: Cambridge University Press, pp. 114–134. Fuller, C., (1994). Legal anthropology, legal pluralism and legal thought. Anthropology Today. 10(3), 9–12. Galgano, F., (2001). Lex Mercatoria. Bologna: Il Mulino. Gledhill, J., (2004). Neoliberalism. In: D. Nugent and J. Vincent, eds. A companion to the anthropology of politics. Malden (MA): Blackwell. pp. 23–48. Gluckman, M. (1963). Gossip and scandal. Current Anthropology. 4(3), 307–316. Goldmann, M., (2008). Inside relative normativity: From sources to standard instruments for the exercise of international public authority. German Law Journal. 9(11), 1866–1908. Günther, K., (2008). Legal pluralism or uniform concept of law? Globalisation as a problem of legal theory. NoFo. Journal of Extreme Legal Positivism. 5, 5–21. Halliday, T. C. and Osinsky, P., (2006). Globalization of law. Annual Review of Sociology 32, 447–70. Harvey, D., (2007). Neoliberalism as creative destruction. Annals of the American Academy of Political and Social Science. 610, 21–44. Held, D., ed., (1995). Democracy and the global order. From the modern state to cosmopolitan governance. Cambridge: Polity. Holmes, D. R. and Marcus, G. E., (2005). Culture of expertise and the management of globalization: Toward the re-functioning of ethnography. In: A. Ong and J. Collier, eds. Global assemblages. Technology, politics, and ethics as anthropological problems. Oxford: Blackwell. pp. 236–252. Katcherian, J., (2010). Softening culture, opening Europe. The European civil society platform for intercultural dialogue. Focaal. Journal of Global and Historical Anthropology. 56, 19–34. Kirton, J. J. and Trebilcock M. J., eds., (2004). Hard choices, soft law: Voluntary standards in global trade, environment and social governance. Aldershot, UK: Ashgate. Klabbers, J., (2007). Reflections on soft international law in a privatized world. Finnish yearbook of international law. 16, 313–328.

Anthropology and soft law  85 Likosky, M., ed., (2002). Transnational legal processes. Globalization and power disparities. Cambridge: Cambridge University Press. Lorenzo, R. A., (2010). Untangling the ‘transnational social’. Soft affirmative action, human rights, and corporate social responsibility in Brazil. Focaal. Journal of Global and Historical Anthropology. 56, 49–61. MacClancy, J., ed., (2002). Exotic no more. Anthropology on the front lines. Chicago: University of Chicago Press. Mattei, U. and Nader, L., (2008). Plunder. When the rule of law is illegal. Oxford: Blackwell. Maurer, B., (2009). From the revenue rule to soft law and back again: The consequences for ‘society’ of the social governance of international tax competition. In: F. von Benda-Beckmann, K. von Benda-Beckmann, and J. Eckert, eds. Rules of law and laws of ruling. Routledge, pp. 217–236. Merry, S. E., (2006a). Anthropology and international law. Annual Review of Anthropology. 35, 99–116. Merry, S. E., (2006b). Human rights and gender violence: Translating international law into local justice. Chicago: University of Chicago Press. Merry, S. E., (2007). Human rights law as a path to international justice: The case of the Woman’s Convention. In: M. Dembour and T. Kelly, eds. Paths to international justice. Cambridge: Cambridge University Press. pp. 161–185. Merry, S. E., (2015). Firming Up Soft Law. The impact of indicators on transnational human rights legal orders. In: T. C. Halliday and G. Shaffer, eds. Transnational Legal Orders. Cambridge: Cambridge University Press. pp. 374–399. Moore, S. F., (1973). Law and social change: The semi-autonomous social field as an appropriate subject of study, Law & Society Review. 7(4), 719–746. Moore, S. F., (2000). Law as process. An anthropological approach. 2nd ed. Münster-Hamburg and Oxford: LIT Verlag and James Currey. Moore, S. F., (2001). Certainties undone: Fifty turbulent years of legal anthropology, 1949–1999. Journal of the Royal Anthropological Institute. 7(1): 95–116. Moore, S. F. (2002). An international legal regime and the context of conditionality. In: M. Likosky, ed., Transnational legal processes. Globalisation and power disparities. Cambridge: Cambridge University Press. pp. 333–354. Moore, S. F., ed., (2005). Law and anthropology: A reader. Oxford: Blackwell. Nader, L., (1972). Up the anthropologist. Perspectives gained from studying up. In: Dell Hymes, ed. Reinventing Anthropology, New York: Random House. Nader, L. (2005). The life of the law. Anthropological projects. Berkeley: University of California Press. Ong, A. and Collier, S. J., eds., (2005). Global assemblages. Technology, politics, and ethics as anthropological problems. Oxford: Blackwell. Penglase, B., (2009). States of insecurity: Everyday emergencies, public secrets, and drug trafficker power in a Brazilian favela. PoLAR—Political and Legal Anthropology Review 32(1), 47–63. Radaelli, Claudio M., (2003). The Open Method of Coordination: A new governance architecture for the European Union? SIEPS Report 1. Stockholm: Swedish Institute for European Policy Studies. Riles, A., (2000). The network inside out. Ann Arbor: University of Michigan Press. Riles, A., (2004). Property as legal knowledge: Means and ends. Journal of the Royal Anthropological Institute. 10(4), 775–795. Riles, A. (2005). A new agenda for the cultural study of law: Taking on the technicalities. Buffalo Law Review. 53, 973–1033. Sandvik, B. K., (2011). Blurring boundaries: Refugee resettlement in Kampala between the formal, the informal, and the illegal. PoLAR—Political and Legal Anthropology Review. 34(1), 11–32. Sassen, S., (2006). Territory, authority, rights. Princeton: Princeton University Press. Sharma, A. and Gupta, A., (2006). Introduction: Rethinking theories of the state in an age of globalization. In: The anthropology of the state: A reader. Oxford: Blackwell. pp. 1–41. Shore, C., (2008). Audit culture and illiberal governance: Universities and the politics of accountability. Anthropological Theory. 8(3), 278–99. Smart, A. and Zerilli, F. M., (2014). Extralegality. In: D. M. Nonini, ed. A Companion to Urban Anthropology. Malden (MA), Wiley Blackwell. pp. 222–238. Snyder, F., (1993). The effectiveness of European community law: Institutions, processes, tools and techniques. Modern Law Review. 56(1), 19–54.

86  Research handbook on soft law Snyder, F., (2005). Governing economic globalization: Global legal pluralism and European Union law. In: S.F. Moore, ed. Law and anthropology: A reader. Oxford: Blackwell. pp. 313–329. Somma, A., ed., (2009). Soft law e hard law nelle società postmoderne. Torino: Giappichelli. Suh, D., (2002). Situating liberalism in transnational legal space. Duke Journal of Comparative International Law. 12, 605–630. Teubner, G., (1997). Global Bukowina: Legal pluralism in the world society. In: G. Teubner, ed. Global law without a state. Brookfield: Dartmouth. pp. 3–28. Trubek, D. M., Cottrell, P. and Nance M., (2005). ‘Soft Law,’ ‘Hard Law,’ and European integration: Toward a theory of hybridity. Jean Monnet Working Papers 2, Jean Monnet Chair. Trubek, D. M. and Trubek L. G., (2005). Hard and soft Law in the construction of social Europe: The role of the Open Method of Co-Ordination. European Law Journal. 11(3), 343–364. Tsing, A., (2002). The global situation. In: J. X. Inda, and R. Rosaldo, eds. The anthropology of globalization. A reader. Oxford: Blackwell. pp. 453–85. Twining, W., (2000). Globalisation and legal theory. Cambridge: Cambridge University Press. Westbrook, D., (2004). City of gold: An apology for global capitalism in a time of discontent. New York: Routledge.

6. Democracy and soft law Ulrika Mörth

1. INTRODUCTION Throughout this volume we have read how the development, history, and uses of soft law in legal and social sciences research are diverse and complex (see e.g., chapters by Snyder and Terpan, this volume). Many concepts in legal and social sciences serve multiple functions and have connotations that vary over time and across research traditions. One of the main functions and advantages of soft law is that it allows researchers and practitioners to go beyond a binary world of legitimate rules – laws or nonlaws. This flexibility could, however, potentially be problematic when we are dealing with a concept and phenomenon that has democratic implications. To whom, for what and how can actors be democratically accountable in a transnational world characterized by an ‘increasing scope and breadth of regulatory activities and governance activities of all kinds’ (Djelic and Sahlin-Andersson 2006, p. 1; Avant et al. 2010; Büthe and Mattli 2011; Holzer, this volume)? Rules and regulations are about power and attempts to influence, albeit by different actors (public or private) and in diverse domains (state, civil society, or business). The emergence of multilateralism, multistakerholderism and global governance ‘introduce a whole new set of governance actors and new processes for making global “laws and regulations”’ (Gleckman 2018, p. 1). These rules and regulations can be legitimate, but they often derive their main legitimacy from different sources – political, legal, economical, and democratic (Baldwin et al. 2011; Holzer, this volume). This chapter analyzes whether and, if so, how it matters from a democratic point of view how we conceptualize rules and regulations that go beyond the binary world of rules. I argue that we need to take a closer look at what we mean by democracy before we can try to answer this broad question. My focus lies on the theoretical arguments within different theories and ideals of democracy rather than on democratic performance and assessments thereof. To be sure, there is a vast literature on how political systems perform in relation to different democratic criteria and democratic ideals. Most of this literature and research concerns the nation state and the performance of basic democratic criteria, electoral voting for instance (Diamond and Plattner 2006), the determinants of democracy (Teorell 2010) and the emergence of elections without democracy (Diamond 2002, 2019). There is also a rich literature on whether and how the European Union (EU) performs in terms of different democratic ideals. This literature and research includes the debate on whether and, if so, how there is a democratic deficit in the EU (Hix 2008; Beetham and Lord 2014, see also Karlsson 2001 on how Dahl’s classic democratic criteria can be empirically assessed in the EU), studies on the balance between performance and accountability in the EU (Scharpf 1999; Mörth 2008) and whether and how deliberation and deliberative democracy are practised in the EU (Eriksen and Fossum 2000; Korkea-aho 2016). Assessments on whether and how democracy works at the global level is more limited. This literature often discusses theoretical blueprints and ideals (a literature that I also address in this chapter) with some 87

88  Research handbook on soft law empirical illustrations rather than conducting systematic evaluations (e.g., Bexell and Mörth 2010; Zelli et al. 2021). The chapter is organized in three sections. The specific question section 2 is whether and how soft law has any intrinsically democratic values or whether it is more about ‘the need for action’ in a political system (Elster 1988, p. 4)? This question is answered by introducing and discussing three classic interpretations of liberal representative democracy. Drawing from each tradition of democracy I attempt to deduce the major principles and theoretical arguments on the relationship between democracy and soft law. The following section discusses whether and how we should assess the democratic implications of soft law from the perspective of normative ideals of democracy or taking the actual democratic performance the regulations may potentially have into account (e.g., Börzel and Risse 2005)? It also deals with the possible trade-offs between democratic values and other criteria for good regulation (Dingwerth 2007; Bexell and Mörth 2010; Baldwin et al. 2011). Here I will discuss the performative aspects of democratic ideals and the most difficult question within democratic theoretical thinking, namely whether there is a case for democratic hybridity that can match transnational and regulatory hybridity (Snyder, this volume; Gustafsson and Tamm-Hallström, this volume)? This question derives from the fact that so much of the theoretical thinking of democracy is about a state-centric world. What answers, if any, do these theorists have on the practice of transnational regulation and governance?

2.

THREE INTERPRETATIONS OF DEMOCRACY

Popular sovereignty is the well-established meaning of democracy which has the aim to curb the arbitrary exercise of power and to hold decision-makers accountable (Goodin 2010). Different theories of democracy, however, emphasize various values within this basic idea of what democracy is. There exists a vast number of competing definitions and conceptualizations of democracy, which Sartori (1987) characterizes as the age of confused democracy (see also Held 2006). In this section of the chapter, I use the classic ideal types of democracy: liberal-aggregative, republican-communitarian and deliberative democracy, to outline three interpretations of liberal representative democracy as summarized in Table 6.1 (see below). The three interpretations are not mutually exclusive. They are all based on the liberal representative democratic system but with differences in emphasis on democratic norms/values, rationalities, types of political process/entity and the most important source of legitimacy for the political system. The three interpretations can therefore be characterized in terms of governance for, by and through the people, inspired by the famous ‘Gettysburg address’ by Abraham Lincoln in 1863.1 Governance for the people refers to a democratic system that does not require active participation of the voters between the elections. Governance by the people refers to a system where active participation of the voters between the elections is essential for the democratic system. Governance through the people refers to a democratic system that requires constant deliberation and public reasoning among the voters. I first introduce the three interpretations and then analyze the overall question whether or not soft law has any intrinsically democratic values, and if it does, how soft law has come to have them. It is through investigating the democratic ideals that we can try to identify the potential democratic risks or democratic opportunities with soft law. The very meaning,

Democracy and soft law  89 Table 6.1

Three interpretations of liberal representative democracy

Overall Interpretation Liberal-Aggregative (Governance for the people) Republican-Communitarian (Governance by the people)

Main Democratic

Dominant Conception

Main type of Political

Most Important Source

Norm/Value

of Rationality

Process/entity

of Legitimacy

Problem Solving

Efficiency

Accountability

Instrumental/ Strategic

Participation

Social Context

Value-based

Public reasoning

Communicative

Rights-based

Collective Self-Understanding

DeliberativeDiscursive (Governance through

Justice and Norms of Fairness

the people)

Sources: Habermas (1996); Eriksen and Fossum (2004).

conceptualization and perception of soft law might even differ between various democratic interpretations and ideals. My starting point is that soft law is about non-binding rules (Snyder, this volume). The concept of norms refers to democratic norms or to norms of the rule of law (depending on which democratic interpretation and model, see Table 6.1 above). The concept of values refers not only to democratic values, but also to cultural and social ones (depending on which democratic interpretation and model, see Table 6.1). The first interpretation is the liberal-aggregative model. This model emphasizes governance for the people. This democratic governance is primarily accomplished through the elections of political representatives in parliaments and other legislative bodies. The rationality is strategic and instrumental which means that professional expertise rather than laymen knowledge is important. The overall aim of the political process is problem solving. Consequently, the source of democratic legitimacy is efficiency, and the primary democratic value is accountability. The republican and communitarian view of democracy is the second interpretation. Democratic governance by the people occurs through general elections but is complemented by the importance of creating a social and a cultural national community and identity to achieve participatory democracy. The rationality is about following the social context and the social norms on which the political process is based. The significant democratic value and norm is participation by the people in the value-based community, for instance in recurring referendums. The most important source of legitimacy is that the political system reflects the collective self-understanding. The third interpretation of democracy is deliberative democracy with its emphasis on public reasoning and consultation, including the use of direct democracy or referendums (Fishkin 2009). In general, this model gives less weight to parliamentarian systems of governance than the other two traditions. It has therefore been labelled as a tradition that features government through the people and reflective democracy (Goodin 2003, 2005, 2008). The focus here is on preference-formation and collective political will-formation and not preference aggregation as in the first interpretation of democracy. Deliberation can broadly be defined as a process in which individuals in joint discussions sincerely deliberate competing arguments (Fishkin 2009). Deliberative democracy is often referred to as a discursive process because it is social and inter-subjective and involves communicative rationality based on the better argument.2 In contrast to the republican-communitarian interpretation of democracy, the deliberative inter-

90  Research handbook on soft law pretation emphasizes the concept of civic citizenship (rights-based political entity) and that the most important source of legitimacy are justice and norms of fairness. There are two main common assumptions in the three interpretations of democracy. The first assumption is that each democratic model specifies what actors are to be designated as legitimate participants in the making of ‘collective decisions for those decisions to qualify as democratic’ (Agné 2006, p. 433). In the first model of the liberal-aggregative democracy, it is first and foremost the voters. Here, political equality is conceptualized as the principle of ‘one person one vote.’ The decision-making consists of the aggregation of individual preferences by governors held accountable before the citizens in regular, free, and fair elections. An extreme position within this notion of preference-aggregation is that of Schumpeter and his minimalistic interpretation of democracy (Schumpeter 1952; see also Fishkin 2009). In this interpretation of democracy, voting is the only manifestation of democracy because people are not supposed to be interested in engaging in the political process or cognitively capable of engaging in complex societal and political matters (Schumpeter 1952). The citizens in the second model of the republican-communitarian model form a demos that takes part in direct democratic processes. The political process that needs to be democratically legitimate is thus considered to be much more comprehensive than voting in general elections only. The rationality or the behavioural logic among voters is not necessarily about strategic choices (as in elections) or accepting the better argument (as in public reasoning) but to follow and uphold the sense of collective self-understanding. In the third interpretation of liberal democracy (deliberative-discursive), the legislative decisions by the elected representatives will be preceded by extensive preference formation and the formation of the collective will among the voters. This is also a much more ambitious and comprehensive political process than counting votes in general elections. The electorate is here perceived as both interested and cognitively capable of being engaged in complex social and political matters. Habermas argues, for instance, that everyone must be treated with equal consideration and that the respect for the better arguments must prevail (Habermas 1996). The constitutional state must therefore be organised in a manner that secures a free debate among its citizens.3 The second common assumption in these three interpretations is the importance of upholding the congruence between legitimate power and accountability in order to qualify as democratic. This congruence is crucial in any liberal representative democratic system with its emphasis on constitutions, laws and the formal hierarchical order of command and control (Dryzek 2000). It is of particular importance in the liberal-aggregative model because a well-functioning delegation between the people and its representatives can be seen as crucial in a model that emphasizes the importance of accountability as its main democratic value. On the other hand, this value needs to be balanced with output legitimacy, that is, that the system performs. Drawing on Elster’s (1988) famous triangle it is argued that political systems need to strike a balance between rule of law, democracy and making things happen (March 1994). This balance is also relevant for the other two interpretations of democracy with the addition of the requirements of the possibility for participation in the second model and making sure that there is qualitative public deliberation among equals in the third model, respectively. So, what conclusions can we draw from the theories of democracy for the relationship between democracy and soft law? I draw four broad conclusions based on these three ideals of democracy. First, soft law might be democratically risky in the liberal aggregative model, but that it may also be seen as a democratic asset. The risk has to do with how this model

Democracy and soft law  91 emphasizes the importance of that ideal of accountability according to the hierarchical model of command and control. The democratic risks concern the uncertainty of who, for what and how accountability can be achieved if the rules are perceived as ambiguous and vague. This risk is high for any rule, soft or hard, but it might increase if the rules are perceived as legally non-binding, and especially if the regulators are not the traditional legislators. The congruence between power and accountability according to the chain of delegation is crucial in the liberal-aggregative model of democracy. The possibility or asset has to do with how this model focuses on solving problems and increasing output legitimacy. The democratic possibilities of soft law in that model have to do with whether and how soft law might be instrumental for solving problems in the political system. Here, soft law might not have any deeper democratic intrinsic values, but it can be seen as more as an instrument for democratic and political systems (e.g., Elster 1988). This instrument of soft law for making things happen in political systems has been an especially salient theme in the research on the EU (Senden 2004; Héritier and Rhodes 2010; Mörth 2015). Second, soft law in the republican-communitarian model could have democratically intrinsic values. This is so because the main democratic value and norm is participation in a social and cultural normative context (and not a strategic instrumental rationality) in a value-based political entity. The most important aspect of legitimacy for the political system is that it reflects and creates collective self-understanding based on cultural values and social norms. The question of who, for what and how accountability is achieved is of less importance. Democracy is more about other values, participation in particular. This participation takes place in a political community in which social and cultural norms are of great importance. Here, soft law can be part of this social and cultural self-understanding of how democracy works. If the source of the democratic legitimacy is challenged by new ways of perceiving the collective self-understandings, and the norms of the democratic system, this could lead to a situation in which the democratic status of non-legally binding rules needs to be re-negotiated. One empirical example of this renegotiation is how the EU is gradually pursuing a hard law position towards Poland and Hungary when they are perceived to break with the democratic norms of the EU, and the very collective self-understanding of how democracy works. Third, in deliberative democracy soft law can both be regarded as a democratic risk and a democratic possibility. How either of these scenarios plays out has to do with the kind of deliberative democracy we envision. In the most dominant theoretical thinking of deliberative democracy, the liberal constitutionalist version, envisioned by Habermas (Habermas 1998; Dryzek 2000), soft law is as problematic as discussed in relation to the liberal-aggregative model. The legality of the Habermasian democratic model – that democracy is unimaginable without a constitutional state – does not mean that soft law is, however, a priori problematic. It could be a democratic asset with democratically intrinsic values especially if one emphasizes the importance of a rights-based political entity and that justice and norms of fairness are crucial for democratic legitimacy. Here, soft law is about important values and norms in systems of rule of law and public legal-authority-making. The fourth and final conclusion concerns the relationship between democracy and the level of political system. The classic interpretations of representative democracy are mainly focused on either democracy within the nation state or whether and how the liberal and representative democratic institutions can be established beyond the nation state. Decades of strengthening democratic institutions in the EU, especially with respect to the European Parliament, suggest that liberal and representative democracy is possible beyond the nation state. The complexity

92  Research handbook on soft law of establishing legislative and parliamentary systems at the global level is often perceived as difficult and even impossible, as argued by one of the most prominent scholars of democracy (Dahl 1979, 1989). However, other scholars of liberal democracy find that electoral democracy can be achieved at the global level. Theories of cosmopolitan democracy are oriented towards global institution building, especially concerning how the United Nations (UN) can be democratically reformed (Held 1995; Archibugi et al. 2011). These theories are based on the liberal-aggregative understanding and are being criticized for either being too unrealistic or too traditional (Dryzek 2000, 2010, 2011). One major problem with the cosmopolitan model of democracy is that it would require global political parties or at least reforms to narrow the gap between the electorate and the elected politicians in the global institutions. Another critique that comes from the republican-communitarian thinking of democracy is the lack of a global demos (Dahl 1989). A possible ‘solution’ to these problems is to organize democracy beyond the nation-state by emphasizing the civic and rights-based citizenship (Habermas 1998, 2012). However, this version of deliberative democracy often resembles a liberal constitutionalist deliberative democracy (Habermas 1998; Dryzek 2000). In fact, the legality of the Habermasian democratic model – that democracy is unimaginable without a constitutional state – is akin to the idea of a global cosmopolitan democratic model (Held 2006). With the idea of a European constitutional patriotism (Habermas 1998), Habermas may have resolved the problem of having no global or European demos, as formulated by the republican-communitarian tradition. His thinking on democracy is highly state-centred, though, and depends on political institutions and a constitutional framework within or beyond the nation-state (Habermas, 1996). This means that deliberative democracy depends on the functioning of the traditional liberal and representative institutions and that votes count (Dryzek 2010). This leaves me with the question of whether there is a case for democratic legitimacy in a world characterized by private governance (Dingwerth 2007; Bexell and Mörth 2010; Cutler and Dietz 2017), market multilateralism (Bull and Neill 2007), an emerging global public domain (Drache 2001; Ruggie 2004; Helgesson and Mörth 2013), multistakeholderism (Gleckman 2018), and global regulatory rivalry (Zelli et al. 2021)? What answers do democratic theories have to this practice of complex transnational regulatory activities and the fight for attention, resources and authority (Djelic and Sahlin-Andersson 2006, p. 5; Holzer 2010, this volume). This question is discussed in the next section.

3.

DEMOCRACY BEYOND THE NATION STATE AND THE IMPLICATIONS FOR SOFT LAW

For-profit actors perform governmental tasks, and they do so in the transnational domain. Businesses and states are becoming more embedded in a broader global arena concerned with rules and regulations in various policy fields (Baldwin et al. 2011). The public is no longer equivalent to the state (Drache 2001; Mörth 2009). A new type of global domain emerges, cutting across the traditional public-private divide (Ruggie 2004). Transnational corporations are central actors in many areas of global governance (Jordana and Levi-Faur, 2004; Djelic and Sahlin-Andersson, 2006; Kobrin 2009; Holzer, this volume). The literature that normatively prescribes and explains the political duties and activities of business actors disputes the classic Friedman approach towards business: ‘that businesses should stick to the knitting and attend

Democracy and soft law  93 exclusively to the interests of their shareholders’ (Néron and Norman 2008, p. 1). This has led to new conceptualizations of private international authority (Cutler 2003; Hall and Biersteker 2002; Bexell and Mörth 2010; Helgesson and Mörth 2013). Holzer argues that there is a development towards corporate citizenship that is bound up with public expectations (Holzer 2010, 2013; see also Gerencser 2013). For these reasons it can be argued that the traditional institutional values and logics between the state, the market and the civil society (Thornton and Occasio 2008) are blurred and that the actors from these domains are turned into institutional hybrids (Pache and Santos 2013; Thomann et al. 2017; Mörth and Pierre 2021). However, this practice of blurred boundaries between actors in transnational governance and regulation has little support in normative democratic theories. Democratic theories and theorists are occupied with the relation between the state and civil society (Walzer 1991; Warren 2012). One of the important positions in the deliberative theory that has its roots in critical theory is its engagement with questions of how to emancipate individuals and society from oppressive forces, such as the market or the economy. Dryzek argues that liberal democracy fails to understand that constitutional structures ‘are not the only forces that order deliberation in the liberal polity’ (Dryzek 2000, p. 18). There exist material and ideological discourses that often emanate from market liberalism (Dryzek 2000). An illustration of this thinking of democracy is how Habermas separates between civil society, the lifeworld, on the one hand, and the two subsystems, market/economy and the state, on the other (Habermas 1996). The purpose of the lifeworld is ‘to establish a democratic barrier against the colonization of civil society by money and bureaucracy’ (Eriksen and Weigård, 2003, p.84). Indeed, for Habermas, private businesses belong to the economic system, a system that is to be kept apart from the lifeworld of public deliberation. The private in the public spheres comprises of ‘nongovernmental and non-economic connections and voluntary associations that anchor the communication structures of the public sphere in the society component of the life world’ (Habermas 1996, p. 366). In sum, Habermas has a clear view of who the legitimate participants are in public deliberations. He recognizes that business actors have political and economic influence but, and because of this influence, he excludes them from the democratic process. Businesses may be political actors and they may be engaged in societal deliberations, but they are not legitimate participants in public deliberations, neither in civil society nor in political institutions, from his perspective. Hence, business actors may take part in elite deliberation (Fishkin 2009), but this has little to do with discourse democracy in the Habermasian sense. Public deliberation is not necessarily the same thing as democratic public deliberation (Lee and Romano 2013). Democratic legitimacy requires, at a minimum, conditions of political equality and political bindingness (Erman 2012; Føllesdal 2011). So where does this leave us with respect to the question of democratic legitimacy in a world of transnational governance and regulation, and on the relationship between democracy and soft law in particular? I see two possible answers to this question. First, given the normative ideals of democracy, we need to assess possible trade-offs between democratic values and other criteria for good regulation (Dingwerth 2007; Mörth 2008, 2009; Baldwin et al. 2011). Second, the democratic assessment of transnational regulations could potentially be based on new ways of understanding democracy and democratic legitimacy. There is a rich literature regarding the first response to the dilemma of how to handle the discrepancy between regulatory practice and normative ideals. The trade-offs between input legitimacy and output legitimacy are well elaborated in the literature on global governance and regulation (for an

94  Research handbook on soft law overview, see Bexell and Mörth 2010; Zelli et al. 2021).4 For this reason, I elaborate on the second alternative. As discussed in the previous section, most models and interpretations of democracy, including the Habermasian model of deliberative democracy, are about ideals. They can therefore be considered as blueprints for democracy but impossible to realize fully in practice, as argued by MacDonald (2008). However, the history of democracy is a history of pragmatism and compromises (Goodin 2010). What are then the pragmatic normative interpretations of democracy? Dryzek argues that we first need to abandon the idea that democracy is equivalent to voting and the representation of persons or interests. Instead, we need to conceptualize democracy in terms of deliberation and communication within networked governance (Dryzek 2010). It is by breaking with electoral democracy and the constitutional state model of global democracy (or the cosmopolitan model) that Dryzek (2000, 2010) theorizes on transnational democracy. He goes on to state that the lack of institutional hardware – formal institutions and rules at the global level – and the resort to ‘institutional software’ – could rather turn out to be favourable for the democratic purposes at the transnational level (Dryzek 2000, p. 122). This interpretation of discursive democracy is connected to a strong conception of civil society, and to the idea of a public sphere and discourse representation (Dryzek 2010). Thus, like many other theorists he puts a lot of democratic faith in the transnational civil society that is expected to be fearless, and to be able to constitute and reconstitute the discourses in international politics (Dryzek 2000, p. 131). He poses business and transnational market forces as something against which civil society should react. In that analysis, he stresses the importance of existing practices such as ‘transnational social movement activism, self-appointed unelected popular representation, and deliberation within international negotiations such as the G20 or the World Economic Forum’ (Dryzek 2011, p. 216). Empowered spaces of transnational financial networks, and associated discourses on market liberalism, may be contested through the cultivation of alternative discourses in public space (Dryzek 2011). Drawing on this networked governance and the importance of transnational discourses, MacDonald (2008) argues for the idea of global stakeholder democracy. She shares the notion with Dryzek that democracy is possible at the transnational level even without a grand constitutional state, although MacDonald emphasizes the need for liberal democratic accountability mechanisms at the global level (MacDonald 2008). However, in the absence of electoral mechanisms of authorization and accountability, empowered public discourses, strong participation among those that are affected by certain decisions, and concrete measures of enhancing the transparency of what the political public agents are doing can function as important democratic mechanisms (MacDonald 2008; see also Uruena et al., this volume). For the relationship between democracy and soft law it means that soft law may have democratically intrinsic values. These values can be assessed through different democratic mechanisms that are not or are to a lesser extent based on electoral voting, political equality, political representation and the congruence between legitimate power and democratic accountability in the hierarchical democratic system of command and control. Whether this could work in practice is of course hard to tell. It might function as a complement to the traditional system of the liberal and representative democracy in situations when the electoral system of accountability is either flawed (Diamond 2002, 2019) or insufficient (e.g., Goodin 2010; MacDonald 2008; Dryzek 2010).

Democracy and soft law  95

4.

CONCLUSIONS AND FURTHER RESEARCH

The short answer to the complex question posed in the introduction of this chapter – whether it matters from a democratic point of view of how we conceptualize rules and regulations that go beyond the binary world of rules? – is that it does matter. Or to be more precise, what matters is how we conceptualize democracy and democratic legitimacy. It is through an analysis of different interpretations of democratic values, rationality, political process and overall aims for the political system, that we can identify the relationship between democracy and soft law. This relationship is sometimes unproblematic because soft law is instrumental for the democratic and political system to perform and achieve results. It is also a relationship that could strengthen the democratic and representative system when soft law reflects deeper social, cultural norms (the republican-communitarian model) and rights-based values (deliberative model) in the political system. The democratic risk with soft law is most salient when accountability is the predominant value, as in the liberal-aggregative interpretation of democracy, especially if the regulators are not the traditional legislators. Paradoxically, it is also in this type of political process that soft law is a political resource and can make things happen. In the analysis on whether there is a case for democratic hybridity that can match transnational and regulatory hybridity, the answer is that normative democratic theories are based on ideals rather than on the actual regulatory practice. The importance of for-profit actors or hybrid actors in transnational governance and regulation has very little resonance in democratic normative theories. These theories make a clear separation between private and public actors or between state, market and the civil society. My reading of the more pragmatic democratic theorists, such as Dryzek and MacDonald, suggests that decoupling democracy from the constitutional state and the chain of delegation in the liberal and representative system still requires some type of liberal and representative democratic accountability mechanisms. The attempts to include participants that are affected and that have stakes in the decision-making process are an addition to this democratic representative system. The main conclusion of this analysis of democratic theories and soft law is that the democratic status of non-legally binding rules depends on whether and how these rules are coupled to the liberal representative system of democracy. I suggest that further research on the relationship between democracy and soft law should focus more on the gap or decoupling between regulatory practices and liberal representative democracy. There are two ways of understanding this gap. The first has to do with the gap between the private transnational practice of regulation and governance, on the one hand, and the democratic ideals upon which the institutions rest, on the other. For-profit actors may be insiders in transnational regulation and governance, but they are outsiders in the democratic accountability system. The chapter shows that the extent to which research on democracy presents any solutions to this incongruence between transnational power and accountability is rather limited. The main problem according to the research on democracy is the importance of upholding the separation between different domains and actors. However, the challenge ahead is that political systems need to handle the hybrid regulatory practice with the requirements for democratic legitimacy. One dilemma on how to do that is to decide on what democratic requirements that are necessary. Should we emphasize the democratic values of accountability, active participation and deliberation or are we to evaluate the democratic effects of the performance that the regulations may potentially have?

96  Research handbook on soft law The second way of understanding this gap is the incongruence between political decision-making and democratic accountability. Here, we can identify different ways of strengthening the electorate, by voting, participation and deliberation, as suggested by the various interpretations of democracy. Perhaps the biggest challenge in contemporary politics on how to square public decision-making with democracy is in the EU. The scholarly discussion in recent years on the emergence of executive federalism in the EU (Habermas 2012), new intergovernmentalism (Bickerton et al. 2015; Jordana Casajuana and Solanes Mullor, this volume) and de-constitutionalization in the EU (Scharpf 2017) illustrate an increasing gap between political decision-making and democratic accountability. The strong role of the European Council may enhance collective decision-making (the value of performance) but may weaken the link to the European constitutional rules and to the representative democratic system of command and control. Further research on democracy and soft law should discuss whether and how the lack of electoral mechanisms of authorization and accountability can be handled through empowered public discourses, strong participation among those that are affected by certain decisions, and concrete measures of enhancing the transparency of what the political public agents are doing.

NOTES 1. In the famous ‘Gettysburg Address’ in 1863 the US President Abraham Lincoln expressed the distinction between government of the people, by the people, for the people (Lincoln 1973). 2. One important discussion in the literature on deliberative democracy is whether deliberation should be oriented to consensus or whether it should simply precede voting (Dryzek 2000). Another matter of debate is whether deliberation ought to be restricted to rational arguments or if other types of communication can be considered legitimate (Dryzek 2000; see also Fishkin 2009). 3. In relation to questions of participation, one pertinent issue in the deliberative model is how one is to determine the quality of the deliberation and configure the better argument. Fishkin argues that the quality of the deliberative process is determined by five criteria (Fishkin 2009), including criteria such as information (to what extent the participants get correct and relevant information), substantial balance (to what extent arguments are presented by counter arguments), and equal consideration (to what only arguments count and no other factors). The last of these criteria, how it is possible to disregard other factors, such as socio-economic differences among participants, is one of the reasons behind the massive critique of deliberative democratic thinking from the sides of liberal and republican-communitarian scholars (for an overview of the critique, see Dryzek 2010; see also Fishkin 2009). 4. In the literature on the EU, the most prominent scholar on the trade-offs between input legitimacy and output legitimacy is Fritz Scharpf (1999, 2002, 2017).

REFERENCES Agné, H., (2006). A dogma of democratic theory and globalization: why politics need not include everyone it affects. European Journal of International Relations. 12(3), 433–458. Archibugi, D., Koenig-Archibugi, M. and Marchetti, R., (2011). Global democracy. Cambridge: Cambridge University Press. Avant, D., Finnemore, M. and Sell, K. S., eds., (2010). Who governs the globe? Cambridge: Cambridge University Press. Baldwin, R., Cave, M. and Lodge, M., (2011). Understanding regulation: theory, strategy, and practice. Oxford: Oxford University Press. Beetham, D. and Lord, C., (2014). Legitimacy and the European Union. London: Routledge.

Democracy and soft law  97 Bexell, M. and Mörth, U., eds., (2010). Democracy and public-private partnerships in global governance. London: Palgrave Macmillan. Bickerton, C. J., Hudson, D. and Puetter, U., eds., (2015). The new intergovernmentalism. states and supranational actors in the post-Maastricht era. Oxford: Oxford University Press. Börzel, T. and Risse, T., (2005). Public-private partnerships: Effective and legitimate tools of transnational governance. In: E. Grande and L. W. Pauly, eds. Complex sovereignty: reconstituting political authority in the twenty-first century. Toronto: Toronto University Press. pp. 285-300. Bull, B. and McNeill, D., (2007). Development issues in global governance: public-private partnerships and market multilateralism. London: Routledge. Büthe, T. and Mattli, W., (2011). The new global rulers – the privatization of regulation in the world economy. Princeton: Princeton University Press. Cutler, A. C., (2003). Private power and transnational authority: transnational merchant law in the transnational political economy. Cambridge: Cambridge University Press. Cutler, C. and Dietz, T., eds., (2017). The politics of private transnational governance by contract, London: Routledge. Dahl, R., (1979). Procedural democracy. In: P. Laslett and J. Fishkin, eds. Philosophy, politics and society. Oxford: Blackwell. pp. 97–133. Dahl, R., (1989). Democracy and its critics. New Haven: Yale University Press. Diamond, L., (2002). Elections without democracy: thinking about hybrid regimes. Journal of Democracy. 13(2), 21-35. Diamond, L., (2019). Ill Winds: Saving democracy from Russian rage, Chinese ambition and American complacency. London/New York: Penguin Press. Diamond, L. and Plattner, M. F., eds., (2006). Electoral systems and Democracy. Baltimore: Johns Hopkins University Press. Dingwerth, K., (2007). The new transnationalism: Transnational governance and democratic legitimacy. Basingstoke: Palgrave Macmillan. Djelic, M.-L. and Sahlin-Andersson, K., eds., (2006). Transnational governance. Institutional dynamics of regulation. Cambridge: Cambridge University Press. Drache, D. (ed.)., (2001). The market or the public domain: global governance and the asymmetry of power. London: Routledge. Dryzek, J., (2000). Deliberative democracy and beyond. Liberals, critics, contestations. Oxford: Oxford University Press. Dryzek, J., (2010). Foundations and frontiers of deliberative governance. Oxford: Oxford University Press. Dryzek, J., (2011). Global democratization: Soup, society, or system? Ethics and International Affairs. 25(2), 211-234. Elster, J., (1988). Introduction. In: J. Elster and G. Hernes, eds. Constitutionalism and democracy – studies in rationality and social change. Cambridge: Cambridge University Press. pp. 1-18. Eriksen, O. E. and Fossum, J. E., (2000). Democracy in the European Union: integration through deliberation? London: Routledge. Eriksen, O. E. and Weigård, J., (2003). Understanding Habermas. Communicative action and deliberative democracy. London: Continuum. Eriksen, O. E. and Fossum, J. E., (2004). Europe in search of legitimacy: Strategies of legitimation assessed. International Political Science Review. 25(4), 435-459. Erman, E., (2012). “Democratic agents” and “agents of democracy” in multi-layered governance. In: L. Beckman and E. Erman, eds. Territories of citizenship. London: Palgrave Macmillan. pp. 60–79. Fishkin, J. S., (2009). When the people speak: Deliberative democracy and public consultation. Oxford: Oxford University Press. Føllesdal, A., (2011). The legitimacy challenges for new modes of governance: Trustworthy responsiveness. Government and Opposition. 46(1), 81–100. Gerencser, S., (2013). Corporations and clientelism: The problem of democratic accountability for corporate citizenship. In: K. Svedberg Helgesson and U. Mörth, eds. The political role of corporate citizens: an interdisciplinary approach. Basingstoke: Palgrave Macmillan. pp. 618–640. Gleckman, H., (2018). Multistakeholder governance and democracy – a global challenge. London: Routledge.

98  Research handbook on soft law Goodin, R., (2003). Reflective democracy. Oxford: Oxford University Press. Goodin, R., (2005). Sequencing deliberative moments. Acta Politica. 40(2), 182–196. Goodin, R., (2008). Innovating democracy: Democratic theory and practice after the deliberative turn. Oxford: Oxford University Press. Goodin, R., (2010). Global democracy: In the beginning. International Theory. 2(2), 175–209. Habermas, J., (1996). Between facts and norms. London: Polity. Habermas, J., (1998). Die postnationale Konstellation. Politische Essays. Frankfurt: Suhrkamp. Habermas, J., (2012). The crisis of the European Union – a response. Cambridge: Polity. Hall, R. B. and Biersteker, T., eds., (2002). The emergence of private authority in global governance. Cambridge: Cambridge University Press. Held, D., (1995). Democracy and the global order. Cambridge: Cambridge University Press. Held, D., (2006). Models of democracy. London: Polity. Helgesson Svedberg, K. and Mörth, U., eds., (2013). The political role of corporate citizens. an interdisciplinary approach. Basingstoke: Palgrave Macmillan. Héritier, A. and Rhodes, M., eds., (2010). New Modes of Governance in Europe: Governance in the shadow of hierarchy. Basingstoke: Palgrave Macmillan. Hix, S., (2008). What’s wrong with the European Union and how to fix it. Cambridge: Polity Press. Holzer. B., (2010). Moralizing the corporation: transnational activism and corporate accountability. Cheltenham: Edward Elgar. Holzer, B., (2013). Inadvertent citizens: Corporate citizenship and moral actorhood. In: K. Svedberg Helgesson and U. Mörth, eds. The political role of corporate citizens. An interdisciplinary approach. Basingstoke: Palgrave Macmillan. pp. 65–83. Jordana, J. and Levi-Faur, D., eds., (2004). The politics of regulation: Institutions and regulatory reforms for the age of governance. Cheltenham: Edward Elgar. Karlsson, C., (2001). Democracy, legitimacy and the European Union. Uppsala: Uppsala University Press. Kobrin, S. J., (2009). Private political authority and public responsibility: Transnational politics, transnational firms and human rights. Business Ethics Quarterly. 19(3), 349–374. Korkea-aho, E., (2016). Adjudicating new governance: deliberative democracy in the European Union. London: Routledge. Lee, C. W. and Romano, Z., (2013). Democracy’s new discipline: Public deliberation as organizational strategy. Organization Studies. 34(5–6), 733–753. Lincoln, A., (1973). The Gettysburg Address. NetLibrary, Inc. ISBN 0585006652. MacDonald, T., (2008). Global stakeholder democracy – Power and representation beyond liberal states. Oxford: Oxford University Press. March, J., (1994). A primer of decision making: how decisions happen. New York: Free Press. Mörth, U., (2008). European public-private collaboration. a choice between efficiency and democratic accountability? Cheltenham: Edward Elgar. Mörth, U., (2009). The market turn in EU governance – the emergence of public-private collaboration. Governance. 22(1), 99–120. Mörth, U., (2015). Sweden in a multigovernance polity. In J. Pierre, ed. Oxford Handbook of Swedish Politics. Oxford University Press Mörth, U. and Pierre, J., (2021). Can anyone implement the law? The discourse and practice of externalizing legal authority. Administration & Society. 53(9), 1315–1336. Néron, P.-Y. and Norman, W., (2008). Citizenship, Inc. Do we really want businesses to be good corporate citizens? Business Ethics Quarterly. 18(1), 1–26. Pache, A-C. and Santos, F., (2013). Inside the hybrid organization: Selective coupling as a response to competing institutional logics. Academy of Management Journal. 56(4), 972–1001. Ruggie, J. G., (2004). Reconstituting the global public domain — Issues, actors and practices. European Journal of International Relations. 10(4), 499–531. Sartori, G., (1987). The theory of democracy revisited. New Jersey: Chatham Publishers Inc. Scharpf, F., (1999). Governing in Europe: Effective and democratic? Oxford: Oxford University Press. Scharpf, F., (2002). Legitimate diversity: the new challenge of European integration. Cahiers Europeens de Sciences po, Paris, No 1.

Democracy and soft law  99 Scharpf, F., (2017). De-constitutionalisation and majority rule: A democratic vision for Europe. European Law Journal. 23(5), 315–334. Schumpeter, J., (1952). Capitalism, socialism and democracy. London: George Allen and Unwin Ltd. Senden, L., (2004). Soft law in European Community Law. Bloomsbury Publishing. Teorell, J., (2010). Determinants of democratization: Explaining regime change in the world 1972–2006. Cambridge: Cambridge University Press. Thomann, E., Hupe, P. and Sager, F., (2017). Serving many masters: Public accountability in private policy implementation. Governance. 31(1), 299–319. Thornton, P. H, and Occasio, W., (2008). Institutional logics. In: J. Greenwood, C. Oliver, K. Sahlin and R. Suddaby, eds. The SAGE Handbook of Organizational Institutionalism. London: SAGE Publications. pp. 99–129. Walzer, M., (1991). Constitutional rights and the shape of civil society. In: R. E. Calvert, ed. The Constitution of the people: Reflections on citizens and civil society. Lawrence: University Press of Kansas. pp. 113–126. Warren, M. E., (2012). Civil Society and Democracy. In: M. Edwards, ed. The Oxford handbook of civil society. Oxford/New York: Oxford University Press. pp. 377–390. Zelli., F., Bäckstrand, K., Nasiritousi, N., Skovgaard, J. and Widerberg, O., eds, (2021). Governing the Energy-Climate Nexus. Institutional Complexity and its Challenges to Effectiveness and Legitimacy. Cambridge: Cambridge University Press.

PART II DISCIPLINES, REGIMES AND AUTHORITY

7. Soft law in European public law Bruno de Witte

1. INTRODUCTION Bindingness is a key element of the European Union’s (EU) legal order, as it is of international law more generally (d’Aspremont 2019), and of national legal systems. The EU legal order is based on a rule of recognition indicating which written utterances of the Union’s institutions (and of the Member States acting together) shall be considered as legally binding, and which ones not. On a traditional understanding, this binary distinction does not leave room for the existence of soft law, that is, norms that, without being binding, are still part of ‘law’. However, in the system of sources of EU law there are clear indications that soft law is, in fact, part of the EU legal order, whilst being categorically different from what can, correspondingly, be called ‘hard law’. The key provision of primary EU law, in this respect, is Article 288 TFEU, whose text still corresponds largely to its original wording in the EEC Treaty of 1957. That provision makes an explicit distinction between EU legal instruments that are binding in various ways (namely, regulations, directives, and decisions) and EU legal instruments that are described as having ‘no binding force’ (namely recommendations and opinions). The latter belong to the EU legal order, as is shown by the fact that Article 288 expressly authorizes their use but also by the fact that the relevant section of the TFEU is entitled ‘The legal acts of the Union’. In other words, soft law is indeed law, as far as the EU legal order is concerned, even though it is categorically different from hard law. The term ‘soft law’ does not appear in the European Treaties, nor in the EU’s Charter of Fundamental Rights, which together form the written constitution of the EU. The concept of soft law was coined, instead, by legal writing. It was borrowed from scholarship in the field of public international law and introduced in the EU legal literature in the 1990s (Snyder 1994; Wellens and Borchardt 1999). The term proved to be popular as it captured the EU’s institutional reality, which is marked by the multiplication of non-binding normative texts other than opinions and recommendations (the two instruments mentioned in Article 288) with a wide range of denominations such as: guidelines, communications, resolutions, conclusions, notices and frameworks. ‘Soft law’ is a useful shorthand for all those instruments, capturing their common characteristics. A widely cited definition of soft law was offered by Linda Senden in her monograph on the subject (Senden 2004, p. 112): soft law rules are ‘rules of conduct that are laid down in instruments which have not been attributed legally binding force as such, but nevertheless may have certain (indirect) legal effects, and that are aimed at and may produce practical effects’. In this definition, it is rightly stated that legal effects are just a possibility (‘may’). Instead, what is essential in this definition is that soft law does not have binding force, that it consists of rules of conduct, and that these rules aim at producing practical effects – namely, that the addressees of the rules should ‘respond’ to those rules and should be persuaded to behave in accordance with them. These three characteristics justify the use of the term ‘law’ for rules that are formally defined as non-binding. They serve to distinguish soft law instruments from 101

102  Research handbook on soft law policy documents that set out the thinking of an EU institution on a certain matter without containing concrete rules of conduct that aim to steer the behaviour of other public or private actors. The soft law instruments considered in this chapter are those of public law, that is, those adopted by the institutions of the EU, excluding therefore the soft law instruments adopted by private actors in the framework of EU policies. This chapter will sketch the way in which soft law fits in the system of EU public law. In section 2, the main constitutional constraints that potentially limit the recourse to soft law will be discussed, namely the principles of conferral and of institutional balance. Section 3 will describe the place of soft law in the EU legal order by distinguishing different ways in which it relates to binding law. Section 4 will illustrate how the categorical divide between binding and non-binding law, whilst being a key characteristic of the EU legal order, is mitigated in several ways through the action of the EU’s judiciary and its political institutions.

2.

THE CONSTITUTIONAL EMBEDDING OF EU SOFT LAW: THE PRINCIPLES OF CONFERRAL AND INSTITUTIONAL BALANCE

2.1

The Principle of Conferral

The EU’s institutions must act within the limits of the competences conferred on the Union by the Treaties, and the concrete expression of that principle of conferral is the legal basis requirement: every action of an EU institution must be authorized, directly or indirectly (in the case of implementing acts), by a specific article of the TEU or TFEU. Those legal basis provisions, spread across the TEU and TFEU, do not only indicate the policy domain in which the EU institutions may act, but they also determine the decision-making procedure to be followed and sometimes they also indicate the legal instrument(s) to be used for that purpose. The latter feature affects the use of soft law instruments. Some legal bases generically refer to the power for EU institutions to adopt ‘measures’ on a given subject, and that generic term can include both hard law and soft law measures. When, however, the legal basis specifies that the Union shall act (only) by means of soft law, that excludes the use of binding legal instruments. For example, where Article 148 TFEU provides for the adoption by the Council of guidelines and recommendations for the employment policies of the Member States, this excludes the possibility to adopt binding legal instruments in that policy domain. Conversely, where the legal basis specifies the use of one or more instruments of hard law (say, directives or international agreements), the EU institutions may not adopt a soft law instrument instead of the hard law instrument. The Court of Justice thus held that the EU Council could not adopt soft law ‘conclusions’ establishing the EU’s position in an international negotiation when the relevant Treaty article (Article 218(9) TFEU) required the Council to adopt a binding ‘decision’ for that purpose. By doing so, the Council had violated an ‘essential procedural requirement’ and the principle of legal certainty (C-687/15, Commission v Council, para. 44). However, as long as the soft law act does not replace what should have been a binding act, the EU institutions can freely adopt it even when the legal basis fails to mention soft law. A well-known example of this is the guidance given by the Commission to the Member States on the kinds of state aid that are acceptable or not. Such guidance takes the form of

Soft law in European public law  103 variously named instruments (communications, guidelines or notices), which amount to an informal system of rule-making in the field of state aid, even though the adoption of these soft law instruments is not mentioned in the Treaty Articles 107–109 TFEU that deal with state aid. This practice is based on the assumption that, when the EU institutions can adopt binding measures in a given policy area, they can also adopt soft law measures which are less constraining on the Member States, according to an argument a maiore ad minus. In the particular case of state aid, since the Commission can adopt binding decisions authorizing or prohibiting particular instances of state aid, it can also provide soft law guidance to the Member States as to which kinds of state aid it finds acceptable. The same reasoning applies in all other policy domains in which the EU institutions can adopt binding measures: That power also includes the lesser power of adopting additional soft law measures. A trickier question is whether the European institutions can adopt soft law instruments in areas outside the EU’s competence as defined by the Treaties, that is, whether the principle of conferral constrains not only the adoption of binding rules but also the adoption of non-binding rules by the Union. The latter view seems compelling in light of the overall system of division of competences in the Treaties. Article 5(2) TEU defines the principle of conferral as meaning that ‘the Union shall act only within the limits of the competences conferred upon by the Member States in the Treaties’, whereby the verb ‘act’ sounds broad enough to comprise both binding and non-binding measures. In addition, Article 2(5) TFEU, the Treaty provision that defines the category of supplementary EU competences, distinguishes between the adoption of legally binding acts and other acts, which presupposes that the rules governing the definition and delimitation of Union competences indeed apply to non-binding acts as well. Some doubt about that conclusion could arise from Article 292 TFEU. That Treaty provision grants to both the Council and the Commission a seemingly general power to adopt recommendations, independently from a particular legal basis in one of the policy chapters of the TFEU. This can be inferred from the fact that Article 292 states that the ECB can adopt recommendations ‘in the specific cases provided for in the Treaties’. Since this limitation is only mentioned for the ECB and not for the Commission and the Council, it can be inferred that their power to adopt recommendations does not depend on whether it is authorized by a specific legal basis. Institutional practice shows that both these institutions adopt recommendations for which the only legal basis, mentioned in the preamble, is Article 292.1 However, this does not mean that the Commission and the Council can freely adopt recommendations on any random subject. The content of the recommendation must relate, rather, to one of the EU’s policy competences as defined elsewhere in the Treaties, and it would be good practice if recommendations based on Article 292 would systematically refer explicitly (as some of them do) to the substantive legal basis that justifies the adoption of this soft instrument.2 2.2

The Principle of Institutional Balance

Soft law instruments that remain within the limits of the EU’s overall competence can nevertheless be problematic from the perspective of another basic principle of EU constitutional law, that of institutional balance. According to Article 13(2) TEU, each EU institution ‘shall act within the limits of the powers conferred on it in the Treaties, and in conformity with the procedures, conditions and objectives set out in them’. However, as was noted earlier, the institutional practice is marked by the adoption of soft law instruments that are often not expressly based on the ‘procedures and conditions’ set out in the text of the Treaties. The adop-

104  Research handbook on soft law tion of soft law measures can therefore potentially disturb the institutional balance as struck by the text of the EU Treaties (see, generally, Senden 2005). In the case of the European Council, the text of the Treaties almost entirely ignores its soft law production, even though that soft law carries important political weight. There are a few Treaty provisions that allude to the adoption of conclusions by the European Council (Articles 121(2) and 148(1) TFEU), but Article 15 TEU, which describes the general role of the European Council, does not specify the legal form taken by the results of the European Council’s deliberations. In reality, the main output of these deliberations takes the form of conclusions, published on that institution’s website but not in the Official Journal of the European Union. The European Council conclusions are essentially policy documents, and they do not have binding legal force. Nevertheless, in addition to a number of inconsequential but politically desirable statements, the conclusions may contain important policy guidelines and rules of conduct addressed to the Member States or to the other EU institutions. Often those rules of conduct subsequently lead to the adoption of binding legal instruments by the other institutions, according to the rules and procedures laid down in the Treaties. The European Council thus produces soft law in the form of ‘conclusions’. This can be seen as implicitly authorized by the fact that the TEU describes its role as to ‘provide the Union with the necessary impetus for its development’ and to ‘define the general political directions and priorities thereof’ (Article 15(1) TEU). A potential threat to the institutional balance is formed by the abundant soft law activity of the Commission. When, in a given policy domain, the Union is allowed to take legislative action, the main role of the Commission is to propose such legislation, whereas its final content is decided by the Council and European Parliament (at least when the ordinary legislative procedure applies). When, in that same policy domain, the Commission adopts a recommendation or other soft law instrument, it decides on the content of that measure independently from the views of Council and Parliament and even if the Member States are under no legal obligation to implement that measure, the Commission still hopes to ‘softly’ but effectively steer national policies. The institutional justification for such initiatives can be found in Article 17 TEU, stating that ‘the Commission shall promote the general interest of the Union and take appropriate initiatives to that end’. However, the practice may appear problematic when the Commission adopts soft law measures where legislative action is not yet ‘ripe’ because there is no sufficient political support for it from the side of the Council and/or the European Parliament. The adoption of a soft law instrument can then appear as a means for the Commission to try to achieve some of the policy objectives that its legislative initiative sought to achieve or as a means to let new policy ideas ‘sink in’ gradually with reluctant Member States. For example, the Commission adopted a Recommendation in 2017 ‘on enhancing legal pathways for persons in need of international protection’ (EU Commission 2017). The recommendation stated among other things that the Member States should ‘offer at least 50.000 resettlement places’, ‘communicate to the Commission the numbers on a monthly basis’ and use EU financial support from the Asylum, Migration and Integration Fund. One year earlier, the Commission had proposed a legislative act to create a ‘Union Resettlement Framework’ (EU Commission 2016), but there was no prospect for its adoption by the co-legislators in the immediate future. From an institutional balance perspective, this begs the question whether the Commission’s power of initiative in the field of asylum includes a sort of ‘preparatory power’

Soft law in European public law  105 to make recommendations, even though this is not mentioned in the relevant legal basis (which is, in this case, in Article 78(2) TFEU). The Council could also disturb the legislative procedure by the adoption of soft law. That institution used to adopt non-binding resolutions in which it set out its views on a particular subject connected to a pending proposal for legislation, thus formulating the Council’s ‘red lines’ in that ongoing legislative procedure. Article 296, third sentence, TFEU, which states that ‘when considering draft legislative acts, the European Parliament and the Council shall refrain from adopting acts not provided by the relevant legislative procedure in the area in question’, now prohibits this practice. The Council still adopts resolutions (which are, in fact, its most frequently used soft law instrument) but not when their subject matter is part of an ongoing legislative procedure. Soft law can also affect the institutional balance in the context of the implementation of EU legislation by the EU institutions. Articles 290 and 291 TFEU only mention binding acts (namely, delegated and implementing acts) as the tools to be used for the central implementation of EU legislation. However, the EU institutions also use other implementation tools. On this point, the practice is overwhelmingly pointing to the conviction of the EU institutions (and in particular the Commission) that they can also exercise their executive role by means of soft law instruments, whenever that seems useful. In many cases, though, a legislative act provides itself for its implementation by means of soft law. In such a case, one could say that the institutional balance is preserved, as the EU’s legislator has openly accepted that other institutions or bodies (usually the Commission or one of the EU agencies) could steer the implementation by means of soft law measures. Consider, for example, the Commission recommendation of 7 May 2009 on the regulatory treatment of fixed and mobile termination rates. It is based on Article 19 of Telecom Directive 2002/21 allowing Commission recommendations for the ‘harmonised application’ of the directive. Similarly, the Guidelines on common procedures and methodologies for the supervisory review and evaluation process (SREP) adopted by the European Banking Authority on 19 December 2014 were based on a power provided by the founding act of that European Banking Authority.3 Another example is the Regulation establishing the new European Asylum Agency, which authorizes it to adopt two kinds of soft law measures: guidance notes to assist the Member States in the assessment of asylum applications on a country-of-origin basis, and recommendations to a Member State when the monitoring of that State’s asylum policy discloses particular problems (Regulation 2021/2303). Finally, soft law may affect the institutional balance in the field of the EU’s external relations. The EU institutions use a separate range of soft law instruments in their relations with third countries or international organizations, corresponding to the established practice in public international law. Typical examples thereof are the ‘memorandum of understanding’ (e.g., Memorandum 2007) and the ‘memorandum of cooperation’ (e.g., Memorandum 2011), but other, more EU-specific denominations are also found, such as the ‘mobility partnerships’ concluded with a number of countries. The wording of most of these instruments clearly indicates that they do not aim at creating rights or obligations under international law, thereby affirming their soft law status. Whereas soft law measures can be found in all domains of EU external policy, they have played a central role in some domains, such as the EU’s neighbourhood policy (Van Vooren 2009) and its external migration policy (Slominski and Trauner 2020; García Andrade 2018a, pp. 192–197). One can, more generally, observe an increase in the recourse to such soft law instruments (Ott 2020; Wessel 2021). The European institutions use such instruments for the same reasons as States, namely for the greater convenience that

106  Research handbook on soft law they may offer, compared to fully fledged treaties, in terms of the lack of formality for their conclusion, amendment and termination (Aust 2012, pp. 60–62). The institution whose power is most clearly affected by an uncontrolled proliferation of soft law measures is the European Parliament (EP). For instance, when the EU institutions adopted, in the Spring of 2020, a series of emergency soft law measures to deal with the COVID-19 pandemic, the European Parliament was almost entirely absent from the discussions (Ştefan 2020, p. 666), because its own functioning was also severely hampered by the pandemic. In the external relations domain, the EP has acquired an almost complete right to approve or reject formal treaties concluded by the EU with third States, but when informal agreements are being negotiated it is effectively sidelined from the decision-making process (García Andrade 2018b). That being said, soft law can also be used to strengthen the interactions between the institutions, as happened with the Interinstitutional Agreement concluded between Commission, Council and the European Parliament in which they defined non-binding criteria to be used by the EU legislator for the choice between delegated and implementing powers (Interinstitutional Agreement 2019). According to its point 1, the criteria seek to provide ‘guidance to the three institutions as to whether in legislative acts an empowerment should be of a delegated or an implementing nature, and should thus be given pursuant to Article 290 TFEU for the adoption of a delegated act or Article 291 for the adoption of an implementing act’. Interinstitutional agreements can be binding or non-binding, depending on the wording or the intent of the parties. In this case, the title of the agreement clearly indicates its soft law nature. The soft law practice of the EU institutions thus develops rather undisturbedly in an institutional universe that runs parallel to that of formal rule-making. The argument that this practice may occasionally undermine the institutional balance is made in the literature but seldom reaches the courtroom. Occasionally though, the Court of Justice seeks to protect the institutional balance against soft law practices. In some recent judgments, for example, it denied that European Council conclusions could bind the other EU institutions when they act in their legislative capacity (C-643/15 and C-647/15, Slovak Republic and Hungary v Council, paras 145–149; C-5/16, Poland v European Parliament and Council, paras 83–89). The European Council was thus firmly corralled within its political role, as prime agenda-setter and crisis-manager in the EU’s institutional system, but without the power to modify the formal decision-making procedures as they are set out in the Treaties.

3.

THE PLACE OF SOFT LAW IN THE EU’S LEGAL SYSTEM

Soft law plays a more important role in some European policy areas than in others (see also Radaelli and Tafoni in this volume; van Gerven and Stiller, this volume). In some domains, such as employment policy, soft law is the sole legal instrument of European governance, as is clear from the text of Article 148 TFEU. In this field, the Council adopts annual guidelines addressed to the Member States and may also adopt recommendations as a follow-up measure if it considers that a Member State has not been sufficiently responsive to the guidelines (Ashiagbor 2005, chapter 5). The EU institutions have, right from the start, sought to enhance the normative force of these guidelines by adopting them in the form of a decision, which is normally an instrument of binding EU law. However, this ‘hard law’ cover is deceptive, as the only things imposed on the Member States by those decisions is that they shall, in their

Soft law in European public law  107 employment policies, ‘take into account’ the guidelines set out in their annex, and that they shall report back to the Council on how they dealt with the guidelines (EU Council 2022, Article 2). There are no sanctions for non-compliance. The governance mechanism in the field of employment policy served as a template for a more general mode of governance, the Open Method of Coordination (OMC), which was launched in 2000 at a European Council meeting in Lisbon. The OMC was gradually extended to a number of policy domains other than employment policy, such as social protection and social inclusion (Armstrong 2010), culture (Psychogiopoulou 2018), education (Gornitzka 2018) and healthcare (de la Rosa 2018). These are all policy domains in which the EU’s competence to adopt binding legislation is rather limited. Today, the Open Method of Coordination (OMC) is still in use (Vanhercke and Pochet 2022, see also van Gerven and Stiller in this volume), although its place in the EU legal order is somewhat shadowy. In a glossary which is part of the Eur-Lex website of the EU, there is a short description of the OMC containing the statement that ‘the open method of coordination (OMC) in the European Union may be described as a form of “soft law”’. However, this statement does not adequately describe the OMC’s policy output, which is often neither hard nor soft law. The OMC in the field of culture, for one, is marked by the absence of instruments formulating rules of conduct. Its policy tools are the sharing of good practices and mutual learning between the member states rather than soft law (Psychogiopoulou 2018). In the field of economic governance, soft law is less dominant but still very present (see also Türk as well as Jordana and Solanes Mullor in this volume). The European surveillance of national fiscal policies and their macro-economic situation takes the form of opinions and recommendation adopted by the Commission and the Council. The most well-known of those instruments are the Country-Specific Recommendations (CSRs) adopted by the Council in the context of the ‘European Semester’ procedure of budgetary surveillance. However, unlike in the field of employment policy, these soft law measures of economic governance are embedded in a framework of hard law formed by detailed Treaty rules and by a number of regulations setting out the procedures leading to the adoption of these soft law instruments (Keppenne 2019). In this domain, soft law thus plays a complementary role to hard law. The complementary role of soft law is, in fact, pervasive in almost all EU policies in the form of post-legislative rule-making through soft law (Senden 2013), mentioned above in section 2. After the adoption of a directive or a regulation, the Commission or an EU agency frequently adopt further general rules that give guidance to the national authorities or to private parties, as the case might be, on how certain provisions of that directive or regulation should be transposed, applied or interpreted. Sometimes, soft law acts more like a precursor of binding law than as a complement to it, when the conditions for the adoption of binding legislation are not yet ripe. It even happens that a given soft law act is both a complement of existing legislation and a precursor of future legislation, as is shown by the Commission’s policy concerning national equality bodies. In 2018, the Commission adopted a recommendation setting out detailed quality standards for national equality bodies (EU Commission 2018). This was a case of post-legislative guidance, as the creation of these equality bodies was mandated by European anti-discrimination directives that had described the role of these bodies in very general and vague terms. More recently, the Commission proposed two directives on standards for equality bodies, arguing that the recommendation of 2018 had not sufficiently led the Member States to improve the structure, functioning and financial resources of their equality bodies (EU Commission 2022a

108  Research handbook on soft law and 2022b). The content of the proposed directives is very similar to the earlier recommendation, but the ‘should’ mode of formulation is transformed in a ‘shall’ mode, indicating the transition from soft to binding law. In this way, the recommendation of 2018 was both a supplement to earlier legislation and a precursor of (possible) future legislation. In the field of competition and state aid law, soft law plays a distinctive and important role in framing the application of hard law (Ştefan 2013; Georgieva 2015). These are policy areas in which the European Commission develops its own policy to apply the broadly worded Treaty provisions that prohibit anti-competitive market behaviour by firms or anti-competitive subsidies granted by national authorities to their undertakings. Although legislative acts dealing with competition and state aid are occasionally adopted, the main form of action in this domain is binding decisions adopted by the Commission to deal with individual cases. These individual decisions have been framed, for many years now, by the adoption of a large number of soft law instruments in which the Commission sets out its general policy in particular domains of competition or state aid law. A first such instrument was adopted in 1962, and since then a large number of variously denominated instruments have followed.4 These instruments allow the firms and the Member States to adapt their behaviour to the standards set out in the soft law instruments. Although the instruments are not legally binding, the addressees typically internalize their content since they know that, if they fail to comply with them, the Commission might adopt a binding individual decision prohibiting and sanctioning their actions. Because of this effect, the soft law approach can be seen as a genuine enforcement technique of EU competition law (Petit and Rato, 2009). Finally, soft law performs an additional function in all policy domains, namely that of permitting the rapid adoption of policy measures in the face of an emergency. Due to their informal status, soft law measures can be adopted more quickly than formal laws. That function was prominently displayed during the COVID pandemic. Among many other soft law measures, the Council adopted a recommendation to coordinate the actions of the Member States that, due to the pandemic, had restricted the free movement of persons between EU countries. As the pandemic evolved, the original recommendation from October 2020 could be rapidly and repeatedly adapted to the changing circumstances.5 The pandemic measures also illustrated the downside of this informal and flexible nature of soft law, namely that the process of its adoption lacks transparency and fails to involve participation by European or national parliaments, or by organized civil society (Eliantonio and Ştefan 2021; see also Mörth in this volume).

4.

MITIGATION OF THE HARD/SOFT LAW DISTINCTION IN THE EU’S JUDICIAL AND INSTITUTIONAL PRACTICE

The public law of the EU is predicated on a categorical divide between binding law and soft law. In the hierarchy of sources, all binding norms rank higher than all soft law norms; soft law cannot derogate from binding EU law. Soft law norms do not benefit either from the primacy and direct effect that characterize the relation between EU law and national law; there is no duty for national courts to disapply national legal rules for conflict with EU law. However, this strict divide between hard law and soft law has been somewhat mitigated by the judicial and institutional practice of the Union. The Court of Justice of the European Union, as the ultimate guardian of the integrity of the legal order, has upheld the binding/

Soft law in European public law  109 non-binding distinction but it has also accepted that, in particular circumstances, soft law instruments could entail legal obligations either for their addressees or for their authors. Apart from, and in addition to, this judicial policy, the practice of the EU’s political institutions and even, in some cases, the text of the European Treaties, have also contributed to mitigate the distinction between hard and soft law. This happens in two opposite ways: sometimes soft law instruments acquire a hard edge, and in other cases hard law instruments are filled with soft legal content. In this section, the judicial and institutional practice mitigating the binding/ non-binding distinction will be discussed in turn. 4.1

Judicial Practice

The role of the European Courts in exercising judicial review of soft law instruments and in defining the conditions for their application by national courts is discussed in the chapter by Eliantonio and Korkea-aho (this volume). Suffice is to say here that, in the context of their judicial review of the legality of EU acts, the Court of Justice and the General Court of the EU have taken a non-formalist view of the distinction between binding and non-binding instruments. The EU Courts could have taken the line that non-binding instruments can never impose any legal obligations on Member States or individuals and should therefore never be the subject of judicial review. However, they have rather taken the view that instruments with a non-binding label could nevertheless aim, in substance, at imposing legal obligations, in which case the Courts can examine their legality in the context of a direct action or a preliminary reference. For instance, in a judgment of 2011, the Court of Justice examined in detail the content of Commission Guidelines in the field of electronic communications before concluding that they did not lay down binding obligations, instead of simply stating that Guidelines can never constitute binding norms (C-410/09, Polska Telefonia Cyfrowa, paras 30–35). In a number of cases, it annulled measures that had been formally enacted as soft law. A characteristic early example is a judgment of 1993 in which the Court of Justice reviewed a Commission communication on the transparency of financial relations between the Member States and national public undertakings. The communication set out the Commission’s views on how the Member States should apply an existing directive on the same subject. The Court found that the communication imposed on the States a reporting obligation that was not required by the text of the directive, so that the communication, even though formally non-binding, constituted ‘an act intended to have legal effects of its own’ (C-325/91, France v Commission, para. 23). The Court annulled the communication for breach of the principle of legal certainty, since it had been adopted without referring to a legal basis in the Treaties, whereas this reference is required for any binding legal act. In a number of other cases, the Court of Justice has made sometimes very lengthy assessments of whether or not a formally non-binding act did, in fact, have legal effects. This attitude by the Court seems inspired by the wish to offer extensive judicial protection against EU measures, even though some authors argue that the Court should go further in this direction (Ştefan 2014). The disadvantage of this non-formalist approach is that it blurs somewhat the distinction between binding and non-binding acts, a distinction which, as mentioned above, plays a vital role in the EU’s legal order. A special judicial review regime applies to recommendations. Article 263 TFEU expressly excludes recommendations from the category of acts that can form the object of an action for annulment. Recommendations are thus shielded from direct judicial review as to their

110  Research handbook on soft law compatibility with rules of primary law such as the principle of conferral or the principle of institutional balance. This shielding effect can be illustrated by the European Court’s judgment of 2018 in a case brought by Belgium against the Commission. Belgium sought the annulment of a Commission recommendation on prevention of gambling online, raising arguments relating to the principles of conferral and institutional balance. The Court did not even enter into a discussion of those constitutional questions, as it considered this to be a genuine recommendation and it, hence, declared the action for annulment inadmissible (C-16/16 P, Belgium v Commission). It was critically noted that this attitude of the Court ‘disturbs the institutional balance by offering the Commission a way of circumventing the decision-making processes laid down by the Treaty’ to the detriment of the Council and European Parliament (Arnull 2018, p. 617). However, the Court of Justice does examine the validity of recommendations in the framework of a preliminary reference on validity made by a national court under Article 267 TFEU. This is somewhat odd, as the action for annulment and the reference on validity are traditionally seen (also by the Court itself) as complementary means to review the legality of EU acts. It is therefore curious that the Court finds challenges of the legality of recommendations admissible when they are submitted to it by a national court, but not when they are submitted directly through an action for annulment brought by a Member State or an individual or legal person to whom the recommendation is addressed (Chamon and de Arriba-Sellier 2022, pp. 306–308). The Court based its ‘asymmetric’ approach on the fact that, unlike in Article 263, the text of Article 267 does not expressly exclude recommendations from the ‘acts’ that may form the subject of a preliminary reference. In the field of competition and state aid law, the Court of Justice developed a line of case law that also blurs the divide between binding and non-binding EU acts, but in a different way, namely by finding that soft law measures can be binding on their author (rather than their addressee). It held that soft law documents of the Commission in which it sets out its policy on particular questions of competition or state aid law impose a self-limitation on the Commission with regard to its subsequent decisional practice in individual cases. The Court held that: in adopting such rules of conduct and announcing by publishing them that they will henceforth apply to the cases to which they refer, the institution in question [i.e. the Commission] imposes a limit on the exercise of its discretion and cannot depart from those rules under pain of being found, where appropriate, to be in breach of the general principles of law, such as equal treatment or the protection of legitimate expectations (C-189/02 P, Dansk Rørindustri , para. 211).

That case related to the Commission’s fining policy in competition cases but was also applied to state aid cases (C-526/14, Kotnik, para. 40) and in other policy fields in which the Commission produces soft law guidance, such as the regulation of financial services (C-667/13, Banco Privado, para. 69). Finally, the Court of Justice also held that, despite the fact that soft law does not have primacy over national law, national courts cannot entirely ignore it. Where the Member States have adopted measures based on an EU recommendation, that recommendation may cast light on the interpretation of those measures and should therefore be taken into consideration by the national courts (C-322/88, Grimaldi, para. 18; Korkea-aho 2018). In such cases, the EU soft law measure has acquired legal effect in the national legal system by means of its ‘transposition’ into national law, even though that transposition itself was a voluntary act of the member state concerned. In this manner, EU soft law becomes significant for national judicial practice (Eliantonio, Korkea-aho and Ştefan 2021).

Soft law in European public law  111 4.2

Institutional Practice

The practice of the EU institutions has also contributed to mitigate the hard/soft law distinction: sometimes soft law instruments acquire a hard edge, and in other cases hard law instruments are filled with soft legal content. As for the first category, a ‘hard edge’ is commonly found in recommendations adopted either by the Commission or by the Council and addressed to the Member States that include provisions on monitoring and reporting. These so-called comply-or-explain mechanisms ‘require the domestic authorities to report, to respond, to cooperate or to otherwise engage with ideas, projects or proposals uttered by the EU level’ (Weiss 2022, p. 13). These duties imposed on the Member States are not binding by themselves (since they are contained in soft law instruments) but failing to comply with such procedural duties could possibly be seen as a breach of the principle of sincere cooperation laid down in Article 4(3) TEU. In other cases, the Treaty text itself provides for procedural mechanisms whereby soft law measures may lead, in a further stage of the procedure, to binding rules. This occurs in the field of economic governance: when euro area Member States fail to take ‘appropriate action’ in response to fiscal policy recommendations addressed to them, the Council can eventually adopt financial sanctions, as foreseen in Article 126(11) TFEU (Keppenne 2019, pp. 68–69). These sanctions are of course binding on the Member States concerned. The Council adopted such binding measures only once, against Spain and Portugal, but even then the actual fine was cancelled (EU Council 2017). The blurring of the soft/hard law divide also happens in the opposite direction, when binding EU legislation is filled with content that is rather soft. A recent example of this phenomenon is the EU directive on minimum wages (Directive 2022/2041). As this is a disputed subject matter, at the outer limits of EU competence, the wording of the directive is very cautious. It repeatedly uses the formula that the Member States shall take ‘appropriate measures’ on matters covered by the directive, which leaves the national authorities so much discretion that the obligations, while formally binding, are effectively quite soft. Another form of softening a binding instrument of EU law happens when the EU’s political institutions declare that binding norms adopted by them shall not be judicially enforceable. This denial of direct effect is a frequent phenomenon in the field of EU external relations. Most free trade agreements concluded by the EU since 2008 contain clauses excluding direct effect, and thereby exclude that these agreements may serve as an effective legality constraint in the EU legal order. This is the case, for example, for the Free Trade Agreement with Korea, the Economic Partnership Agreement with the CARIFORUM States, the Trade Agreement with Colombia and Peru, and those concluded with Ukraine and with Singapore (Semertzi 2014). A further example is Article 30.6 of the trade agreement with Canada (CETA), stating that ‘nothing in this Agreement shall be construed as conferring rights or imposing obligations on persons other than those created between the Parties under public international law, nor as permitting this Agreement to be directly invoked in the domestic legal systems of the Parties’ (Agreement 2017). This exclusion of direct effect also happened in the EU’s neighbourhood relations, albeit in a different form: the EU’s association agreements with Ukraine do not contain a no-direct-effect clause, but the EU Council, when signing them, stated that ‘the Agreement shall not be construed as conferring rights or imposing obligations which can be directly invoked before Union or Member State courts or tribunals’ (EU Council 2014a, Article 5; EU Council 2014b, Article 3). The Council thus seeks to pre-empt the Court of Justice from applying its habitual favourable approach to the direct effect of bilateral asso-

112  Research handbook on soft law ciation agreements. In this manner, the provisions of those Treaties are much softer, in their actual operation, than their wording would seem suggest. Something comparable occurs, in a more systematic way, in the other branch of the EU’s external relations, namely the Common Foreign and Security Policy (CFSP). There, it is the text of the Treaties itself (rather than the practice of the EU institutions) that states that legally binding decisions of the Council are not judicially enforceable before the EU’s Court of Justice (Article 24(1) TEU). Compliance by the Member States with those binding decisions is thus as ‘voluntary’ as if these were soft law measures.

5. CONCLUSION The EU legal system lists and defines the sources of binding law. It also recognizes the existence of soft law as part of the legal system. The EU Treaties give some general and specific functions to a selected number of soft law instruments, especially to recommendations and opinions. The legal system does not set quantitative limits to the use of soft law instruments, meaning that each EU institution and EU agency can make as much soft law as it sees fit, with only a few limits (such as the prohibition for Council and Parliament to adopt soft law measures related to the subject of an ongoing legislative procedure). The Court of Justice did not seek to restrain the use of soft law, and did not act as a strict guardian of the ‘legally binding world’ of EU law. It rather accepted the institutional practice of using soft law instruments for a variety of purposes that are either aimed at strengthening the effectiveness of binding EU law or at trying to steer the behaviour of the Member States or of private actors where the Union is legally or politically unable to adopt binding legislation.

NOTES 1. See, for example, Commission Recommendation of 14 July on principles for the protection of consumers and players of online gambling services and for the prevention of minors from gambling online, O.J. 2014, L 214/38, and Commission Recommendation 2015/914 on a European resettlement scheme, O.J. 2015, L 148/32. The latter instrument deals with the resettlement of asylum seekers, but its legal basis is Art 292 rather than, as one would expect, one of the articles in the Treaty chapter on immigration and asylum. 2. See, for example, Council Recommendation 2020/1475 of 13 October 2020 on a coordinated approach to the restriction of free movement in response to the COVID-19 pandemic, OJ 2020, L 337/3. The preamble cites as its legal bases not only Article 292 TFEU but also two substantive legal bases, namely Article 21(2) TFEU (free movement of citizens) and Article 168(6) TFEU (public health). See also Commission Proposal for a Council Recommendation on adequate minimum income ensuring active inclusion, COM(2022) 490 of 28 September 2022, citing Article 292 TFEU as the procedural legal basis in combination with Article 153(1) (combating social exclusion) as the substantial legal basis. 3. Regulation 1093/2010, Art. 16(1): ‘The Authority shall, with a view to establishing consistent, efficient and effective supervisory practices within the EFSF, and to ensuring the common, uniform and consistent application of Union law, issue guidelines and recommendations addressed to competent authorities and financial institutions.’ 4. A random selection of such instruments includes: European Commission, Guidelines on the application of Article 81(3) of the Treaty, O.J. 2004, C 101/97; European Commission, Notice on the notion of state aid as referred to in Article 107(1) of the Treaty on the Functioning of the European Union, O.J. 2016, C 262/1; European Commission, Guidelines on Regional State Aid for 2014-2020, O.J.

Soft law in European public law  113 2013, C 209/1; European Commission, Communication of 20 March 2020, Temporary framework for state aid measures to support the economy in the current COVID-19 outbreak, O.J. 2020, C 91 I/1. 5. See Council Recommendation 2021/961 of 14 June 2021 amending Recommendation (EU) 2020/1475 on a coordinated approach to the restriction of free movement in response to the COVID-19 pandemic, O.J. 2021, L 213I/1. Further Recommendations with the same title, replacing the original one from 2020, were adopted on 25 January 2022 and 13 December 2022 (see, respectively O.J. 2022, L 18/110 and O.J. 2022, L 328/138).

REFERENCES Agreement., (2017). Comprehensive Economic and Trade Agreement between Canada, of the one part, and the European Union and its Member States, of the other part, O.J. 2017 L 11/23. Agreement., (2018). Agreement between the European Union and Japan for an Economic Partnership, O.J. 2018 L 330/3. Armstrong, K., (2010). Governing social inclusion. Europeanization through policy coordination. Oxford: Oxford University Press. Arnull, A., (2018). EU recommendations and judicial review. European Constitutional Law Review. 14, 609–621. Ashiagbor, D., (2005). The European employment strategy. Oxford: Oxford University Press. d’Aspremont, J., (2019). Bindingness. In: J. d’Aspremont and S. Singh, eds. Concepts for international law: Contributions to disciplinary thought. Northampton: Edward Elgar. pp. 67–82. Aust, A., (2012). Alternatives to treaty-making: MOUs as political commitments. In: D.B. Hollis, ed., The Oxford Guide to Treaties. Oxford: Oxford University Press. pp. 46–72. Case C-322/88, Grimaldi, EU:C:1989:646. Case C-325/91, France v Commission, EU:C:1993:245. Joined Cases C-189/02 P, C-202/02 P, C-205/02 P to C-208/02 P, and C-213/02 P Dansk Rørindustri and others v Commission, EU:C:2005:408. Case C-410/09, Polska Telefonia Cyfrowa, EU:C:2011:294. Case C-667/13, Banco Privado Portugues and Massa Insolvente do Banco Privado Portugues, EU:C:2015:151. Case C-526/14, Kotnik and Others, EU:C:2016:570. Joined Cases C-643/15 and C-647/15, Slovak Republic and Hungary v Council, EU:C:2017:631. Case C-687/15, Commission v Council, EU:C:2017:803. Case C-5/16, Poland v European Parliament and Council, EU:C:2018:483. Case C-16/16 P, Belgium v Commission, EU:C:2018:79. Chamon, M. and de Arriba-Sellier, N., (2022). FBF: on the justiciability of soft law and broadening the discretion of EU agencies. European Constitutional Law Review. 18, 286–314. de la Rosa, S., (2018). The OMC processes in the health care field: what does coordination really mean? European Papers. 3, 215–234. Directive., (2022). Directive 2022/2041 of the European Parliament and of the Council of 19 October 2022 on adequate minimum wages in the European Union, O.J. 2022, L 275/33. Eliantonio, M. and Ştefan, O., eds., (2021). The elusive legitimacy of EU soft law: An analysis of consultation and participation in the process of adopting COVID-19 soft law in the EU. European Journal of Risk Regulation. 12, 159–175. Eliantonio, M., Korkea-aho, E. and Ştefan, O., eds., (2021). EU soft law in the member states. Theoretical findings and empirical evidence. Oxford: Hart Publishing. EU Council., (2014a). Council Decision 2014/295, O.J. 2014 L 161/1. EU Council., (2014b). Council Decision 2014/669, O.J. 2014 L 278/6. EU Council., (2017). Council Implementing Decisions 2017/2350 and 2017/2351, O.J. 2017, L 336/24 and 336/27. EU Council., (2022). Council Decision 2022/2296 on guidelines for the employment policies of the Member States, O.J. 2022 L 304/67.

114  Research handbook on soft law EU Commission., (2009). Commission Recommendation 2009/384 of 30 April 2009 on remuneration policies in the financial services sector, O.J. 2009 L 120/22. EU Commission., (2016). Proposal for a Regulation of the European Parliament and of the Council establishing a Union Resettlement Framework, COM (2016) 468 of 13 July 2016. EU Commission., (2017). Commission Recommendation 2017/1803 of 3 October 2017, O.J. 2017 L 259/21. EU Commission., (2018). Commission Recommendation 2018/951 of 22 June 2018 on standards for equality bodies, O.J. 2018 L 167/28. EU Commission., (2022a). Proposal for a Directive on standards for equality bodies (equal treatment of women and men in employment), COM(2022) 688 of 7 December 2022. EU Commission., (2022b). Proposal for a Directive on standards for equality bodies (other grounds of discrimination), COM(2022) 689 of 7 December 2022. García Andrade, P., (2018a). EU external competence in the field of migration: How to act externally when thinking internally. Common Market Law Review. 55, 157–200. García Andrade, P., (2018b). The role of the European Parliament in the adoption of non-legally binding agreements with third countries. In: J. Sánchez Vara and S. R. Sánchez Tabernero, eds. The democratization of EU international relations through EU law. New York: Routledge. pp. 115–131. Georgieva, Z. R., (2015). Soft law in EU competition law and its judicial reception in member states – A theoretical perspective. German Law Journal. 16, 223–260. Gornitzka, A., (2018). Organising the governance in hard times – the unlikely survival of the open method of coordination in EU education policy. European Papers. 3, 235–255. Interinstitutional Agreement., (2019). Interinstitutional agreement between the Commission, the Council and the European Parliament of 18 June 2019 on non-binding delineation criteria for the application of Articles 290 and 291 TFEU, O.J. 2019 C 223/1. Keppenne, J. P., (2019). Fiscal rules. In: Fabbrini, F. and Ventoruzzo, M., eds. Research handbook on EU economic law. Northampton: Edward Elgar, pp. 35–88. Korkea-aho, E., (2018). National courts and European soft law: Is Grimaldi still good law? Yearbook of European Law. 37, 470-495. Memorandum., (2007). Memorandum of Understanding between the Council of Europe and the European Union, Strasbourg, 13 May 2007. Memorandum., (2011). Memorandum of Cooperation between the European Union and the International Civil Aviation Organisation providing a framework for enhanced cooperation, OJ 2011, L 232/2. Ott, A., (2020). Informalization of EU bilateral instruments: categorization, contestation and challenges. Yearbook of European Law. 39, 569–601. Petit, N. and Rato, M., (2009). From hard to soft enforcement of EC competition law – A bestiary of “sunshine” enforcement instruments. In: Gheur, C. and Petit, N., eds. Alternative enforcement techniques in EC competition law. Bruxelles: Bruylant. pp. 183–219. Psychogiopoulou, E., (2018). The cultural open method of coordination: a new but different OMC? European Papers. 3, 257–279. Regulation 2021/2303 of the European Parliament and of the Council of 15 December 2021 on the European Agency for Asylum and repealing Regulation (EU) No 439/2010, O.J. 2021 L 468/1. Semertzi, A., (2014)., The preclusion of direct effect in the recently concluded EU free trade agreements. Common Market Law Review. 51, 1125–1158. Senden, L., (2004). Soft law in European Community law. Oxford: Hart. Senden, L., (2005). Soft law and its implications for institutional balance in the EC. Utrecht Law Review. 1, 79–99. Senden, L., (2013). Soft post-legislative rulemaking: A time for more stringent control. European Law Journal. 19, 57–75. Slominski, P. and Trauner, F., (2020). Reforming me softly – how soft law has changed EU return policy since the migration crisis. West European Politics. 44, 93–113. Snyder, F., (1994). Soft law and institutional practice in the European Community. In: S. Martin, ed. The construction of Europe. Essays in honour of Emile Noël. Dordrecht: Kluwer, pp. 197-225. Ştefan, O., (2013). Soft law in court: Competition law, state aid and the Court of Justice of the European Union. Alphen: Kluwer Law International.

Soft law in European public law  115 Ştefan, O., (2014). Helping loose ends meet? The judicial acknowledgement of soft law as a tool of multi-level governance. Maastricht Journal of European and Comparative Law. 21, 359–379. Ştefan, O., (2020). COVID-19 soft law: voluminous, effective, legitimate? A research agenda. European Papers. 5, 663–670. Vanhercke, B. and Pochet, P., (2022). The Open Method of Coordination: a contested public policy instrument still going ‘soft’. In: Graziano, P.R. and Tosun, J., eds. Elgar encyclopaedia of European Union public policy. Northampton: Edward Elgar. pp. 454-463. Van Vooren, B., (2009). A case study of ‘soft law’ in EU external relations: The European neighbourhood policy. European Law Review. 34, 696–719. Weiss, W., (2022). Pandemic and administrative EU soft law: persistent challenges to the rule of law in the EU and possible solutions. Review of European Administrative Law. 15, 7. Wellens, K. C. and Borchardt, G. M., (1999). Soft law in European Community law. European Law Review 14, 267–321. Wessel, R. A., (2021). Normative transformations in EU external relations: The phenomenon of ‘soft’ international agreements. West European Politics. 44, 72–92.

8. Better regulation as soft law Claudio M. Radaelli and Gaia Taffoni

1.

INTRODUCTION: BETTER REGULATION AND EU LAW

This chapter deals with the ‘better regulation’ agenda of the European Union (EU) as soft law. The topic of ‘better regulation’ (over the years this term has been tweaked with variations around the themes of better law-making and smart regulation) has been firmly on the agenda of the EU since the early 2000s. Its early steps date back to the 1990s. Hence, we are talking of a long-standing element of the soft law activities of the EU. The connection between ‘better regulation’ and EU law is clear, since the former is a set of practices and standards about how to produce and manage law across the policy cycle. Thus, essentially, better regulation is about how to make high-quality legislation. As we shall see, this statement comes with qualifications. Better regulation has been criticized as EU de-regulation, harmful to the goals of social and environmental policies (Van den Abeele 2015; Schömann 2015; Garben 2018). But, no matter how we appraise it normatively, the point about the connection with EU law stands: better regulation is an agenda to reform and change EU law across different policy areas and stages of the policy cycles. It is a set of rules about EU law-making – later we will in fact talk about its meta-regulatory properties. Before we get into the subject matter of this chapter, we need to clarify that to make a statement about the quality of a rule (and say how it should get ‘better’) is not a straightforward, easy step. Even the deceivingly trivial question of what a regulatory burden really is reveals ambiguity and complexity (Lodge and Wegrich 2012, pp. 12–26). Hence, what is the problem that better regulation is supposed to address? Regulation means protection from risks, but it also means rules that in some cases are overwhelming and stifle economic activity and entrepreneurship. In the words of Bruce Doern, the reform design underlying better regulation should address both red tape and red flags (Doern 2007). To do that, we need a fine toothcomb that allows us to establish the quality of a given rule. Yet – here is the thing – regulatory quality depends on the paradigms or conceptual lenses we adopt. For a politician, quality may well mean sufficient levels of consensus around a rule. For an economist, it may mean efficiency. For a bureaucrat, good rules come out of routines and legitimate procedures. In short, quality depends on the ‘theory hat’ one wears (Gunningham et al. 1998; Baldwin and Cave 1999) and the seat one takes at the law-making table (be it the seat of a politician, a civil society organization, or an expert). Some theoretical paradigms suggest that regulation is the public interest response to market externalities. Other theories argue that capture and interest group politics shape the system of rules we have at a given time in a given place (Lodge and Wegrich 2012, pp. 27–46). Consequently, all normative statements about existing regulation need to be clear on the theoretical lenses and presuppositions adopted. Without this clarity upfront, better regulation remains undetermined so to speak. In this undetermined policy space, politics takes advantage of semantic ambiguity (Radaelli 2023). When international organizations or governments decide to reduce the stock of rules 116

Better regulation as soft law  117 and commit regulators to cost ceilings or regulatory budgets in the name of high-quality regulation, we are witnessing an argument in terms of quantity – a de-regulatory argument. When the argument is not about the number of rules but whether they match certain benchmarks or criteria we are closer to the domain of quality. When someone says that beyond a certain quantity the legal system becomes a maze impossible to implement, enforce and adjudicate, one is making the argument that too many rules deteriorate the quality of the system. The EU is not alone in having adopted (and promoted in its Member States) the discourse and policies of ‘better regulation’ (from now onwards without the quotation marks). Indeed, the same language, activities and tools appear on the agenda of the World Bank (2010) and the Organisation for Economic Cooperation and Development (OECD 1995, 2012). The presence of better regulation is an empirical manifestation of an agenda that can be examined without making prior normative assumptions. As we shall see, the EU has adopted soft law to promote this agenda – hence the fit with the volume. To be clear: for us, the fact that an institution defines its regulatory reform policy as better regulation does not necessarily mean that it meets the previously mentioned academic benchmarks, or, is better for European law or for anyone. Neither do we presuppose coherence in what the EU has done with better regulation. Indeed, there are fierce criticisms of better regulation (Kysar 2010; Garben and Govaere 2018). And the EU has pursued quality and quantity objectives at different times, with some contradictions (Radaelli 2021). Objectively, the EU better regulation policy is one possible set of visions, discourses and tools about regulatory reform and the governance of EU legislation. To wrap up and carry on, better regulation is a governance agenda for the reform of law-making. As such, it provides discourse as well activities and tools to manage regulation across the policy cycle. In the EU and its Member States, its stated aims have been to improve on the transparency and quality of law-making, to appraise with evidence-based tools both the stock and the flow of regulation, to simplify, to reduce burdens, and to create regulatory oversight bodies. The specific policy instruments adopted by the European Commission and the Member States have not always been the same, but stakeholder consultation, impact assessment, more recently legislative evaluation and foresight, and variations on the theme of regulatory offsetting and the cull of administrative burdens are pivotal. As mentioned, for lawyers and political scientists, better regulation is also an instance of meta-regulation, that is, a set of rules dedicated to how rules should be made and managed across their life cycle (Scott 2003, 2010; Radaelli 2010). Falling outside the Treaties, better regulation has always been handled by the EU in terms of communications, guidance, reports, and so on. Hence it is exclusively soft law – although the situation in the Member States is, as we shall see, quite different, presenting a combination of soft and hard law. In the next section we examine better regulation from the policy and politics angles focusing on the EU level. Section 3 traces the history of better regulation as politics. Section 4 examines two central tools of better regulation: Impact Assessment (IA) and consultation at the Member State level (and the UK), highlighting their variation and the mix of hard and soft law. In the last section, we draw the main conclusions on better regulation demonstrating that this specific type of soft law has become the field where European institutions measure their power and Member States express their domestic preferences.

118  Research handbook on soft law

2.

POLICY ARENAS AND POLITICS

If we consider the European Commission and more broadly the EU policy process, we find that better regulation has its own distinctive underlying assumptions about the problem(s) it is supposed to solve, actors, processes, inter-institutional agreements, and instruments. Consequently, we can call it a public policy. But, following Lowi (1972), for every policy there is a politics dimension or arena of power. Public policies determine their own political arenas with their distinctive issues, actors, and decision-making processes. The tension between policy and politics is our lens to examine this particular instance of soft law in the EU and its Member States. Let us first focus on the EU level. Here, essentially, the policy space for better regulation is a set of actors, instruments, decision-making processes, and institutions competent to make the decisions. As politics, it is the terrain where the Member States, the European Parliament and the Commission exploit the ambiguities we have seen in the Introduction to define who is in control of the law-making process in the EU (Radaelli 2018). Better regulation politics is particularly intense. We see this in the historical relationship between the Commission and the Member States. Should better regulation be an arena that allows the Member States and the European Parliament to control (or at least to demarcate) the power of the Commission on law-making, qualifying over time the Commission’s Treaty-based right to initiate legislation? Impact assessment and consultation can be used to make the Commission more accountable to Member States and MEPs (and of course pressure groups). Or, alternatively, does better regulation enhance the control of the bureaucracy on the life cycle of EU legislation? This can happen in different ways: 1. by making it more difficult to criticize the Commission’s proposals supported by robust impact assessments and positive opinions of pressure groups gathered via consultation; 2. by building over time the option (for the Commission) to exempt proposals from impact assessment or withdraw proposals without publishing an impact assessment; 3. by leveraging consultation to create synergy on policy issues with some pressure groups; 4. by asking the European Parliament and the Council to carry out impact assessments of their substantive amendments, thus proving that the amendments do not make legislation worse in comparison to the original proposal. EP and Council’s impact assessment are foreseen in the interinstitutional agreement on better law-making of 2016 (Interinstitutional Agreement 2016); 5. by asking Member States to report to the Commission on administrative burdens and compliance costs introduced at the stage of transposition and implementation; and 6. finally by creating a coherent and sound policy inside the Commission via the empowerment of the Secretariat General over the Directorates General. In short, for some, better regulation is a policy arena to make the Commission’s law-making more accountable and controllable, for others it is a tool to preserve and possibly expand the power of the Commission. It follows that even if this soft law as policy was considered irrelevant because Communications and guidelines are toothless (something we are not arguing, but we consider for the sake of presenting an extreme argument), as politics better regulation goes right to the core of who is in charge of the life cycle of legislation.

Better regulation as soft law  119

3.

HOW DID WE GET THERE?

To find a day 1 in EU better regulation we must go back to the Edinburgh European Council of 1992. There, the UK, with Germany, and the Netherlands articulated a concern shared with the business community about the quality of EU law-making and the need to simplify legislation. In those days there were also talks to create a body of ‘guardians of the rules’ (an independent review body) that would one day take the shape of a ‘European Conseil d’Etat’ (Radaelli 1999). The Member States played the first card. No one was asking for hard law measures in this field. Yet the experience of a few leading Member States and above all OECD guidance, codified in a set of principles in 1995 (OECD 1995), showed a possible way forward: a set of commitments and principles taken by the Commission to improve on the quality of law-making, together with the adoption of procedural policy instruments, especially consultation and impact assessment of policy proposals. The Commission did not immediately embrace wholesale the better regulation principles and tools. Rather, initially it responded with individual projects and task forces such as the Business Environment Simplification Task Force (BEST). The Commission experimented with some ad hoc projects for impact assessment (fiches d’impact were rare and not public), business consultation, rules of procedure, legislative drafting manuals and the generic commitment to openness (Radaelli 1999). The Directorates General (DGs) kept their autonomy when handling the checklists, and coordination across services remained low priority. Ex-post policy evaluation remained limited to financial controls only. After an almost a decade since Edinburgh, the politics of better regulation was still unresolved. And indeed it re-appeared neatly with the pressure on the Commission to respond to the resignation of the Santer Commission with the adoption of, among other things, a regulatory reform and governance agenda. The new call on the Commission appeared in the so-called Mandelkern Report (2001) published in November 2001 – this group was staffed by high profile Member States delegates. This report specifically addressed the Commission, asking for a comprehensive policy on regulatory reform, including consultation, regulatory impact assessment, and the consideration of alternatives to traditional command and control regulation. Mandelkern went as far as to propose a deadline asking the Commission to ‘propose by June 2002 a set of indicators of better regulation’ (Mandelkern Report 2001, pp. iii, 59). The Commission had its own internal problems in responding to this call. Some DGs were keen on the regulatory reform agenda, others far less. The Secretariat General saw in better regulation an opportunity to take a more central coordination role in policy formulation and started to like the idea of a single consultation procedure and a template for the appraisal of all proposals for directives and regulations. The approach of the Commission to impact assessment then emerged in the years 2002–05 (Allio 2008). It revolves around the economic, environmental, and social dimensions of regulation. The three-fold articulation reflects the deal between the DGs oriented towards business and small and medium enterprises and the DGs more concerned about environmental and social standards. As magisterially narrated by Lorenzo Allio (2008), the emergence of this specific three-fold articulation of impact assessment had more to do with how to make a deal within this complex organization (the DGs in charge of enterprise, social affairs and environmental policy agreed on the impact assessment template) than with responding to the call (of the Member States, the European Parliament, and business) for transparency and accountability.

120  Research handbook on soft law As for its organizational impact, regulatory reform in the Commission established ‘a focus for strategic and operational management within the Secretariat General’ (Radaelli and Meuwese 2010, p. 142), and a limitation of the silos mentality that prevailed until then. In the first decade of the 2000s, the Secretariat General mutated from a primus inter pares with loose coordination capacity to something like a cabinet office (Radaelli and Meuwese 2010). Among other things, this explains why in the first decade of the 2000s the profile of better regulation within the Secretariat General rose year-after-year, as did its capacity to steer the impact assessment working groups inside the Commission. Thus, the rise of regulatory reform agendas as soft law reveals power dynamics between the Commission and the other institutions as well as dynamics inside the Commission as complex organization. For a short period of time, it looked like this bundle of soft law instruments (including minimum standards for consultation and impact assessment) could be accompanied by a slightly harder tool: regulatory indicators. Here is the story. The then DG Enterprise and Industry (DG ENTR) and the Secretariat General (SECGEN) kicked off the game by publishing a tender for a study on regulatory indicators in 2003. DG ENTR and the SECGEN wanted to keep control of the objectives of better regulation – and keep track of progress both at the EU level and in the Member States. After all, the quality of regulation is decided in a multi-level context: the EU produces policies that are then transposed and implemented by national bureaucracies and regulators. This attempt to harden the soft law tools with indicators was arguably a political twist. Initially demanded by the Member States to control the Commission, in the early 2000s someone in the Commission might have reasoned that the design and implementation of regulatory indicators could have given the driver’s seat to the SECGEN instead. Some argued that, if adopted, regulatory indicators could have created a proper open method of coordination in this domain (Radaelli and De Francesco 2007; Zeitlin 2008; Tholoniat 2010). Open method of coordination is a classic in the toolbox of soft law modes of governance in the EU (see van Gerven and Stiller in this volume). Typically, facilitated coordination includes shared objectives, guidelines, the sharing of best practice, multi-lateral discussions, peer review and an iterative process. Indicators are necessary to make the objectives concrete and to measure progress. In the years 2002–05 the EU had guidelines, tools, a network of Directors and Experts of Better Regulation to host multi-lateral surveillance and peer review. What was missing was a set of indicators designed to support an open method of coordination (Radaelli and De Francesco 2007). However, this element of a fully-fledged open method did not materialize. Around 2003–04, the Commission lost the support of the UK and the Netherlands, two countries that were more interested in experimenting with tools for the reduction of administrative burdens (Radaelli 2020). Two different visions came to clash: on the one hand the broad, governance-inspired vision of the Commission. On the other, there was the Dutch and British-led de-regulatory war on red tape (Gravey 2016). The project to adopt EU-wide regulatory indicators was abandoned – although during the same decade, outside the EU, the OECD consolidated its own system of indicators of regulatory governance (Radaelli 2020). Institutions emerged as the second terrain of confrontation: who should be committed to the adoption and usage of the soft law instruments? Vertically, the tension was between the EU level and the responsibility of the Member States, and who should be accountable to whom. Horizontally, the tension involved the Commission and the other two major institutions. The

Better regulation as soft law  121 better regulation agenda was supposed to bind the Council, the European Parliament and the Commission with the 2003 inter-institutional agreement on better regulation (Interinstitutional Agreement 2003). This agreement was not implemented by the Council, and only in limited ways by the European Parliament. This signals that these two institutions were keen on tools such as impact assessment to make the Commission accountable to them, but preferred to be free to think politically when making their way through EU legislative proposals, without the constraints imposed by impact assessments of their substantive amendments. If anything, during the mid-2000s the national delegations pressed for more control on the Commission. They did so by asking for a regulatory oversight body that would check on how serious the Commission was with its own impact assessments. In the SECGEN and elsewhere there was no a priori opposition to a similar body (Radaelli 2020), given that centralized oversight of rulemaking would go hand-in-hand with centralized control of policy formulation (around the key role of the SECGEN). However, the Commission wanted this body staffed by its officers, whilst the Member States active on the better regulation agenda demanded an independent body. The result was a Commission-staffed Impact Assessment Board (IAB) – created in 2007. Much later, in 2015, the IAB turned into a Regulatory Scrutiny Board (RSB) with three members from the Commission and three temporary agents recruited externally, and a chair from the Commission ranked at the level of Director General. The 2020 decision of the Commission (European Commission 2020) emphasized that the RSB does not take political instructions (on the Board’s impact, see Senninger and Blom-Hansen 2021) and amplified its mandate to the scrutiny of foresight – a crucial element of the Next Generation plan of the EU for recovery and resiliency. In the same decade (the 2010s), the EP increased its capacity to critically appraise the impact assessments and ex-post legislative evaluations of the Commission – with a substantial strengthening of its research service (EPRS) (Radaelli 2018). In 2015 the Commission re-calibrated better regulation (European Commission 2015), always keeping the line firmly on soft law instruments – indeed the key instrument for launching the agenda was a Communication. The 2015 re-launch included the goal of closing the policy cycle with legislative evaluations made necessary before work on new impact assessments could start. There was also more investment in consultations at different stages of the law-making process. Guidance on impact assessment and ex-post legislative evaluation improved, with a proper toolbox introduced in 2015 and then re-adjusted in summer 2017. Until then, ex-post legislative evaluation was rare in the Commission, and not scrutinized by oversight bodies. Finally, as mentioned, the IAB turned into the stronger RSB, because among other things the RSB members were full-time (as opposed to the part-timers of the IAB). In terms of the interinstitutional political arena, the bone of contention was a new interinstitutional agreement negotiated in the second half of the 2010s. In the end the three major institutions finalized a new interinstitutional agreement on better law-making (Interinstitutional Agreement 2016). This agreement relaunched the 2003 agreement. Significantly, it adopts the language of law-making instead of better regulation, to indicate the ambition to streamline the evidence-base for the whole law-making process, from inception to the final agreement on the proposals. Other innovations in the last decade are yet again in the territory of soft law, with simplification carried out by a platform called REFIT – whose mandate was to check that the legislation in force was still fit for purpose.

122  Research handbook on soft law In 2021 the Commission (2021) intervened on better regulation with the classic soft law tool of a Communication, this time adding foresight, a single portal for consultation, and references to environmental sustainability. Better regulation’s importance in the Commission is evidenced by its presence in the working methods. The working methods include the presence of a mechanism of regulatory offsetting (Trnka and Thuerer 2019) called one-in-one-out. This offsetting principle affirms that any regulation introducing new burdens should relieve business and citizens of an equivalent burden existing in EU-level legislation in the same policy area. Offsetting is yet another soft law instrument. But behind the policy instrument of one-in-one-out we find yet again the politics dimension. The Competitiveness Council demanded a commitment of the Commission to limit the growth of regulatory burdens with some mechanisms of regulatory offsetting. Yet the Commission resisted the idea not once, but twice (Radaelli 2021 for the detail of the story). To accept one-in-one-out was a way to accommodate the preferences of governments that had been instrumental in delivering support for von der Leyen – without accepting an explicit de-regulation target (Radaelli 2021). Conceptually, after Covid-19, the suspension of the stability and growth pact, the green deal and the Next Generation EU, one-in-one-out should not be the compass of better regulation. This cannot be the major innovation that the European economy and citizens want from ‘better law-making’. One-in-one-out is a narrow priority when compared with the political priorities of the moment. All the post-pandemic political priorities of the EU institutions are geared towards delivering welfare, hence they are net benefits-oriented. But what has all this activity brought about in the Member States? How is the bundle of soft law instruments of better regulation as developed at the supranational level mirrored at the national level?

4.

THE STATE OF PLAY IN THE MEMBER STATES

We now turn to the Member States level and the UK. It is impossible to examine both the politics and policy dimension for 27 Member States (and the UK), but we are interested in examining how better regulation has been established at the domestic level. Operationally, we trace the soft and hard law characteristics of regulatory reform by focusing on two pivotal policy instruments: IA and stakeholders consultation. We draw on information and OECD data (OECD 2019), as well as data from the European Research Council-funded project Protego, Procedural Tools for Effective Governance (http://​protego​-erc​.eu/​[viewed 12 June 2023]). While the supranational level is dominated by soft law, in the Member States we find a combination of soft and hard. But that is not the whole story. What emerges from the overview of the distribution of both IA and consultation practices is that better regulation was adopted in different times and different ways, thus signalling a rather wide range of political choices made in the Member States. As will become clear in the remainder, the evidence on Member States points to a subtle difference – beyond the choice for soft or hard law. This difference lies in the high degree of formalization of the better regulation agenda of the Commission (and, for that matter, the OECD too). The Commission has adopted soft law in order to create a web of rules that govern every single feature of IA and consultation. The Member States are divided between those who have put down on formal guidance, page after page, and set in hard law the requirements for consultation and IA, and those who have a more informal style of con-

Better regulation as soft law  123 Table 8.1

Hard and soft legal basis for IA and consultation IA Hard Law

IA Soft Law

Consultation

Austria , Belgium , Croatia, Cyprus, Czech

EU, Ireland, UK

Soft Law

Republic , Denmark , France , Italy, Malta, Poland, Sweden

Consultation

Bulgaria, Estonia, Finland, Germany, Greece,

Hard Law

Hungary, Lithuania, Netherlands, Portugal,

Latvia, Luxembourg

Romania, Slovenia, Slovakia, Spain

Source: Our elaboration of Protego data (underlined countries have no legal basis at all for consultation).

sulting and appraising proposals for primary and secondary legislation. Some Member States in fact do not have a legal basis for consultation thus signalling a high degree of informality of consultation practices. Better regulation, we maintain, reflects domestic preferences and legal traditions rather than supranational models. Our cross tabulation (Table 8.1 above) takes into account all the Member States (and the UK) and the EU as a stand-alone case. Specifically, we cross tabulated data on IA and consultation. One caveat applies to countries that do not have a legal basis for consultation at all, neither hard nor soft. For the purpose of providing a general overview, countries with no legal basis for consultation fall in the soft law clusters (these cases are underlined in the table). We start by looking at the two clusters of countries in which one instrument is established in hard law and the other in soft law. The majority of countries fall in the cluster where IA is set in hard law and consultation in soft law. Contrarily, it is uncommon to observe cases with a combination where consultation is set in hard law and IA in soft law, this is perhaps because consultation is the most informal tool of better regulation. This is specifically the case of the EU – not surprisingly as we have showed in the previous sections – but also of Ireland and UK. The most-populated cluster is the one with countries where both IA and consultation are set in hard law. Table 8.1 also highlights that only few countries have a soft legal basis for both IA and consultation. This last observation points immediately to a striking difference between the way Member States have established better regulation tools vis-à-vis the supranational agenda promoting better regulation as soft law. Member States have put down formal requirements and established hard legal acts for better regulation’s tools. 4.1

Stakeholders’ Consultation

When looking at the single tools, we find evidence suggesting a variety of choices made by the Member States. Let us start with stakeholders consultation (Table 8.2 below) – like in the case of the EU in the previous section, we do not cover other types of consultation, such as consultation by parliamentary committees. We consider the dimension of publicity of the comments received and reporting on the consultation exercise. These aspects highlight a certain level of formality in the way governments and regulators carry out consultation. In fact, despite having hard or soft law, consultation procedures can be more or less detailed by setting a number of rules to be followed. A first observation regards those countries that do not have a legal basis for consultation. Among those countries we find Denmark and Sweden. They do indeed have consultation traditions and practices, anchored in hearings and consultative committees for the preparation of new legislation. The absence of a formal procedure indicates the difference between the

124  Research handbook on soft law formalized and proceduralized approach of the Commission (where consultation guidance is highly formal, with dozens of rules in the better regulation toolbox) and the informal, agile approach of Denmark and Sweden. When we turn to whether comments are made public and the publication of a consultation report, we find that these two dimensions signal a certain formality of consultation, whether it is established by soft or hard law. Even when comments are not made public and a final report on consultation is not published, a systematic way to engage with stakeholders may still emerge. Yet again, Denmark signals the distance between informal practice and the formal approach. In Denmark there is high stakeholders’ engagement, but the system relies on trust between government and external stakeholders, not on formal procedure (OECD 2010). In this sense, consultation mainly relies on informality even if some elements of formality exist, such as the online guide on procedures for the development of regulations (OECD 2010). Informality does not seem to affect the level of transparency of the process. In the OECD Indicators of Regulatory Policy and Governance (iReg) (Arndt et al. 2015) Denmark is different from Czech Republic even if in both countries there is no systematic reporting on consultation and comments are made public. The two countries in fact have very different scores in terms of iReg Transparency index. The Czech Republic is below the OECD 2018 average in terms of transparency, while Denmark is on the OECD average. A possible explanation lies in the fact that Denmark complements the lack of published consultation information and data with a strong informal engagement with stakeholders, whilst the Czech Republic does not. Finally, looking at timing, consultation procedures have a long history only in a handful of countries, indicating that they (together with the Commission) were early pioneers. Here we cannot be sure that there is an effect of the Commission on the diffusion of consultation. If anything, existing research suggests that the major effect on the adoption of formal procedures of consultation comes from the OECD rather than the EU (De Francesco 2012). 4.2

Impact Assessment

A significant difference characterizes the scope and breadth of IA (Table 8.3 below). In Ireland, Luxembourg and Malta IA on secondary legislation is either just a checklist (Luxembourg) or is performed only on major pieces of legislation (Ireland) or just on SMEs, as a test (Malta, where stakeholders consultation is a discretionary activity left to individual departments). IA has evolved over time: the revision of guidance is not an exception. Arguably, to revise the IA discipline is an indicator of vitality and good practice (e.g., the EU-OECD SIGMA initiative has recommended this to the Western Balkans, Tunyan 2021). In some cases, IA was adopted with a soft legal basis first and only then hardened. In Hungary and Lithuania, for instance, soft legal bases were first adopted, and only later were the guidance hardened. By contrast, in Finland, France and Poland IA was first grounded in hard law, thus establishing a general mandatory duty, and then followed by specific guidance grounded in soft law. The column on time reveals an interesting perspective on the presence of this specific better regulation tool before the advent of the international pressure for better regulation. Countries like Sweden had elements of IA before the international and EU agenda for better regulation was set. In the Netherlands IA was established by law in 1992, at the same time when the country was among those that indicated the need to simplify EU legislation. The vast majority of countries, however, adopted a legal basis – hard or soft– after the 1995 OECD Council

Better regulation as soft law  125 Table 8.2

Consultation

Countries

Are comments made public?

A consultation report is published Year in which the legal basis was

Austria

No

No

/

Belgium

No

No

/

Bulgaria

Yes

Yes

2007

Croatia

Yes

Yes

2009

Cyprus

No– not systematic

Yes

2009

Czech Republic

No

No

1998

Denmark

No

No

/

Estonia

Yes

Yes

2011

European Union

Yes

Yes

/

Finland

Yes – the summary of comments

Yes

2010

France

No

No

/

Germany

Yes

Yes

1979, 2000

Greece

Yes

Yes

2012

Hungary

Yes

Yes

2010

Ireland

Yes

Yes

2009

Italy

No

No

2017

Latvia

No

Yes

2009

Lithuania

Yes

Yes

1996

Luxembourg

No

No

1978/79

Malta

Yes

Yes

2011

Netherlands

No

No

1994

Poland

No

Yes

2013

Portugal

Yes- – a record

No

1976, 2009, 2015

Romania

Yes

Yes

2003

Slovakia

No

Yes

2016

Slovenia

Yes

Yes

1991 Constitution, 2003

Spain

Yes – a record

Yes

1997, 2015

Sweden

No

No

/

UK

Yes

Yes

2012

introduced

Source: Our elaboration of Protego and OECD 2019 data.

Recommendations (OECD 1995) and the first emergence, in 2002–05, of the EU approach (Allio 2008; De Francesco 2012). A crucial element of better regulation agendas is the presence of a Regulatory Oversight Body (ROB). Regulatory oversight means among other things that there is a body in charge of keeping track of how consultation and IA are carried out. In some countries, the function of this body extends to returning to departments IAs that do not match the standards imposed by the government. Hence, whether the standards for IA and stakeholder’s engagement are set in hard or soft law, the presence of a ROB suggests commitment to enforce the standards. If the ROB does not exist, the oversight and scrutiny functions of IA may still be carried out by a single department or by a unit in central government (such as a better regulation unit in the cabinet office). The political and administrative costs of implementing better regulation from the adoption of IA guidance to its implementation, down to the presence of a strong central unit (independent de jure or de facto) tend to increase as implementation deepens, as shown by previous studies (De Francesco et al. 2012).

126  Research handbook on soft law Two indicators in the Protego dataset illustrate whether the ROB publishes its opinion on IAs and the formal de jure independence of the body. Although what matters is the de facto independence, some governments make ROBs independent by design to increase the credibility of their commitment. Only few countries have set up an independent body that is composed of members coming from outside the administration. But – unsurprisingly perhaps – these Member States belong to the set that is most active in the debate on better regulation. The ROB’s opinions on the quality of IA are published only in Austria, Czech Republic, EU, France, Finland, Italy, Sweden, and the UK. In the Netherlands, the ROB does not issue a formal opinion for every case, but just on specific ones. Conversely, in Lithuania the main regulatory oversight body (the Office of the Government) provides its opinion only to the Prime Minister and the Chancellor of the Government. A final observation on the independence of the ROBs and the establishment of IA and consultation in hard law: a stronger and independent control over IAs is not automatically associated with hard-based IA and consultation (Table 8.3, columns 1 and 3).

5. CONCLUSIONS To wrap up, EU-level better regulation is soft law and applies to EU law-making. It is a kind of meta-regulation, that is, rules about making rules (Radaelli 2010). The EU does not have the legal and political means to produce directives or regulations on how the Member States should go about regulatory reform – although the Commission (and the Regulatory Scrutiny Board) correctly insists on the fact that to succeed, better regulation ought to be a joint commitment of Member States and EU institutions. This choice for soft law chimes with everything we know about soft law being easier to adopt as well as more respectful of Member State sovereignty (for the literature, see Saurugger and Terpan 2021). Soft law has its peculiar properties when applied to better regulation. We demonstrated that this type of soft law, presented as win-win as well as positively accepted by the majority of Member States, has become the field where EU institutions measure their power to control the law-making process. The Council wants to deploy better regulation to make the Commission and the whole law-making process more intergovernmental. The Commission, on the other hand, tells Member States as well as accessing countries that better regulation is not only a set of guidance and recommendations, but also an integral part of the soft acquis. Inside the Commission, better regulation measures the power of the Secretariat General in relation to the DGs. At the domestic level we observe significant differences, also because breadth and scope of IA and consultation vary considerably. IA in some countries covers only specific types of assessments and legislation (e.g., it may apply to primary but not secondary legislation, OECD 2019), in others IA goes deeper. The UK and EU Member States have chosen both soft and hard legal bases, with the UK championing the soft law approach. This is coherent with the Westminster model, where political commitment at the ministerial level is sufficient to embed ways of doing things in departmental modus operandi. We also observe a mix of formality (be it through either soft or hard law) and informality (in countries like Denmark and Sweden). This yet again confirms our knowledge of the administrative and law-making process in these two countries, and the role of IA and consultation therein (Radaelli 2009).

Better regulation as soft law  127 Table 8.3 Countries

Impact assessment Legal basis is grounded Year in which the legal Independent ROB

ROB publishes its opinion

in soft or hard law

on IA

basis was introduced and year(s) of major changes

Austria

Hard and Soft

2013

No

Yes

Belgium

Hard

2013

No

No (not mandatory)

Bulgaria

Hard and Soft

2016, 2017

ROB does not exist

ROB does not exist

Croatia

Soft and then Hard law 2012, 2017

No

No

Cyprus

Hard and Soft

2016

No

No

Czech Republic

Hard and Soft

2011

Yes

Yes

Denmark

Hard

2005

ROB does not exist

ROB does not exist

Estonia

Hard and Soft law

2011, 2012

ROB does not exist

ROB does not exist

European Union

Soft Law

2002, 2017 guidelines. Yes, but only half of the

Finland

Hard and Soft law

Yes

Toolbox 2015

members are external. Chaired

(currently under

by a high-level officer of the

revision)

Commission

Hard 2015, Soft 2017,

Yes

Yes

No

Yes

Yes

No

No

No

ROB does not exist

ROB does not exist

ROB does not exist

ROB does not exist

Introduced 1999,

ROB does not exist – experts

ROB does not exist

changes with primary

are recruited from outside the

legislation 2005, 2018

PA but the body sits within

2007 on IA and 2013 on legislative drafting France

Hard and Soft law

2015 hard law, 2017 soft law

Germany

Hard and Soft law

In the 1990s first attempts with the blue checklist on the quality of legislation. 2006, 2009 working manual on RIA, 1992, 2012 handbook on preparation of law

Greece

Hard and Soft law

2012 key legal text on IA, 2006 Prime Minister circular explaining IA

Hungary

Hard and Soft law

2011, 2016 regulation on IA, guidelines on different types of IA 2010

Ireland

Soft law

2009, 2016 revised guidelines

Italy

Hard Law

decree introduces IA on the department for legislative EU regulations

activity

Latvia

Soft Law

2009

ROB does not exist

ROB does not exist

Lithuania

Hard and Soft Law

2012 hard law, 2003

No

No

ROB does not exist

ROB does not exist

soft law Luxembourg

Soft law

2011

128  Research handbook on soft law Countries

Legal basis is grounded Year in which the legal Independent ROB

ROB publishes its opinion

in soft or hard law

on IA

basis was introduced and year(s) of major changes

Malta

Hard law (only SMEs

2011, 2015

ROB does not exist

ROB does not exist

Hard law 1992/2017.

Yes

No (annual report)

ROB does not exist

ROB does not exist

test) Netherlands

Hard and Soft law

Soft law 2011 Poland

Hard and Soft law

2011, 2014. In 2015 guidelines as part of government legislative process are issued by the Ministry of Economy

Portugal

Hard and Soft law

2017

No

No

Romania

Hard law

2000, 2005, 2009

ROB does not exist

ROB does not exist

Slovakia

Hard and Soft law

2015, 2016

ROB does not exist

ROB does not exist

Slovenia

Hard and Soft law

2001, 2010, 2017

ROB does not exist

ROB does not exist

Spain

Hard and Soft law

1997, 2015, 2017

ROB does not exist

ROB does not exist

Sweden

Hard law and

1974, 2003, 2010

Yes

Yes

Guidelines 2017.

Yes

Yes

ordinances UK

Soft law

Principles 2018, BR framework 2018, Green Book on evaluation 2018

Source: Our elaboration of Protego data and OECD 2019.

Legal-administrative traditions and political priorities have left their mark on the establishment of better regulation in the Member States. It is telling that a country like Italy introduced IA with a law under the assumption, dominant in this country, that if law-making procedures are not embedded in formal legislation, they will not bite. This finding contributes to the vast literature on the efficacy of soft law in specific policy fields (see Slominski 2012; Bekker 2014 on ambiguities, but also Trubek and Trubek 2005 on soft law as a producer of ‘cheap talk’). The remarks on the UK and Italy may be generalized. Assumptions and prevailing views in government circles about the efficiency of soft law might not only explain the differences in legal bases of better regulation’s tools, but also the processes of hardening or softening over time. This conjecture could be further explored thus making future contribution to the literature on soft law and the conditions that underpin the efficacy of better regulation as soft law and meta-regulation.

ACKNOWLEDGMENTS AND NOTE We wish to acknowledge the support of the European Research Council, Project on Procedural Tools for Effective Governance grant number 694632. Website: protego-erc.eu. Protego data were collected in the period 2018-2019; they cover the EU-27, the UK, and the EU as stan-

Better regulation as soft law  129 dalone case. The Protego data were gathered on the legal base in force in 2018 by a team of lawyers using purpose-built protocols.

REFERENCES Abeele, E. van den, (2015). ‘Better regulation’: a bureaucratic simplification with a political agenda. ETUI Working Paper, 2015.04 [online]. Brussels: European Trade Union Institution (ETUI). [Viewed 29 November 2021]. Available from: doi: 10.2139/ssrn.2628067. Allio, L., (2008). The emergence of better regulation in the European Union. PhD thesis. King’s College London (University of London). Arndt, C., Baker, A. C., Querbach, T. and Schultz, R., (2015). 2015 indicators of regulatory policy and governance: design, methodology and key results. OECD Regulatory Policy Working Papers, 1. Paris: OECD Publishing. Baldwin, R. and Cave M., (1999). Understanding regulation: theory, strategy, and practice. Oxford: Oxford University Press. Bekker, S., (2014). EU coordination of welfare states after the crisis: further interconnecting soft and hard law. International Review of Public Administration. 19(3), 296–307. De Francesco, F., (2012). Diffusion of regulatory impact analysis in OECD and EU Member States. Comparative Political Studies. 45(10), 1277–1305. De Francesco, F., Radaelli, C. M. and Troeger, V. E., (2012). Implementing regulatory innovations in Europe: the case of impact assessment. Journal of European Public Policy [online]. 19(4), 491–511. [Viewed 24 November 2021]. Available from: doi: 10.1080/13501763.2011.607342. Doern, B., (2007). Red tape, red flags: regulation for the innovation age. Ottawa: Conference Board of Canada. European Commission., (2015). Better regulation agenda: enhancing transparency and scrutiny for better EU law-making [online]. Press release, IP/15/4988. Strasbourg: European Commission. [Viewed 26 November 2021]. Available from: https://​ec​.europa​.eu/​commission/​presscorner/​detail/​en/​IP​_15​ _4988. European Commission., (2020). Decision of the President of the European Commission on an independent Regulatory Scrutiny Board (P(2020)2). Brussels: European Commission. European Commission., (2021). Communication. better regulation: joining forces to make better laws (COM/2021/219 final) [online]. Brussels: European Commission. [Viewed 24 November 2021]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​?uri​=​CELEX:​52021DC0219. Garben, S., (2018). An ‘impact assessment’ of EU better regulation. In: S. Garben and I. Govaere, eds. The EU better regulation agenda: a critical perspective. Oxford: Hart Publishing. pp. 217–242. Garben, S. and Govaere, I., eds., (2018). The EU better regulation agenda: a critical perspective. Oxford: Hart Publishing. Gravey, V., (2016). Does the European Union have a reverse gear? Environmental policy dismantling, 1992-2014. PhD thesis [online]. School of Environmental Sciences, University of East Anglia. [Viewed 24 November 2021]. Available from: https://​core​.ac​.uk/​download/​pdf/​41994932​.pdf. Gunningham, N., Grabosky, P. and Sinclair, D., (1998). Smart regulation: designing environmental policy. Oxford: Oxford University Press. Interinstitutional agreement 2003/C 321/01 between the European Parliament, the Council of the European Union and the Commission of the European Union on better law-making (2003). Official Journal C 321, p. 1. Interinstitutional agreement between the European Parliament, the Council of the European Union, and the European Commission on better law-making (2016). Official Journal L 123, p. 1. Kysar, D. A., (2010). Regulating from nowhere: environmental law and the search for objectivity. New Haven, C.T.: Yale University Press. Lodge, M. and Wegrich, K., (2012). Managing regulation: regulatory analysis, politics and policy. Basingstoke, Hampshire: Palgrave Macmillan. Lowi, T. J., (1972). Four systems of policy, politics, and choice. Public Administration Review. 32(4), 298–310.

130  Research handbook on soft law Mandelkern Report., (2001). Mandelkern Group on Better Regulation. Final Report 13 November 2001 [online]. Lisbon: European Council. [Viewed 25 November 2021]. Available from: https://​www​ .smartreg​.pe/​reportes/​Mandelkern​%20Report​%20on​%20Better​%20Regulation​%202001​.pdf. OECD., (1995). Recommendation of the Council of the OECD on improving the quality of government regulation (adopted on 9 March 1995) [online]. Paris: OECD. [Viewed 25 November 2021]. Available from: https://​www​.oecd​.org/​officialdocuments/​publi​cdisplaydo​cumentpdf/​?doclanguage​=​ en​&​cote​=​OCDE/​GD(95)95. OECD., (2010). Better regulation in Europe [online]. Paris: OECD. [Viewed 25 November 2021]. Available from: doi: 10.1787/20790368. OECD., (2012). Recommendation of the Council on regulatory policy and governance [online]. Paris: OECD Publishing. [Viewed 25 November 2021]. Available from: doi: 10.1787/9789264209022-en. OECD., (2019). Better regulation practices across the European Union [online]. Paris: OECD Publishing. [Viewed 25 November 2021]. Available from: doi: 10.1787/9789264311732-en. Radaelli, C. M., (1999). Steering the Community regulatory system: the challenges ahead. Public Administration. 77(4), 855–871. Radaelli, C. M., (2009). Desperately seeking regulatory impact assessments: diary of a reflective researcher. Evaluation. 15(1), 31–48. Radaelli, C. M., (2010). Regulating rule-making via impact assessment. Governance. 23(1): 89–108. Radaelli, C. M., (2018). Halfway through the better regulation strategy of the Juncker Commission: what does the evidence say? JCMS: Journal of Common Market Studies. 56(S1), 85–95. Radaelli, C. M., (2020). Regulatory indicators in the European Union and the Organization for Economic Cooperation and Development: performance assessment, organizational processes, and learning. Public Policy and Administration. 35(3), 227–246. Radaelli, C. M., (2021). The state of play with the better regulation agenda of the European Commission. European University Institute, School of Transnational Governance, STG Policy Papers [online]. 2021/06. [Viewed 25 November 2021]. Available from: https://​cadmus​.eui​.eu/​bitstream/​handle/​1814/​ 70901/​PB​_Radaelli​_2021​_06​.pdf. Radaelli, C. M., (2023). Occupy the semantic space! Opening up the language of better regulation. Journal of European Public Policy. 30(9), 1860-1883. Radaelli, C. M. and De Francesco, F., (2007). Regulatory quality in Europe: concepts, measures, and policy processes. Manchester: Manchester University Press. Radaelli, C. M. and Meuwese, A. C. M., (2010). Hard questions, hard solutions: proceduralisation through impact assessment in the EU. West European Politics. 33(1), 136–153. Saurugger, S. and Terpan, F., (2021). Normative transformations in the European Union: on hardening and softening law. West European Politics. 44(1), 1–20. Schömann, I., (2015). EU REFIT machinery ‘cutting red tape’ at the cost of the acquis communautaire. ETUI Policy Brief No. 5/2015 [online]. [Viewed 25 November 2021]. Available from: https://​www​ .etui​.org/​sites/​default/​files/​Policy​%20Brief​%202015​.05​%20Sch​%C3​%B6mann​.pdf. Scott, C., (2003). Speaking softly without big sticks: meta-regulation and public sector audit. Law & Policy [online]. 25(3), 203–219. [Viewed 25 November 2021]. Available from: doi: 10.1111/j.14679930.2003.00148.x. Scott, C., (2010). Reflexive governance, regulation, and meta-regulation: control or learning?. In: J. Lenoble and O. De Schutter, eds. Reflexive governance: redefining the public interest in a pluralistic world. Oxford: Hart Publishing. pp. 43–66. Senninger, R. and Blom-Hansen, J., (2021). Meet the critics: analyzing the EU Commission’s Regulatory Scrutiny Board through quantitative text analysis. Regulation & Governance [online]. 15(4), 1436–1453. [Viewed 25 November 2021]. Available from: doi: 10.1111/rego.12312. Slominski, P., (2012). The ambiguities of legalization and the EU’s strategy of extraterritorial border control. European Foreign Affairs Review. 17(2), 19–34. Tholoniat, L., (2010). The career of the open method of coordination: lessons from a ‘soft’ EU instrument. West European Politics. 33(1), 93–117. Trnka, D. and Thuerer, Y., (2019). One-in, x-out: regulatory offsetting in selected OECD countries. OECD Regulatory Policy Working Papers, 11 [online]. Paris: OECD Publishing. [Viewed 25 November 2021]. Available from: doi: 10.1787/67d71764-en.

Better regulation as soft law  131 Trubek, D. M. and Trubek, L. G., (2005). Hard and soft law in the construction of social Europe: the role of the open method of co-ordination: European Law Journal. 11(3), 343–364. Tunyan, B., (2021). Regulatory impact assessment and EU law transposition in the Western Balkans: a comparative analysis of the practice of ex ante assessment of regulatory proposals and EU law transposition [online]. SIGMA Papers, No.61. Paris: OECD Publishing. [Viewed 25 November 2021]. Available from: doi: 10.1787/2cbdb615-en. World Bank., (2010). Better regulation for growth: regulatory governance in developing countries [online]. Washington, D.C.: The World Bank Group. [Viewed 26 November 2021]. Available from: http://​hdl​.handle​.net/​10986/​27881. Zeitlin, J., (2008). The open method of co-ordination and the governance of the Lisbon Strategy. Journal of Common Market Studies. 46(2), 436–450.

9. Rules and responsibilities: business and social norms in transnational governance Boris Holzer

1. INTRODUCTION According to Lindblom (1977), governments have ‘thumbs’ but no ‘fingers’. They can make rules to encourage some and discourage other economic activities, but in a market system there is no guarantee that this will produce the desired outcomes. Although governments can exert power through legislation, their control over the economic consequences of their decisions remains limited (Luhmann 1988). Any political intervention may have unintended consequences such as increasing unemployment or decreasing economic growth. Lindblom therefore regards the market as a ‘prison’, as ‘an extraordinary system for repressing change’ (Lindblom 1982, p. 326). This structural weakness of the state is brought into sharper relief by economic globalization: Transnational economic transactions and global financial markets are increasingly beyond the control of the nation-state. If regulation imposes new costs on economic actors, they will try to avoid them. Firms do not regard legal norms as ‘precise normative commands requiring obedience’ but interpret them ‘according to the inner logic of the concrete market and the concrete organization’ (Teubner 1994, p. 33; see also Stone 1975). From the perspective of a managing director of a transnational corporation, the resulting division of labour is straightforward: ‘It is the job of governments to make the rules, and ours to find the loopholes’ (Tugendhat 1973, p. 163). The world economy not only offers many loopholes but also a great variety of regulatory regimes, which enables transnational corporations (TNCs) to organize their supply chains to exploit wage inequalities as well as regulatory differences. Despite international conventions, there are only few binding standards addressing the problem that social and environmental standards are often lower in countries of the Global South (Galbraith 2002). The governance gap on a transnational level is to some extent addressed by norms and standards that are constructed ‘from below’ – by non-state actors who devise codes of conduct, engage in stakeholder dialogue, and monitor certification schemes. Such legally non-binding forms of regulation are advocated by transnational activists who can muster the support of consumers in the Global North who demand consistent and fair rules in transnational economic transactions. While state regulation cannot prevent corporations from exploiting the variety and limits of regulatory regimes, other actors may challenge corporations to meet broader moral expectations. Nongovernmental organizations (NGOs) and transnational activists do not have the necessary ‘thumbs’ to enforce rules concerning environmental protection, labour, and human rights. Instead, they have ‘fingers’ to point at corporations which violate such rules. Activists and protest groups call attention to corporate misconduct through ‘naming and shaming’-campaigns which have already targeted the operations of major TNCs and their suppliers, for instance Nike, Nestlé, Shell, and Starbucks (Greenberg and Knight 2004; Holzer 2007; 132

Rules and responsibilities  133 Bair and Palpacuer 2012; Bartley and Child 2014). In the wake of those and other successful anti-corporate campaigns, public criticism and scandalization have become recognized as potential constraints of corporate decision-making. These constraints are of a normative but not of a legal nature: They are based on public claims that articulate rules that corporations are supposed to adhere to. More often than not, these rules only become visible when activists decry that they have been neglected or violated. Engaging with such criticism, corporations have to cope with its normative expectations to minimize the risk of public conflicts about their responsibility. This chapter examines the dynamics of rulemaking, accountability, and blame between TNCs and transnational political activists. On the one hand, many companies have adopted voluntary corporate social responsibility (CSR) policies as well as methods of social and environmental reporting – often in reaction to the reputational risks inherent in doing business across political and legal borders. On the other hand, NGOs constantly observe corporations and use perceived misdemeanours to point out shortcomings of regulation. In some cases, some sort of ‘hybrid’ regulation emerges in which both private and state actors participate in the promulgation, monitoring, and certification of rules (Bair 2017; Bartley 2018). In other cases, the application of rules is less systematic and more conflictual. The chapter shows how such interactions result in the elaboration of a normative framework for corporate decision-making that complements state regulation and interacts with it. Comprising a broad and sometimes fuzzy set of social norms, it includes but is not limited to the rules of transnational ‘soft law’ that are considered binding but lack enforceability (Abbott and Snidal 2000; Shaffer and Pollack 2010). The normative framework of world society also encompasses social norms that are neither legally binding nor formally codified but nonetheless inform the public’s appraisal of business practices. The public scrutiny of corporate behaviour enacts and articulates such social norms, for instance through public debates about the rights and duties of corporate actors. The participation in and the strategic anticipation of such conflicts are important drivers of corporate responsibility and rule-following.

2.

RESPONSIBLE CORPORATIONS

Rules and responsibility go together: Without rules, the question of who is responsible for mistakes and wrongdoings does not arise. Without responsibility, rules cannot be monitored and enforced because no one can be held accountable. Yet it remains contested how much responsibility of economic organizations is necessary and desirable. Economist Milton Friedman (1970) famously claimed that ‘the social responsibility of business is to increase its profits’. From an economic point of view and from the one of shareholders in particular, that statement expresses a mandate that firms obviously have to pursue. They are entrusted by their owners to make profits or to create ‘shareholder value’ (Rappaport 1986). A more far-reaching definition of corporate responsibility would not only threaten the legitimate interests of owners but could also interfere with the responsibilities of others. Without a proper mandate, business leaders should not strive to champion or support values, or provide welfare services which in principle are the responsibility of the state (Levitt 1958). However, such arguments do not prevent other groups from having their own perspectives that let them attribute other, additional responsibilities to corporations. Companies and other organizations may be held responsible for a lot of issues beyond their economic performance – for instance

134  Research handbook on soft law for harm that their decisions may inflict on other people. The pursuit of profit and other organizational objectives is therefore subject to rules both ‘hard’ (based on legislation) and ‘soft’ (based on social norms and expectations). As a reaction to the increasing resonance of such demands, the CSR concept aims to reconcile corporate objectives and societal expectations. Instead of merely complying with legal obligations, corporations try to ‘meet the ethical gaze’ (Crane 2005) by adapting their organizational cultures (Sorell and Hendry 1994; Carmichael 1995; Weiss 1998). Adopting ethical principles to guide their decisions, firms may be compelled to anticipate and consider the harmful side-effects of their operations and also to contribute to societal welfare. While some theorists advocate a fairly limited set of basic moral principles (Sternberg 1994), others favour the wide-ranging incorporation of non-economic criteria and values into the business agenda (Jones and Pollitt 1998).1 A broad interpretation of the social and political responsibilities of business is also associated with the notion of corporate ‘citizenship’, which casts the corporation as a member of a community: Corporations should behave as responsible ‘corporate citizens’ (Zadek 2001; Matten and Crane 2005; Crane et al. 2008). How to become such a citizen is not only the subject of a growing body of literature and expertise (Meyer et al. 2015), but also a matter of practical concern for transnational corporations (Sklair 2001, pp. 149–197). The ‘Milton Friedman view of the world’ (Mitroff 1983, p. 19) is not necessarily a good guide for corporate decision-making. Ownership of a modern public enterprise is widely dispersed and often exercised indirectly, for instance through investment funds. Stockholders may therefore be understood in analogy to citizens, whose demands constitute only ‘loose constraints’ on the actual policies pursued on their behalf (March 1962, p. 674). Others suggest that the relationship between shareholders and a stock-listed company is similar to the one between a bank providing a mortgage and a borrower (Handy 1996): Shareholders may expect management to provide a reasonable return on investment but this need not be interpreted in terms of profit maximization. The obligations of a corporation towards its shareholders are not as clearly defined as the term ‘shareholder value’ sometimes suggests (Marens and Wicks 1999). In fact, any firm is the target of manifold demands and expectations from a variety of groups within and outside the organization. Not only shareholders but different ‘stakeholders’ are affected by corporate performance and may therefore demand that their interests are taken into account.2 There are both ethical and strategic reasons to monitor and manage stakeholder interests (Donaldson and Preston 1995; Weiss 1998). From an ethical perspective, ‘being affected’ justifies the representation of the concerns of such groups in organizational decision-making, for instance by community dialogue and participation processes (Ulrich 1998, pp. 438–449). From a strategic perspective, the potential significance of such groups on business performance cannot be neglected. The ‘reverse-impact model’ (Brummer 1991, p. 38) points out that, regardless of the legitimacy of the interests concerned, they affect economic success and therefore need to be taken into account. Corporate responsibility may be the result of ‘enlightened self-interest’ (Mintzberg 1983, p. 4) based on various strategic considerations: First, the business community as a whole might benefit from socially responsible behaviour. If corporations contribute to the mitigation of social problems or invest in education, they will directly or indirectly benefit in the middle or long run. Second, theories of ‘sound investment’ predict that the stock market will react positively to social responsibility. Since many institutional investors view irresponsible firms (with the concomitant bad reputation) as bad investment. Third, CSR may serve to avoid

Rules and responsibilities  135 outside interference. If corporations demonstrate that they behave in a responsible manner without extensive regulation, they may be able to exert more control over their operations in the future. CSR-based voluntary self-regulation is also promoted by international organizations. For instance, the United Nations’ ‘Global Compact’ initiative calls on companies to respect and protect human and labour rights by subscribing to ten guiding principles (Williams 2004). In some jurisdictions, fundamental principles of comprehensive reporting are already mandatory. For instance, the European Union’s non-financial reporting directive (2014/95/EU) requires EU-based companies with more than 500 employees to report on the impact of their activities on society and the environment. Against this background, the business community has increasingly accepted that a certain commitment to social responsibility is a prerequisite for sustainable profit-making. ‘“Codes of conduct” and “business principles” reflect the insight that added legitimacy can be earned by adopting a policy of “corporate responsibility”’ (Sikkink 1986; Hopkins 1999, pp. 53–70). Some business ethics scholars argue that a truly ‘moral point of view’ must give ethical considerations absolute priority (Ulrich 1998). There are indeed examples of firms according a high priority to ethical criteria, such as the retail cosmetics chain The Body Shop, one of the ‘icons of socially responsible business’ (Jennings and Entine 1998). Founded in 1976 by British entrepreneur Anita Roddick, the company specializes in naturally-based cosmetics products. The company’s philosophy is based on environmental principles, including the rejection of any animal testing and an emphasis on recycling and minimal, environmentally friendly packaging. In so doing, The Body Shop and similar ethically-minded firms seek to comply with a particular set of expectations and values. There certainly are audiences and markets that appreciate such endeavours. But even value orientations, such as environmental and animal protection, are not shared universally. ‘Ethical’ business practice therefore is limited to certain products and services. Firms that do not aim to stand out on account of their ethical performance or cannot afford to do so are confronted with a different situation. For them it makes sense to be responsible to a certain extent – but not more than that. Empirical studies of the correlation of ethical policies and corporate performance show that the marginal utility of being ethical decreases: ‘It pays to be good, but not too good’ (Mintzberg 1983, p. 7). Business firms have to base their decisions on economic, i.e., financial, considerations. But the profit-making motive alone is too general to determine complex decisions.3 Other criteria and values always play a role in everyday organizational practice. Each individual company – unlike the economy as a whole – is free, and sometimes compelled, to consider other aspects than mere economic ones in order to make its decisions, e.g., the environmental or social effects of its operations and investments (Elkington 1997). In doing so, corporations have to take into account the expectations not only of corporate critics but also of customers and consumers since environmental, ethical, and political values inform lifestyles and consumer choices (Gabriel and Lang 1995, p. 164; Cohen 1998). If an organization cannot take for granted that others agree with its objectives, consideration for a variety of group interests must be demonstrated in order to earn the social ‘license to operate’. Hence corporations are forced to anticipate or integrate movements’ and critics’ views into their decision-making processes, for instance by round tables, corporate dialogues, and scenario groups – both to boost the legitimacy of their operations and to protect their profitability (van Es and Meijlink 2000). Multiple and differentiated audiences and stakeholders articulate different and sometimes conflicting demands. Stakeholder engagement therefore does not imply the adaption to some

136  Research handbook on soft law form of societal consensus. Instead, it reflects the differences between the organization and its social environment as well as the differentiated demands within that environment. Organizations do not possess a moral ‘consciousness’ to guide their decisions about such differentiated and sometimes conflicting demands. Even the consciousness of responsible managers does not result in responsible behaviour at the organizational level – ‘having a conscience in the running of a large corporation does not translate automatically into running a conscientious corporation’ (Goodpaster 1983, p. 305). Whether being good pays off or whether behaving badly is costly depends on external constraints, i.e., on laws and potential sanctions, on protest actions, and on the decisions of owners and investors. Accordingly, the scope of corporate responsibility is not determined by the corporation alone but is also a matter of social control and attribution.

3.

THE ATTRIBUTION OF RESPONSIBILITY

The dominant conception of responsibility in the CSR debate is a voluntary one: Responsibility is conceived as a stance taken by business in the face of external demands, including but not limited to the responsibility towards owners or shareholders. Supplementing and driving voluntary forms of responsibility favoured by CSR policies, there are external sources of responsibility: on the one hand, the more specified and circumscribed, ‘hard’ rules of legal responsibility, for instance in terms of liability; on the other hand, the much broader and ‘soft’ rules of publicly attributed responsibility, for instance, if corporations are publicly held accountable for accidents or scandals. There is little doubt that business has to confirm with ‘hard’ law that codifies behavioural constraints and the extent of liability. Such rules are tied to the threat of sanctions to enforce them. ‘Soft’ law, for instance codes of conducts promoted by international organizations such as the OECD’s ‘Guidelines for Multinational Enterprises’, may be legally binding yet not judicially enforceable (Davarnejad 2010). Even ‘softer’ are those rules that are not codified at all but widely supported by a (transnational) public and NGOs. Despite their differing degrees of codification and enforceability, all those rules articulate normative expectations: They are based on expectancies about corporate behaviour that are not simply discarded or adapted in the face of disappointment but rather retained and defended (Luhmann 1985). They thus presume that disappointment does not mean that the expectancy was wrong but that someone did something wrong. Against this common normative background, social norms differ in the ways in which wrongdoings are dealt with, be it through sanctions or admonishments, public outrage or scandalizations or simply denial and neglect; yet they converge in the way rule violations are dealt with: perpetrators need to be identified in order to attribute responsibility and blame. Laws define corporate responsibility in both prescriptive and causal terms. They shape the extent of corporate responsibility through duties of ‘due diligence’ that are designed to prevent undesired actions and outcomes and through tort law that establishes liability for damages inflicted on others (Bonnitcha and McCorquodale 2017; Wagner 2021). In the context of transnational regulation, defining responsibility in global supply chains remains problematic: Agency and causalities are regularly distributed among several actors and jurisdictions. Some countries have already adopted legislation that mandates corporate responsibility for human rights and labour standards along the supply chain. In France, companies have been obliged

Rules and responsibilities  137 since 2017 to reduce environmental and human rights violations and may be liable for them. In 2019, the Netherlands introduced a law on the due diligence of companies with regard to child labour (‘Wet zorgplicht kinderarbeid’). And Germany’s ‘Act on Corporate Due Diligence Obligations in Supply Chains’ (Lieferkettensorgfaltspflichtengesetz, LkSG), adopted in June 2021 after long political debates, similarly standardizes behavioural obligations that are subject to fines. The difference between ‘hard’ and ‘soft’ rules is sometimes fuzzy because common-sense understandings of responsibility also inform and shape legal notions of liability. The social attribution of responsibility on the one hand and the legal definition of liability on the other are dynamic and influence each other reciprocally: Legal norms that legitimate claims for injunctive relief or damages may be invoked strategically to highlight problematic business practices (Graser and Helmrich 2019). But moral claims may also go beyond what is legally enforceable. If acting responsibly by exercising due diligence fails to prevent undesired consequences, one can still be held responsible. Responsibility, then, amounts to the retrospective demand to ‘answer for mistakes’ (Luhmann 1964, p. 172). Companies are held accountable not only for the direct consequences of their decisions, but also for ecological or societal side-effects whose causal connection with corporate action is not immediately evident. Yet the social attribution of corporate responsibility does not have to conform with legal definitions. It merely relies on plausible criteria such as the ability to communicate and to act (Sombetzki 2014) – and helps them to take effect performatively. If companies are attributed the responsibility for the consequences of their decisions, they are in fact held responsible – regardless of whether the corporation would like to portray itself as responsible. Any kind of public self-presentation, be it in terms of ‘responsible’ business practice or not, gives rise to the expectation that a corporation accepts its ‘actorhood’ and thus a moral frame of reference for evaluating its decisions. Activists pursue the public attribution of responsibility by ‘moralizing’ economic decisions (Holzer 2010). In contrast to the legal definition of responsibility through liability laws, a ‘moralizing’ discourse does not depend on a fixed and narrow understanding of diligence and causality. The social attribution of responsibility invokes common-sense notions of accountability that may well transcend the extent of legal liability, for instance if a corporation is blamed for long-term consequences of its decisions or for far-away incidents along its supply chain. Lacking institutional power, protest groups and activists have to rely on public opinion to support such claims and allegations. In order to challenge economic decisions due to their social or environmental impact, they have to mobilize public opinion. Activists can capitalize on the fact that national regulations differ considerably and are often considered insufficient to deal with environmental problems of global relevance. In a nation-state context the democratic legislative procedure confers not only legality but also legitimacy to positive law. Following Kant, the principle of self-government of the people in a democratic state ensures the legitimacy of legislative acts: no one would willingly do himself an injustice, and since democratic government means that the people govern themselves, democracy makes the illegitimate use of legislative power impossible (Habermas 1989, p. 466). Yet on a transnational level the legitimacy of decisions is more difficult to achieve. The problem arises from the fact that legal norms, even if agreed upon in a democratic procedure, remain confined to a national context. Groups in other places which consider themselves affected may doubt the legitimacy of a certain norm. Transnational corporations are con-

138  Research handbook on soft law fronted with a situation in which the legitimacy of their operations may be challenged no matter what their legal status is. According to Phil Watts (1998, p. 25), former chairman of Shell International, conflicting demands and the fact that activities anywhere in the world are scrutinized by an increasingly transnational and diverse public pose ‘major dilemmas for business’. Compared with the institutional framework provided by nation-states, the normative structure of world society is weakly institutionalized and fragmented. Rather than uniform legal regulation it provides a patchwork of different, and sometimes even contradictory, demands across different cultures and value systems.4 Perspectives ‘out of place’ therefore play an important role in defining the factual responsibility of corporations.

4.

NEGOTIATING THE RULES OF ACTORHOOD

Based on the self-regulation of business, the scrutiny of transnational NGOs and a patchwork of national legislation, a new kind of transnational governance has emerged. In contrast to ‘governance without government’ (Rosenau and Czempiel 1992) based on non-state actors, new forms of regulation are characterized by a networked, ‘hybrid’ interaction of state, private and civil society actors (Bair 2017). The interaction of states, civil society and corporations constitutes a ‘governance triangle’ (Abbott and Snidal 2009) that involves the development and enforcement of legislation but also the governance through non-binding ‘soft law’: Business principles, codes of conduct, and industry standards presume that companies voluntarily assume responsibility and participate in the monitoring of compliance. Based on private as well as state norms and actors, a ‘hybrid’ form of regulation of transnational economic affairs and supply chains has emerged (Bartley 2007; Büthe and Mattli 2011; Rühmkorf 2018). The public accountability of business within such a hybrid regime of regulation goes beyond legal regulations. Definitions of acceptable and responsible practice are contested and vary from place to place. Living up to the norms of ‘hard’ law may not satisfy the standards of ‘soft’ law or the broader normative expectations of civil society. Neither can self-styled codes of ethics reliably persuade the public that business acts in accordance with wider societal goals. Such efforts may always be deemed insufficient by critical observers. In this case, TNCs have no ultimate authority to appeal to. They have to deal with the criticism as actors in the public sphere, either defending themselves or seeking a compromise with their opponents. In this sense, corporations become more and more ‘political’ as they have to engage in ‘talk’ to justify their ‘actions’ (Brunsson 1989). Because of their global presence, TNCs are likely targets for protest action (Yearley and Forrester 2000). They often assume top positions in individual markets and seek to foster their public image and reputation through extensive and expensive advertising campaigns. Therefore, they usually are exemplary cases of what has been called ‘public exposure’. The idea of ‘public exposure’ refers to the fact that enterprises touch on public interests and, vice versa, they are themselves afflicted by measures taken in the name of public interest (Dyllick 1989, p. 209). Especially in areas bearing environmental impact such as the chemical industry, the public has grown wary of the side-effects and long-term consequences of decisions. The ensuing scrutiny and distrust of business practice has transformed multinational corporations into ‘quasi-public institutions’ (Kristol 1975; Ulrich 1977). That calls into question the traditional concept of business decisions as essentially private decisions made by or on behalf of the owners of a company. Decisions are increasingly becoming public in nature

Rules and responsibilities  139 because of their alleged impact on other people. The larger the company, the more likely it is to have such an impact. ‘The price of successful economic growth for a company is that it gains increased public visibility. It is thus more subject to public scrutiny and public criticism than a small company’ (Willetts 1998, p. 225). Ultimately, all companies depend on the public acceptance of their actions and thus have to position themselves in relation to public perceptions, standards, and etiquette (Brunsson 1989, p. 216). Conflicts between corporations and social movements are not only about conflicting values but also about the rules of being an ‘actor’. Those rules are largely informal and therefore not as binding as formal law; but they are nonetheless rules that organizations cannot ignore if they are to maintain the legitimacy of their operations in institutionalized environments (Meyer and Rowan 1977; DiMaggio and Powell 1983). Transnational activists and social movements successfully direct their claims at corporations as corporate actors liable to moral evaluation. This moralization of corporate behaviour requires that the corporation be framed as a motivated, goal-directed actor rather than a non-social entity. In other words, corporate responsibility and accountability presuppose a certain degree of ‘actorhood.’ Actors and their actions are ‘artefacts of processes of attribution’ (Luhmann 1995, p. xliv), and it is common to attribute intentions, decisions, and actions to organizations. Corporations are observed and scrutinized as planners and perpetrators of their decisions. Moreover, they are treated as moral actors subject to the evaluation of both the motivation and outcomes of their actions (Holzer 2010). However, organizations do not act according to a single purpose. Frictions and contradictions among various objectives, audiences, and temporal horizons are common. Increasing brand loyalty through high-quality products might hamper profits. Targeting an urban milieu may disappoint rural clients. Measures to ensure short-term profitability could deter long-term investors. To deal with conflicting expectations organizations may simply adopt opportunistic strategies: Incompatible expectations of different constituencies are dealt with by first doing one thing and then the other, i.e., by the ‘sequential attention to goals’ (Cyert and March 1963, p. 118).5 Such opportunistic behaviour may satisfy various audiences with specific interests. Investors will be inclined to welcome or at least to tolerate some attention to environmental criteria if that does not affect the bottom line. But an audience that emphasizes and monitors the moral quality of actorhood necessarily employs a more wholistic perspective on business practice and denounces inconsistent behaviour. Confronted with such demanding observers, corporations are unable to comprehensively comply with those expectations but may instead seek to reflect them in their ‘talk’, i.e., by a compatible and acceptable ideology. The result is a constitutive discrepancy between talk and practice: ‘organizational hypocrisy’ (Brunsson 1989). If hypocrisy, as La Rochefoucauld claimed, is the homage that vice pays to virtue, the present conjuncture of hypocritical window-dressing and green-washing hints at the fact that there are indeed implicit norms of virtue and propriety at work that such efforts seek to address. While the responsible self-presentations of more and more corporations must be regarded as efforts to deflect criticism, they may also set in motion a ‘virtuous spiral’ (Kaler 2000): The manifest affirmation of societal values raises expectations and lays the foundation for scrutiny of the self-proclaimed principles. To counter criticism, corporations need to both address shortcomings and further elaborate their self-presentation, which may again lead to rising expectations. Anti-corporate activism exploits inconsistencies between self-presentation and behaviour by focusing on companies that portray themselves as benevolent and ethical: ‘good firms’ make ‘good targets’ (King and McDonnell 2015). The fact that organizational

140  Research handbook on soft law hypocrisy may attract criticism as much as it deflects it shows that the public discourse about the rules pertaining to corporate behaviour is lopsided in an unexpected way. From the perspective of economic and political resources, it appears to pit the proverbial David against a corporate Goliath. From a discursive angle, however, corporations are in a disadvantaged position since the question of responsibility imposes the burden of proof on them. Whether certain actions are criticized or justified, corporations and protest groups appeal to conflicting aspects of modern rationality: While TNCs invoke a ‘liberal’ script of rational progress and economic growth with their own agency at the very centre, NGOs follow a ‘communitarian’ script drawing on notions of human rights and quality of life which enables them to act as the disinterested advocates of universal values – as ‘rationalized others’ (Meyer 1996). As the representatives of altruistic ‘otherhood’, NGOs enjoy a high degree of legitimacy. The pursuit of economic profitability, in comparison, inevitably appears as egoistic – which is acceptable, too, but does not afford a particularly advantageous moral position. Conflicts and debates about corporations’ responsibility for accidents, labour standards, and environmental problems abound. Despite persisting legal ambiguities about corporate accountability and liability in transnational economic networks, the public discourse about corporate responsibility has established and institutionalized the actorhood of corporations based on a ‘“beyond-compliance” notion of responsibility’ (Favotto and Kollman 2020, p. 209). In addition to the traditional legal expertise in governance through ‘hard’ law, corporations are also subject to the ‘soft’ rules that legitimate the claims and normative expectations of activists and other critical observers who attribute responsibility to them and insist on holding them responsible for alleged misdemeanours within their scope of influence.

5. CONCLUSION The responsibilities of corporations have various sources: legal regulations, voluntary corporate responsibility, binding yet non-enforceable agreements, and diffuse but consequential notions of actorhood scrutinized by public observers. The rules that emanate from these diverse sources of normative expectations do not necessarily converge. They institutionalize a complex normative framework of responsibility that is differentiated according to the types of sanctions that come into play when violations are observed. Sanctions range from fines over exclusion from associations and other meta-organizations or entire markets to the public outrage about scandals and accidents. Notwithstanding the multiplicity of specific rules and sanctions, the basic parameters of this normative framework pivot around the baseline of corporate actorhood: As decision-making actors, corporations are held responsible for consequences that can be attributed to their decisions, be it in terms of legal liability or of a more broadly conceived moral responsibility. Retrospectively calling corporations to account, for instance in the wake of accidents or scandals, is a powerful yet episodic vehicle of articulating the social norms underpinning the attribution of responsibility. Public pressure cannot be upheld continuously because the attention span of the public sphere is short-lived, and activists change their focus frequently. However, recent developments in both international soft law and national legislation suggest that future rules that institutionalize and regulate the responsibility of business in transnational networks and supply chains will support and intensify rather than decrease the demands on corporate actorhood.

Rules and responsibilities  141

NOTES 1.

The field of business ethics has grown considerably over the last 30 years. Among a wealth of publications see Brummer (1991), Carmichael (1995), George (1999), Carroll and Buchholtz (2009), Crane et al. (2019). 2. The term ‘stakeholder’ was developed at the Stanford Research Institute (SRI) in the 1960s and subsequently used to denote ‘individuals or groups who depend on the company for the realisation of their personal goals’ (Rhenman 1968, p. 25). According to Freeman (1984, p. 25), a stakeholder as ‘(a)ny group or individual who can affect or is affected by the achievement of the firm’s objectives’. See Abraham (2021) for a critical appraisal of the stakeholder concept in the context of participatory politics. 3. Cf. Ott (1977, p. 167): ‘The profit motive too … has lost its unambiguousness. In the conditions of modern economic and enterprise structures, it can a priori no longer mean maximal, but only optimal, profit, guaranteed in the long term. Inclusion of this time dimension also implies including those other factors in the business decision process, making the profit motive relative. Long-term profit maximization in fact means partial renunciation of profit from regard for the firm’s “social environment.” The profit approach is limited by principles of controlled growth, maintenance of market share and stability of the company’ (translated in Breunung and Nocke 1994, p. 273). 4. This is partly because they are used to relying on ‘expert culture’, i.e., the possibility to establish the validity of claims in a given field of knowledge unambiguously. In a limited domain such as economic rationality this may be possible to some extent. However, if the boundaries between, for instance, economic and ecological considerations are ignored or politicized this is bound to fail. Corporations increasingly have to deal with the problem that ‘politics and morality are gaining priority over expert reasoning’ (Beck 1996, p. 20). 5. ‘Organizations resolve conflict among goals, in part, by attending to different goals at different times. (For instance) the business firm is likely to resolve conflicting pressures to “smooth production” and “satisfy customers” by first doing one thing and the other’ (Cyert and March 1963, p. 118).

REFERENCES Abbott, K. W. and Snidal, D., (2000). Hard and soft law in international governance. International Organization. 54(3), 421–456. Abbott, K. W. and Snidal, D., (2009). The governance triangle: regulatory standards institutions and the shadow of the state. In: W. Mattli and N. Woods, eds. The politics of global regulation. Princenton, N.J.: Princeton University Press. pp. 44–88. Abraham, K. J., (2021). Midcentury modern: the emergence of stakeholders in democratic practice. American Political Science Review [online]. First View, 1–14. [Viewed 15 April 2022]. Available from: doi: 10.1017/S0003055421001106. Bair, J., (2017). Contextualising compliance: hybrid governance in global value chains. New Political Economy. 22(2), 169–185. Bair, J. and Palpacuer, F., (2012). From varieties of capitalism to varieties of activism: the antisweatshop movement in comparative perspective. Social Problems. 59(4), 522–543. Bartley, T., (2007). Institutional emergence in an era of globalization: the rise of transnational private regulation of labor and environmental conditions. American Journal of Sociology. 113(2), 297–351. Bartley, T., (2018). Rules without rights: land, labor, and private authority in the global economy. Oxford: Oxford University Press. Bartley, T. and Child, C., (2014). Shaming the corporation: the social production of targets and the anti-sweatshop movement. American Sociological Review. 79(4), 653–679. Beck, U., (1996). World risk society as cosmopolitan society? Ecological questions in a framework of manufactured uncertainties. Theory, Culture & Society. 13(4), 1–32. Bonnitcha, J. and McCorquodale, R., (2017). The concept of ‘due diligence’ in the UN Guiding Principles on Business and Human Rights. European Journal of International Law. 28(3), 899–919.

142  Research handbook on soft law Breunung, L. and Nocke, J., (1994). Environmental officers: a viable concept for ecological management?. In: G. Teubner, L. Farmer and D. Murphy, eds. Environmental law and ecological responsibility. The concept and practice of ecological self-organization. Chichester: Wiley. pp. 267–295. Brummer, J. J., (1991). Corporate responsibility and legitimacy. An interdisciplinary analysis. Westport, C.T.: Greenwood Press. Brunsson, N., (1989). The organization of hypocrisy. Talk, decisions, and actions in organizations. Chichester: Wiley. Büthe, T. and Mattli, W., (2011). The new global rulers: the privatization of regulation in the world economy. Princeton, N.J.: Princeton University Press. Carmichael, S., (1995). Business ethics: the new bottom line. London: Demos. Carroll, A. B. and Buchholtz, A. K., (2009). Business and society: ethics and stakeholder management. 7th ed. Mason, O.H.: South-Western Cengage Learning. Cohen, M., (1998). Evidence of a new environmental ethic: assessing the trend towards investor and consumer activism. In: I. Jones and M. Pollitt, eds. The role of business ethics in economic performance. London: Palgrave Macmillan. pp. 111–140. Crane, A., (2005). Meeting the ethical gaze: challenges for orienting to the ethical market. In: R. Harrison, T. Newholm and D. Shaw, eds. The ethical consumer. London: Sage. pp. 219–232. Crane, A., Matten, D. and Moon, J., (2008). The emergence of corporate citizenship: historical development and alternative perspectives. In: A. G. Scherer, and G. Palazzo, eds. Handbook of research on global corporate citizenship. Cheltenham: Edward Elgar. pp. 25–49. Crane, A., Matten, D., Glozer, S. and Spence, L. J., (2019). Business ethics: managing corporate citizenship and sustainability in the age of globalization. 5th ed. Oxford: Oxford University Press. Cyert, R. M. and March, J. G., (1963). A behavioral theory of the firm. Englewood Cliffs, N.J.: Prentice Hall. Davarnejad, L., (2010). The impact of non-state actors on the international law regime of corporate social responsibility: blessing or curse?. In: M. Noortmann and C. Ryngaert, eds. Non-State actor dynamics in international law: from law-takers to law-makers. Abingdon: Routledge. pp. 41–65. DiMaggio, P. J. and Powell, W. W., (1983). The iron cage revisited: institutional isomorphism and collective rationality in organizational fields. American Sociological Review. 48(2), 147–160. Donaldson, T. and Preston, L. E., (1995). The stakeholder theory of the corporation: concepts, evidence, and implications. Academy of Management Review. 20(1), 65–91. Dyllick, T., (1989). Politische legitimität, moralische autorität und wirtschaftliche effizienz als externe lenkungssysteme der unternehmung. grundvorstellungen einer gesellschaftsbezogenen managementlehre. In: K. Sandner, ed. Politische prozesse in unternehmen. Berlin: Springer. pp. 205–230. Elkington, J., (1997). Cannibals with forks: the triple bottom line of 21st century business. Oxford: Capstone. Es, R. van and Meijlink, T. L., (2000). The dialogical turn of public relation ethics. Journal of Business Ethics. 27(1/2), 69–77. Favotto, A. and Kollman, K., (2020). An expanding conception of social responsibility? Of global norms and changing corporate perceptions. In: H. Hansen-Magnusson and A. Vetterlein, eds. The rise of responsibility in world politics. Cambridge: Cambridge University Press. pp. 188–212. Freeman, R. E., (1984). Strategic management: a stakeholder approach. Marshfield, M.A.: Pitman. Friedman, M., (1970). The social responsibility of business is to increase its profits. New York Times Magazine. 13 September 1970, 17. Gabriel, Y. and Lang, T., (1995). The unmanageable consumer: contemporary consumption and its fragmentations. London: Sage. Galbraith, J. K., (2002). A perfect crime: inequality in the age of globalization. Daedalus. 131(1), 11–25. George, R. T. de, (1999). Business ethics. 5th edn. Upper Saddle River, N.J.: Prentice Hall. Goodpaster, K. E., (1983). The concept of corporate responsibility. In: T. Regan, ed. Just business. New introductory essays in business ethics. Philadelphia: Temple University Press. pp. 292–323. Graser, A. and Helmrich, C., eds., (2019). Strategic litigation: begriff und praxis. Baden-Baden: Nomos. Greenberg, J. and Knight, G., (2004). Framing sweatshops: Nike, global production, and the American news media. Communication and Critical/Cultural Studies. 1(2), 151–175. Habermas, J., (1989). Volkssouveränität als verfahren. ein normativer begriff von öffentlichkeit. Merkur. 43(6), 465–477.

Rules and responsibilities  143 Handy, C., (1996). Beyond certainty: the changing worlds of organizations. Boston, M.A.: Harvard Business School Press. Holzer, B., (2007). Framing the corporation: Royal Dutch/Shell and human rights woes in Nigeria. Journal of Consumer Policy. 30(3), 281–301. Holzer, B., (2010). Moralizing the corporation: transnational activism and corporate accountability. Cheltenham, UK: Edward Elgar. Hopkins, M., (1999). The planetary bargain: corporate social responsibility comes of age. Houndmills: Palgrave Macmillan. Jennings, M. E. and Entine, J., (1998). Business with a soul: a reexamination of what counts in business ethics. Hamline Journal of Public Law & Policy. 20(1), 1–88. Jones, I. and Pollitt, M., (1998). Introduction. In: I. Jones and M. Pollitt, eds. The role of business ethics in economic performance. Houndmills: Palgrave Macmillan. pp. 1–20. Kaler, J., (2000). Reasons to be ethical: self-interest and ethical business. Journal of Business Ethics. 27(1/2), 161–173. King, B. G. and McDonnell, M.-H., (2015). Good firms, good targets: the relationship among corporate social responsibility, reputation, and activist targeting. In: K. Tsutsui and A. Lim, eds. Corporate social responsibility in a globalizing world. Cambridge: Cambridge University Press. pp. 430–454. Kristol, I., (1975). On corporate capitalism in America. The Public Interest. 41(Fall), 124–141. Levitt, T., (1958). The dangers of social responsibility. Harvard Business Review. 36(5), 41–50. Lindblom, C. E., (1977). Politics and markets: the world's political-economic systems. New York: Basic Books. Lindblom, C. E., (1982). The market as a prison. The Journal of Politics. 44(2), 324–336. Luhmann, N., (1964). Funktionen und folgen formaler organisation. Berlin: Duncker und Humblot. Luhmann, N., (1985). A sociological theory of law. Translated byM. Albrow, ed. London: Routledge & Kegan Paul. Luhmann, N., (1988). Die wirtschaft der gesellschaft. Frankfurt am Main: Suhrkamp. Luhmann, N., (1995). Social systems. Translated by J. Bednarz Jr. and D. Baecker. Stanford, C.A.: Stanford University Press. March, J. G., (1962). The business firm as a political coalition. Journal of Politics. 24(4), 662–678. Marens, R. and Wicks, A., (1999). Getting real: stakeholder theory, managerial practice, and the general irrelevance of fiduciary duties owed to shareholders. Business Ethics Quarterly. 9(2), 273–293. Matten, D. and Crane, A., (2005). Corporate citizenship: toward an extended theoretical conceptualization. Academy of Management Review. 30(1), 166–179. Meyer, J. W., (1996). Otherhood: the promulgation and transmission of ideas in the modern organizational environment. In: B. Czarniawska and G. Sevón, eds. Translating organizational change. Berlin: De Gruyter. pp. 241–252. Meyer, J. W. and Rowan, B., (1977). Institutionalized organizations: formal structure as myth and ceremony. American Journal of Sociology. 83(2), 340–363. Meyer, J. W., Pope, S. M. and Isaacson, A., (2015). Legitimating the transnational corporation in a stateless world society. In: K. Tsutsui and A. Lim, eds. Corporate social responsibility in a globalizing world. Cambridge: Cambridge University Press. pp. 27–72. Mintzberg, H., (1983). The case for corporate social responsibility. Journal of Business Strategy. 4(2), 3–15. Mitroff, I. I., (1983). Stakeholders of the organizational mind. Toward a new view of organizational policy making. San Francisco: Jossey-Bass. Ott, C., (1977). Recht und realität der unternehmenskorporation: ein beitrag zur theorie der juristischen person. Tübingen: J.C.B. Mohr (Paul Siebeck). Rappaport, A., (1986). Creating shareholder value: the new standard for business performance. New York: Free Press. Rhenman, E., (1968). Industrial democracy and industrial management: a critical essay on the possible meanings and implications of industrial democracy. London: Tavistock. Rosenau, J. N. and Czempiel, E.-O., eds., (1992). Governance without government: order and change in world politics. Cambridge: Cambridge University Press. Rühmkorf, A., (2018). Corporate social responsibility in der lieferkette: governance und verantwortlichkeiten. Zeitschrift für Unternehmens- und Gesellschaftsrecht. 47(2-3), 410–432.

144  Research handbook on soft law Shaffer, G. C. and Pollack, M. A., (2010). Hard vs. soft law: alternatives, complements, and antagonists in international governance. Minnesota Law Review. 94(3), 706–799. Sikkink, K., (1986). Codes of conduct for transnational corporations: the case of the WHO/UNICED code. International Organization. 40(4), 815–840. Sklair, L., (2001). The transnational capitalist class. Oxford: Blackwell. Sombetzki, J., (2014). Verantwortung als begriff, fähigkeit, aufgabe: eine drei-ebenen-analyse. Wiesbaden: Springer VS. Sorell, T. and Hendry, J., (1994). Business ethics. Oxford: Butterworth-Heinemann. Sternberg, E., (1994). Just business. Business ethics in action. London: Little, Brown. Stone, C. D., (1975). Where the law ends: the social control of corporate behavior. New York: Harper & Row. Teubner, G., (1994). The invisible cupola: from causal to collective attribution in ecological liability. In: G. Teubner, L. Farmer and D. Murphy, eds. Environmental law and ecological responsibility. The concept and practice of ecological self-organization. Chichester: Wiley. pp. 17–47. Tugendhat, C., (1973). The multinationals. Harmondsworth: Penguin. Ulrich, P., (1977). Die großunternehmung als quasi-öffentliche institution. eine politische theorie der unternehmung. Stuttgart: Poeschel. Ulrich, P., (1998). Integrative wirtschaftsethik. grundlagen einer lebensdienlichen ökonomie. 2nd ed. Bern: Haupt. Wagner, G., (2021). Tort law and human rights. In: M. Saage-Maaß, P. Zumbansen, M. Bader and P. Shahab, eds. Transnational legal activism in global value chains: the Ali Enterprises Factory fire and the struggle for justice. Cham: Springer. pp. 209–233. Watts, P., (1998). The international petroleum industry: economic actor or social activist?. In: J. V. Mitchell, ed. Companies in a world of conflict: NGOs, sanctions, and corporate responsibility. London: Earthscan. pp. 23–31. Weiss, J. W., (1998). Business ethics. a stakeholder and issues management approach. 2nd ed. Fort Worth: Dryden Press. Willetts, P., (1998). Political globalization and the impact of NGOs upon transnational companies. In: J. V. Mitchell, ed. Companies in a world of conflict: NGOs, sanctions, and corporate responsibility. London: Earthscan. pp. 195–226. Williams, O. F., (2004). The UN Global Compact: the challenge and the promise. Business Ethics Quarterly. 14(4), 755–774. Yearley, S. and Forrester, J., (2000). Shell, a sure target for global environmental campaigning?. In: R. Cohen and S. M. Rai, eds. Global social movements. London: Athlone Press. pp. 134–145. Zadek, S., (2001). The civil corporation: the new economy of corporate citizenship. London: Earthscan.

10. Soft law, technical standards and European private law Hans-W. Micklitz

1.

THE CLAIM AND THE ARGUMENT

There is no common understanding of what soft law might mean (Cafaggi 2006). This chapter approaches technical standards as soft law and focuses on their interaction with private law. Technical standards are overwhelmingly regarded as soft law. The present claim is the following: the European authorities, the European legislature and in particular the European Commission are establishing technical standards as default rules that underpin contracts in business-to-business (b2b) but also in business-to-consumer (b2c) relations. The argument to be developed in this chapter is that technical standards function as a substitute for the failed attempt to find political support for the development of a European Civil Code (Micklitz 2010). I demonstrate how the European Union (EU) has established and will further strengthen a contractual underground in which technical standards determine and underpin contractual transactions in Europe. Technical standards will also open up new perspectives for the EU as a global player (van Gestel and van Lochem 2020, p. 27). The rising importance of technical standards in EU ‘law making’ goes along with the development of private law beyond the nation state and thereby beyond the EU (Michaels and Jansen 2006). What I will try to demonstrate in this chapter is that technical standards serve different purposes at the same time: first they fill regulatory gaps in the EU, the gaps that cannot be closed through legal harmonization due to lack of competence and political resistance of the Member States, second they allow the EU to open up sectorial markets for European companies by relying on and promoting EU-based soft law in the transnational arena, third, they enable the EU to play to its long-standing strength in standardization and to integrate concerns linked to fundamental rights and labour, environmental and consumer protection in the standardization process. Thereby technical standards establish a common level playing field that underpins European private law.

2.

THE SURFACE AND THE UNDERGROUND

European private law is the surface, with technical standards belonging to the underground. This claim needs to be explained. Private law, whether national or European, and ‘the law of standards’ (Schepel 2005; see also Vallejo 2021) are regarded as two separate worlds. The EU has managed to adopt a dense net of private law rules. These exist mainly in the field of b2c relations, but there are rules also on b2b relations and in the field of regulated markets, finance, telecom, energy and transport (Basedow 2021; Stürmer 2021). The legislative history of secondary EU law and academic initiatives, especially the Academic Draft Common Frame of Reference, underpin the metaphor of the two worlds apart. Whilst such an assessment is 145

146  Research handbook on soft law certainly true for the horizontal regulation of b2c relations and b2b relations, the picture looks slightly different when it comes to regulated markets as explained below.1 The EU entered the field of regulated markets not via private law, but via economic regulation in order to liberalize and (partly) privatize markets, break down statutory monopolies and guarantee access for potential competitors beyond the incumbents. Once the markets were opened, the EU legislature entered into private law relations, both with respect to b2b and b2c. This process started after the adoption of the Single European Act in 1986 and is still ongoing. During the making of regulated markets, the EU legislature realized that it has to dive into the field of standardization, as technical standards are a classical tool for opening up markets in order to establish a common level playing field. But metaphorically, the worlds of private law on the one hand and of standards on the other have not been brought together. In regulated markets, where the technical standards and private law rules are close to one another and often enshrined in the same set of secondary EU law, there has not been any explicit attempt to reflect the interaction between technical standards and private law, neither in politics nor in legal scholarship.2 The same oversight applies to what has become known as the ‘law of standards’ (Schepel 2005), the intermingling of legal requirements and private regulation, the development what Vallejo (2021) also calls ‘private administrative law’. The crucial importance of technical standards for the building of the EU common market, later the internal market, became clear after the EU had managed to abolish customs. In 1969, the EU had already set up its first programme to deal with technical standards as ‘non-tariff barriers to trade’. It promoted the adoption of secondary EU law to overcome potential barriers to trade through the harmonization of technical requirements (Joerges et al. 1991). The 1969 programme, however, ended in a deadlock. Harmonizing technical standards by way of EU directives burdened the legislature with political debates about technical specifications. In 1985, in line with the Single European Act, the Commission developed the New Approach. In 2012, the New Approach was turned into the New Legislative Framework (NLF),3 currently under revision so as to better integrate the digital economy and the European Green Deal in the Roadmap for a new Standardisation Strategy (see European Commission 2021a and comprehensively Micklitz 2023). The regulatory rationale behind the European Standardisation System (ESS) – the name given to it by the Commission – is the following. The Memorandum of Understanding between the Commission and the European Standardization Bodies, (European Committee for Standardisation (CEN), European Committee for Electrotechnical Standardization (CENELEC) and later European Telecommunications Standards Institute (ETSI)), enables the Commission to mandate the elaboration of a standard in exchange for taking over parts of the production costs. These so-called ‘harmonized standards’ are published in the Official Journal after approval of the Commission (Articles 2(1)(c) and 10(5)–(6) of Regulation (EU) No 1025/2012). They have to be distinguished from stand-alone ‘voluntary industry standards’, which are self-financed and not mandated. The 1985 New Approach established an interaction between binding secondary EU law and complementary technical standards. EU legislature no longer regulated technical details, but limited itself to the formulation of legally binding ‘general requirements’ which were to be concretized through technical standards, either through voluntary or harmonized technical standards. The manufacturers are not obliged to use the Europeanized level of technical standardization. They remain free to choose a different technical option, as long as it complies with the ‘binding general requirements’.4 Compliance with the European standards, both with those

Soft law, technical standards and European private law  147 mandated by the Commission or voluntary industry standards, secures access to the European market. As a rule, it is for the manufacturer to ensure that its products respect the technical requirements and document compliance through self-certification via the CE label. The EU may also require certification by a third party via secondary EU law. Third-party certification bodies need to be accredited (see Tamm-Hallström and Gustafsson in this volume). The Commission understands the harmonization of technical requirements for the manufacturing of products as a success story (European Commission 2022a). The relevance of product standardization for private law is most obvious when looking at the liability for the safety of products. Here technical standards gained ground as a reference point in potential litigation. The ‘law of standards’ came much closer to private law when the EU adopted in 2006 the Services Directive 2006/123/EC (2006). The original political aim of the EU was to set up an overall frame for all those services that are not subject to particular sector or subject-related rules. In the field of services, technical standards are easily transgressing the legal boundaries in formulating legal rights and obligations which determine the scope and the content of the services to be performed. One might interpret the interaction of the rules in the Services Directive and the underpinning standardization of services as an early attempt of the Commission to compensate for the lack of common rules on contracts for services, a deficit which is equally visible in the Draft Common Frame of Reference (see von Bar et al. eds., (2009)). However, the Commission did not really succeed in pushing the technical standardization of services either in b2b or in b2c relations. This is visible from the fact that 70 per cent of EU’s gross income results from services, but only 2 per cent of all EU standards are dealing with services (European Commission 2021a). The few existing standards of services are, nevertheless, of major interest for private lawyers, as they demonstrate the extent to which technical standards can substitute private laws rules, thereby providing for a much broader understanding of ‘contract’, covering the pre-contractual and the post-contractual stage (Micklitz 2007; van Leeuwen 2017). In its Roadmap for a new Standardisation Strategy, the Commission (2022) is referring explicitly to universal services as a potential field of standardization. The other envisaged areas of activities are those with a shaky competence: social policy, education, sport, and health. Strongly interlinked to the Roadmap is the ongoing attempt of the Commission to lay down an overall framework for the digital economy and digital society (sic!) through the Digital Governance Act (DGA), the Digital Market Act (DMA), the Digital Services Act (DSA), and the Artificial Intelligence Act (see European Commission 2021c). In the language of the ESS, the three Acts and the proposal on AIA are mirroring the 1980’s New Approach rationale, where broadly-worded general requirements will have to be specified and concretized through harmonized standards and voluntary industry standards. The envisaged standards, provided that the four proposals find political support, will have to integrate not only physical harm, but also psychological harm and fundamental rights. In short, the new four will strengthen the ‘underground’ world of standards, now in the form of algorithms (Micklitz, Helberger and Rott 2021).

148  Research handbook on soft law

3.

THE LINKAGE BETWEEN THE SURFACE AND THE UNDERGROUND

Why then bring European private law and European technical standards together, if the two fields seem to be rather distant from each other? What is the impact of technical standards on private law and vice versa, and where does it show? What exactly connects the ‘surface’, that is, private law to the ‘underground’, that is, the technical standards, or vice versa? In legal terminology, technical standards seem to belong to the realm of trade law, which in the European understanding covers the four freedoms and competition law. There is an abundant literature on technical standards as ‘non-tariff barriers to trade’. This is very much a public law perspective, which is by and large mainstream in legal scholarship. For such scholarship, the major question then is whether and to what extent technical standards as soft law can be and should be subject to judicial review. From a private law angle, it is particularly interesting to investigate potential parallels between the judicial control of technical standards and the judicial control of (private law) standard terms through the case law of the European Court of Justice. I will, however, start the analysis with the assessment of the interaction of technical standards with contract and tort law, before I turn to judicial review. 3.1

Contract Law

The Treaty of Rome laid down the four freedoms embedded in a competitive market order, but did not deal with private law, with one notable exception. It provided for the possibility of the Member States to elaborate rules on jurisdiction, which was realized in the Brussels Convention in 1980. Overall, however, the Treaty of Rome was built on the existence of national private law regimes which enables private parties to make use of the four freedoms in cross-border transactions. This might explain why most of the preliminary references based on the four freedoms enjoy a contractual background (Davies 2013). Technical standards come into picture when the parties to a contract argue about deviations from the agreed quality of the product. Historically, technical standards dealt with measuring techniques so as to make sure that the bits and pieces needed for the composition of a product fit together. The ancient story goes that the Romans invented standardization in the making of the galleys. In modern day Europe, standardization is linked to industrialization. The move towards mass production and the overall dependence on electricity required and enhanced standardization techniques to rationalize the manufacturing process and to secure interoperability and electrical safety. The leading industrial states established national standardization bodies, the UK and the US in 1901, Germany in 1917, and France in 1926.5 These countries also promoted the establishment of international counterparts, the International Organisation for Standardisation (ISO) in 1947 and the International Electrotechnical Commission (IEC) in 1901. Their European counterparts were set up after the birth of the EU: European Committee for Standardisation (CEN, from French Comité Européen de la Normalisation) in 1961, European Committee for Electrotechnical Standardization (CENELEC, from French Comité Européen de la Normalisation Électrotechnique) in 1978, and European Telecommunication Standards Institute (ETSI) in 1988 under the initiative of the Commission. Over time, however, and not least because of the pressure from what is today called the Global South, technical standardization expanded to more than mere measuring techniques,

Soft law, technical standards and European private law  149 and now routinely covers defining the composition of products and even the way in which they should be produced (process standards). This transformation of technical standards from mere measuring into defining the quality of the end product and even process standards is well documented and has led to a rather sophisticated categorization (Marburger 1979; Vieweg 2021; for the literature on transnational private rule-making which reaches far beyond technical standardization, see Büthe and Mattli 2011, and Contreras 2019). The legal benchmark for assessing the defectiveness of the product is the agreement between the parties, independent of the legal order in question and independent of whether continental or common law applies. Private law orders are built on the premise that the parties to a contract enter into negotiations and define jointly the conformity of the product. That is why technical standards can only serve as a reference point if the parties are referring to the respective technical standards explicitly. Only if there is no such agreement, the question arises whether technical standards may be taken into account implicitly. The private law orders around the world are translating this distinction into subjective vs. objective conformity requirements. Article 35 of the United Nations Convention on Contracts for the International Sale of Goods (1980, CISG) comes close to a kind of a common, nearly worldwide agreed understanding of what conformity requires.6 It formulates a benchmark for b2b sales contracts. If there is no explicit agreement between the parties, the products must be ‘fit for purpose for which goods of the same description would ordinarily be used’ (Article 35(2)(a) CISG) – thereby the conformity requirements are objectified. There is no explicit reference in the CISG to whether and to what extent technical standards can contribute to an objectivation of the conformity requirements, or whether and under what conditions they may play an implicit role. For long, the scholarly debate revolved around ‘business customs’, into which technical standards might be integrated (Schwenzer 2013; for an early discussion on the linkage between mass production, standardization and the assessment of conformity in sales law, see Knieper, 1981, p. 144). However, in 2018, the CISG Advisory Council (2018) published Opinion No 19 Standards and Conformity of the Goods under Article 35 CISG. The legally non-binding Opinion fills a crucial gap by focusing on the conditions under which technical standards may serve as a reference point if agreed upon implicitly by the parties or even in situations where there is neither an explicit nor an implicit agreement. The Opinion builds on the competence of the members of the advisory board and the existing case law around the world where national courts are using technical standards in the conformity assessment under Article 35 CISG. The Opinion 19 does not go as far as arguing that compliance with technical standards – whether agreed upon explicitly or implicitly – establishes a presumption which releases the seller from contractual liability. The wording of Article 35 leaves the courts a large margin of appreciation in the assessment of the facts of the case in both directions: compliance as presumption of conformity or non-compliance as presumption of non-conformity. The situation in the EU looks different. There is no such thing as a common European sales law, which would have been the equivalent to the CISG. Under pressure from six Member States, the EU Commission withdrew its proposal for a Common European Sales Law (CESL), which is the remainder of the much more ambitious project on a European Civil Code. None of the two projects invested efforts into conceptualizing a revised mechanism for a conformity assessment with due regard for technical standards. The scholarly debate resembles the one around the interpretation of Article 35 CISG,7 but it is Opinion 19, which really breaks new and much more systematic ground.

150  Research handbook on soft law What remains in the EU, however, is the now fully harmonized sales law for b2c transactions. Up until 1999, the Consumer Sales Directive 1999/44/EC (1999) defined the conformity requirements very much in line with Article 35 CISG, subject to certain consumer-related deviations (Wilhelmsson 2008). In 2019, the EU and the Member States did not only agree to move from minimum harmonization to full harmonization, but they also introduced a far-reaching change in the conformity requirements. First and foremost, the new Articles 6 and 7 of the Consumer Sales Directive (EU) 2019/771 (2019) are no longer defining a clear preference for subjective agreement, but advocate for a mixed approach, one where subjective and objective elements interact (see Recital 25 of the Directive). Article 7 ‘Objective requirements for conformity’ reads: 1.

In addition to complying with any subjective requirement for conformity, the goods shall: (a) be fit for the purposes for which goods of the same type would normally be used, taking into account, where applicable, any existing Union and national law, technical standards or, in the absence of such technical standards, applicable sector-specific industry codes of conduct (emphasis added).

This is an amazing move, in particular as the recitals do not provide for any further explanation or justification. So far the EU Directive on consumer sales seems to be the only sales law at least in Europe that ties contractual conformity to technical standards and industry codes of conduct (Faber 2019). The wording ‘taking into account’ makes clear, however, that the courts are not bound to the technical standards of codes, and that there is not even a presumption of conformity in case the products comply with the standards. At the time of writing in 2022, it is open how the national courts will react to Article 7 and whether they are ready to take technical standards as an objective requirement seriously. Two decades ago, Schepel and Falke provided for insights into the case law of Austria, Belgium, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Luxemburg, Netherlands, Norway, Portugal, Spain, Switzerland/Liechtenstein and the United Kingdom (Schepel and Falke 2000a and 2000b; and Falke 2000). In their comparative analysis, they argued that national courts are ready to accept that contracts may refer to technical standards either implicitly or explicitly. Technical standards are automatically binding only in the latter case, but courts may – depending on the circumstances of the case – come to the conclusion that the parties were silently relying on technical standards as a benchmark. However, there are no empirical data on the use and reference to technical standards through courts or through complaint handling outside courts, which would allow to make any statement on the practical reach of technical standards in the relevant business sectors, let alone on the relationship between b2b and b2c contracts. To summarize, there are two ways of looking at the role and importance of technical standards in contracts. The first is the one which by and large dominates the legal discourse: what matters is the agreement between two parties on what the product should be used for. Such a view remains within the ideology of contract law – it is for the parties to individually engage with the quality of the product. The second is stressing the need to dive deeper into the contractual underground, into the practical relevance of technical standards. In such a perspective what really matters is that most of the products sold over the counter are mass products, produced along the line of the New Approach – the voluntary harmonized European technical standard sets the benchmark as compliance guarantees access to the Internal Market. There is not much room for individual negotiations, certainly not in b2c contracts. But there is another

Soft law, technical standards and European private law  151 reason. The few cases which reach the courts do not tell much about the relevance of technical standards. B2b conflicts are only occasionally resettled before courts. Most of the conflicts never reach the courts, they are solved via arbitration or simply settled. 3.2

Tort Law

In tort law, the role and function of technical standards is more obvious due to the strong linkage between liability and product safety. Historically, the origins of the link between standardization and tort law are dating back to the use of electricity in the mass manufacturing in the late 19th century and to the rise of the consumer society between the two World Wars. Ever more products were electrified, which made it necessary for the producers to integrate safety into the manufacturing process. This was the beginning of the elaboration of product safety-related technical standards. That is why research on technical standards usually includes a description of the linkage between technical standards and tort liability (Marburger 1979, p. 429; Schepel and Falke 2000a, p. 231). The development in the EU can be characterized by drawing attention to the parallel development of product safety regulation and product liability. The 1985 New Approach was designed so as to include the elaboration of technical standards that respect ‘general safety requirements’ laid down in EU directives. The debate first turned around pressure vessels, machines and toys, but over time extended to broader areas of the economy in which product safety played a role (see Directive 2014/29/EU). The Commission had started originally from the premise that the New Approach regulates product safety exhaustively through product-related general safety requirements. However, it had to give way to consumer policy concerns. In 1992, the EU adopted the Product Safety Directive 92/59/EEC (1992), which was amended in 2001 and which is now replaced by the EU Product Safety Regulation. The fierce political debate which preceded the adoption of the 1992 Directive focused on the role and function of technical standards in the design of product safety and was largely disconnected from the implications on the liability regime. It was in essence a French-German conflict (Joerges et al. 1991). France had taken the lead in product safety regulation and had introduced the obligation that only safe products should be marketed without any reference to the use and place of technical standards. To the contrary, Germany relied on its Maschinenschutzgesetz, which in its national regulatory design came close to the New Approach – the interplay between binding general (safety) requirements and the elaboration of voluntary technical standards. Article 4 of the Directive mirrored the compromise. It prioritized ‘specific national rules’ over ‘technical standards’. This basic compromise was kept in the next amendment. Article 3 of the Directive 2001/95/EC (2001) did not touch the original compromise, but gave more weight to technical standards, which are elaborated by the European Standardisation bodies and then transposed into national technical standards. Article 7 of the Product Safety Regulation reverses the order and establishes a presumption of conformity if the products comply with European technical standards.8 Article 7: Presumption of safety 1. For the purpose of this Regulation, a product shall be presumed to be in conformity with the general safety requirement laid down in Article 5 in the following cases: (a) if it conforms to relevant European standards or parts thereof as far as the risks and risk categories covered are concerned, the references of which have been published in the Official Journal of the European

152  Research handbook on soft law Union in accordance with Article 10(7) of Regulation (EU) 1025/2012; (b) in the absence of European standards referred to in point (a), as regards the risks covered by health and safety requirements laid down in the law of the Member State where the product is made available on the market, if it conforms to such national requirements.

The paradigm shift reflects the original intention of the Commission, already enshrined in the New Approach in the 1980s. The growing number of European technical standards might have provided the necessary legitimacy to upgrade the role and function of technical standards in the legal design of product safety. The Directive 85/374/EEC (1999) on product liability complements the New Approach type directives and the Product Safety Directive. The Directive has not undergone any major revisions despite many attempts to initiate a political debate with the Member States on, inter alia, the potential addressees of product liability (Machnikowski 2016). There is no explicit link between the use and place of technical standards and the definition of a defective product (Article 6). So far, the European Court of Justice has had no opportunity to debate the degree to which technical safety standards should be taken into consideration (Micklitz 2014; with regard to the more recent case law, see also Micklitz and Kas 2018). One might assume, nevertheless, that the European Court of Justice will follow the commonly agreed reasoning of national courts. Compliance with technical standards leads to the ‘indication’ that the products are safe, but no more than that. Non-compliance leads to the presumption that the product is defective, however, again no more than that (Schepel and Falke 2000a, pp. 231 and 235–236 with regard to the Product Liability Directive). There remains a strong mismatch between the design of product safety in the Product Safety Directive and the definition of defect in the Product Liability Directive. The former refers to technical standards as the yardstick of compliance, the latter does not. However, there are also good reasons why technical standards have to be treated differently in liability cases. First, the technical standards may not be in line with the latest technological development. Second, technical standards are just one aspect in ex post factum analysis of what the manufacturer should have done to prevent risks. There is ample evidence from Austrian and German supreme courts who have refuted the presumption of conformity in light of the concrete circumstances (Schepel and Falke 2000a, pp. 231, 235–236, with reference to Austrian and German case law). A clarification is, however, needed. It needs to be kept in mind that, first, the Product Liability Directive is of limited importance in practice. National courts usually apply national tort law in product liability cases. Second, whilst technical safety standards have an undeniable role to play in tort litigation, there is no research that has systematically evaluated the degree to which technical safety standards play a role and whether the technical safety standards in question are European standards elaborated by the European standardization bodies. This would require an empirical analysis of the role and function of health and accident insurances, which seem to cover most of the restoration costs. Contrary to the political attempts of the Commission to harmonize sales law, it did not even try to present to the Member States a proposal to harmonize tort law. The Academic Draft Common Frame of Reference provided in Book VI rules on ‘Non-contractual liability arising out of damage caused to another’ (von Bar 2009). Chapter 1 lays down the fundamental provisions and the basic rule which requires ‘fault’, Chapter 3 defines accountability and reproduces the key elements of the Product Liability Directive, including the notion of defect. The back-

Soft law, technical standards and European private law  153 ground material prepared by Christian von Bar et al. (2006) does not indicate that technical standards played a role in the conceptualization of the draft rules on liability.

4.

JUDICIAL REVIEW OF TECHNICAL STANDARDS

Whilst the role and function of technical standards in contract and tort law litigation before courts remains rather opaque and hard to assess and qualify, European technical standards are coming under pressure through judicial review. The New Approach, later turned into the New Legislative Framework, is being challenged as a form of ‘law’ which comes under the jurisdiction of the European Court of Justice. The self-understanding of the ‘club members’ (van Gestel and Micklitz 2013), that is, all those involved in the elaboration of standards such as the national and the European standardization bodies, the business which is financing the adoption of standards, the EU Commission and the Member States, relied on the assumption that technical standards are immune from judicial review. The first line of defence can be summed up as follows: technical standards are not ‘law’, they are only defining mere technicalities. It is an argument that relates back to the late 19th century, when standardization became an integral part of the industrialization, and national administrations accepted the supremacy of business expertise and relied on it as an integral part of the economic order (Ladeur 2011). The second line of defence against judicial review asserts that technical standards, drafted by the standardization bodies and sold to the business, enjoy copyright protection. The legal recognition of copyright protection is said to underpin their non-law character. Law needs to be publicly accessible. This is currently not the case, not even for harmonized standards. Technical standards have to be bought by the users. The third defence line relates to private self-regulation, to the autonomy of private parties and the voluntary character of technical standards. Seen through the lenses of private regulation, business is said to be autonomous to set up the institutional frame for collaboration in the making of technical standards, to define the procedural rules on who is allowed to participate and under what conditions and, last but not least, a possible voting procedure as well as potential dispute settlement mechanism in case of disagreements. This three-tier defence line has not changed over the last 100 years, although the character of standardization has considerably altered due to the integration of product safety and of quality requirements beyond mere measuring. It is amazing that the political shield lasted for so long in Europe, in the EU and the US (Vallejo 2021). There have, of course, been judicial attacks under property rights law from various sides. Consumer lawyers argue that technical standards are ‘law’ and law must be publicly accessible, in particular when standards intermingle with public policy objectives such as product safety. Industry circles are equally forcefully pursuing possible infringements of the copyrights. The EU’s General Court has not been ready to break up the system in recognizing technical standards as ‘law’, which would have excluded copyright protection and secured public availability so far (Case T-185/19 Public.Resource. Org and Right to Know v Commission; see also Volpato 2021). The judgment is not yet final (see the appeal Case C-588/21). It is equally amazing that the ‘law on unfair terms control’, which developed rapidly in Europe from the 1970s onwards was not instrumentalized to test the judicial review of technical standards. The Unfair Terms Directive 93/13/EEC (2011) covers standard terms and individually negotiated terms. At first glance, both variants exist in the field of technical standards.

154  Research handbook on soft law There are European standards which lay down a common level playing field and then there are individual technical standards, which serve as an alternative for businesses. Contrary to modern EU secondary law – it only suffices to look at the EU draft regulations on the legal framework for the digital economy – Directive 93/13/EEC does not provide for a definition of standard terms. Article 3 defines standard terms as those which are not individually negotiated, with little guidance of what exactly the characteristics of standard terms are. Article 1 requires that the standard terms are ‘related to a contract concluded between a seller or supplier and a consumer’. Article 7 intends to prevent the use of unfair standard terms, also in situations when they have become part of a contract between two individual parties. Does it then seem far-fetched to argue that technical standards which define the level of safety and maybe even the quality of a product may eventually be treated as standard terms which establish and/or specify rights and duties so that the analogous application of the Directive might be considered? Standardization of services brings technical standards even closer to contract law, because in service contracts the rights and duties have to be concretized due to the lack of appropriate default rules. Or does Article 1(2) of Directive 93/13/EEC apply so that ‘contractual terms which reflect mandatory statutory or regulatory provisions and the provisions or principles of international conventions’ are excluded … from the scope of application (for the problems of this exemption, see Micklitz 2014, p. 136)? To my knowledge, neither public authorities nor consumer organizations which have been given legal standing to exercise control, have ever considered using the Unfair Contract Terms Directive as means to break up the shield which currently protects technical standards from judicial control. In sports law (just think of case C-415/93 Bosman), FIFA rules – which are in fact voluntary collective agreements – could be submitted to judicial review (Komisar 1994). One might wonder whether something similar is currently happening in the field of technical standards. The European Court of Justice has not yet tested the applicability of the Unfair Terms Directive, but has begun to examine whether and under what conditions technical standards should be regarded as ‘law’ which has to respect primary and secondary EU law. Fra.bo (Case C-171/11) was the first blow. The European Court of Justice had to decide whether technical standards are apt to restrict free movement. The Court confirmed the applicability of Article 30 TFEU, holding that the standardization bodies were exercising public powers in concreto. The facts behind the case, however, tell a different story. There is a private law conflict lying behind the preliminary reference sent by the German court. In essence, Fra.bo was a conflict between two competitors, a German and an Italian, where the German competitor used German technical standards to block access of the Italian competitor to the German market. After the delivery of the judgment of the European Court of Justice, the Italian competitor claimed compensation before German courts. Once the Court of Appeal of Düsseldorf had decided in favour of Fra.bo, the Italian competitor, the two parties settled the case (see Case VI-2 U (Kart) 15/08).9 The second blow and maybe the much more important one came with James Elliott (Case C-613/14, paras. 34 and 40). The Court declared that harmonized standards are ‘part of EU law’ as measures implementing or applying an act of EU law. Again the facts are of utmost interest. The case concerned an action for contractual liability brought by James Elliott Construction against Irish Asphalt, its supplier of a particular type of product aggregate to be used for building the internal floors of a Youth Facility in Dublin (specified in the contract as ‘Clause 804 hardcore’). Once the facility was delivered, and after cracks appeared on the floor

Soft law, technical standards and European private law  155 and ceilings that rendered the building unusable, James Elliott Construction had to perform remedial work amounting to EUR 1.5 million. The damage in question was caused by the presence of pyrite in the Clause 804 aggregate supplied by Irish Asphalt, which was not in compliance with the underlying European standard.10 That is why James Elliott sought compensation from Irish Asphalt. Irish Asphalt in turn tried to use the claimed compliance with the European technical standards as a shield against the compensation claim. The European Court of Justice confirmed that it has jurisdiction over harmonized standards. In so far it looks as if harmonized standards are now subject to judicial review. However, the jurisdiction of the Court is far from being complete. The Court qualified harmonized standards as acts that cannot be ascribed to EU ‘institutions, bodies, offices or agencies’, but that nevertheless ‘implement or apply an act of EU law’. Therefore, the Court established that it only has ‘jurisdiction to interpret [these] acts’ (Case C-613/14, para. 34). One may speculate whether James Elliott is the first step into a legality test of technical standards. If this would be the case, technical standards, more precisely harmonized European standards, could be tested against primary and secondary EU law (Hettne 2017; Lundqvist 2017; Gnes 2017; Medzmariashvili 2017). Both Fra.bo and James Elliott tie the judicial control to the ‘public character’ of particular types and forms of technical standards (i.e., harmonized standards). Therefore, the European Court of Justice makes it difficult to extend its adjudication to technical standards where the private power prevails or where the technical standards are voluntary industry standards. The latter two could only be submitted to judicial control if they are regarded as standard terms in the meaning of Directive 93/13/EEC. In conclusion, currently private technical standards benefit from a regulatory gap. They are not covered by the Fra​.bo/​James Elliott doctrine and the applicability of the Unfair Terms Directive is doubtful.

5.

THE NEXT MOVE: TECHNICAL STANDARDS AND PRIVATE LAW IN THE DIGITAL ECONOMY

At the time of writing in 2022, it is not clear what kind of changes will be introduced in the Commission proposal on the AIA in the trialogue. However, it is equally clear that the EU, the Member States, business and civil society organizations are by and large in favour of some sort of a binding regulatory framework for AI, with conflicting perspectives. My hypothesis is that the rationale will not change, and this rationale follows the logic of the current interactions between technical standards and private law in the old economy. The five proposals, which might be complemented by an Al Liability Act, start from a twofold premise: first, the EU contract law acquis does not need to be complemented through EU further regulation and second, the national contract law rules and the Brussels Regulation and the Rome I and Rome II Regulations can properly handle also cross-border transactions. However, 1957 is different from 2022. The EU failed to adopt a European Civil Code, a Common European Sales Law and appropriate rules on contracts and liability for services, although there are valid economic and political reasons to have at least a common set of principles, which facilitates cross-border exchange and contributes to the building of a European society (Hesselink 2021). Moreover, the digital economy has established a new type of contract, which Palka (2018) has termed ‘as if’ contracts, because the customers interact with the service provider without a contract that meets the idea and the ideology of two

156  Research handbook on soft law parties which negotiate and agree on a common content. Article 3 of the Omnibus Directive (EU) 2019/2161 (2019) treats data as a form of remuneration, breaking away from a traditional understanding of contract law, where a contract presupposes the existence of an agreement on the price to be paid. In its proposals the EU Commission follows the regulatory approach of the New Approach/ New Legislative Framework. In exactly this type of thinking the DGA, the DSA and the Artificial Intelligence Act lay down ‘binding general requirements’ which should then be concretized through technical standards. Technical standardization serves as a substitute of further Europeanization of contract law (Ebers 2022; Helberger et al. 2021, p. 17). The DGA and the DSA rely on ‘voluntary industry standards’, the Artificial Intelligence Act and the Data Act on ‘harmonised European standards’. The European standardization bodies – this is the decisive move – are requested not only to take the physical integrity of the customers into account, but in the Artificial Intelligence Act also the psychological integrity and in all five proposals the respect for the Charter of Fundamental Rights. Technical standardization is no longer the business of engineers but also of psychologists and lawyers. The five proposals strengthen the contractual underground through the promotion of technical standards in the digital economy. Entrusting private standardization bodies with ever more public policy issues, first physical safety, now psychological safety and the respect for fundamental rights, does not only raise questions as to the legitimacy of EU co-regulation (Eliantonio and Cauffmann 2017; Lee 2018). It also demonstrates how the EU is compensating for the lack of harmonized private law rules through an ever-stronger reliance on technical standards, now in the form of algorithms, which remain by and large exempted from any form of judicial control. This is all the more true, as the Artificial Intelligence Act will be of very limited importance in practice (Veale and Borgesius 2021), which means that voluntary industry standards will be the rule and harmonized standards the exception.

NOTES 1. We – the research team of my ERC grant on European Regulatory Private Law developed the argument jointly in various economic sectors, mainly in energy (Almeida), telecommunications (Cantero Gamito), and financial services (Della Negra and Sedano). 2. In the late 1990s, Schepel and Falke had been mandated by the European Commission and the European Free Trade Association to execute a comparative study on ‘Legal Aspects of Standardisation’ (see Schepel and Falke 2000a, Schepel and Falke 2000b, and Falke 2000). The study deserves particular attention. It is the only attempt to comparatively evaluate, inter alia, the relationship between technical standardization and private law. To the best of my knowledge, I am not aware of any update that comes even close to the comprehensive comparative analysis undertaken by them. 3. The so-called New Approach was approved by the Council on 7 May 1985 in its Resolution on a New Approach to technical harmonization and standards (OJ, 1985, C136, p. 1). In 2008, this approach was updated by the so-called New Legislative Framework, which comprises Regulation (EC) 765/2008 setting out the requirements for accreditation and the market surveillance of products (OJ, 2008, L218, p. 30), Decision 768/2008 on a common framework for the marketing of products (OJ, 2008, L218, p. 82), and Regulation (EU) 2019/1020 on market surveillance and compliance of products (OJ, 2019, L169, p. 1). 4. However, the voluntary standards may unfold a de facto binding effect. See, e.g., Case C-171/11 Fra.bo ECLI:EU:C:2012:453. 5. British Standards Institution, National Bureau of Standards (today National Institute of Standards and Technology), Deutsches Institute für Normung, Association Française de Normalisation.

Soft law, technical standards and European private law  157 6.

Article 35 CISG: (1) The seller must deliver goods which are of the quantity, quality and description required by the contract and which are contained or p­ ackaged in the manner required by the contract. (2) Except where the parties have agreed otherwise, the goods do not conform with the contract unless they: (a) are fit for the purposes for which goods of the same description would ordinarily be used; (b) are fit for any particular purpose expressly or impliedly made known to the seller at the time of the conclusion of the contract, except where the circumstances show that the buyer did not rely, or that it was unreasonable for him to rely, on the seller’s skill and judgement; (c) possess the qualities of goods which the seller has held out to the buyer as a sample or model; (d) are contained or packaged in the manner usual for such goods or, where there is no such manner, in a manner adequate to preserve and protect the goods. (3) The seller is not liable under subparagraphs (a) to (d) of the preceding paragraph for any lack of conformity of the goods if, at the time of the conclusion of the contract, the buyer knew or could not have been unaware of such lack of conformity. 7. For a stock-taking of the scholarly debate on the interaction between subjective and objective requirements, see von Bar, Clive and Schulte-Nölke (eds.) 2006, section IV.A – 2:302 (p. 1287); and Hondius et al. 2008, p. 191. Both documents lack a reference to the function of technical standards. 8. Brussels, 30.6.2021. COM(2021) 346 final. 9. The judgement is only available in German. There is, however, a website in English set up by Fra.bo which explains the history, the facts and the outcome. See: Frabo., (2022). The Fra.bo Case [online]. Frabo. [Viewed 19 April 2022]. Available from: https://​www​.frabo​.com/​eng/​frabo/​the​-frabo​-case. 10. The High Court held that the product was not fit for purpose, the Supreme Court held instead that the product lacked merchantable quality, for details see Case C-613/14 James Elliott [2016] EU:C:2016:821, paras 23–29; and the Opinion of AG Campos Sanchez-Bordona, paras 25–29.

REFERENCES de Almeida, L., (2017). Integration through self-standing European private law: insights from the internal point of view to harmonization in energy market. PhD thesis. European University Institute. [Viewed 13 April 2022]. Available from: https://​doi​.org/​10​.2870/​23996. von Bar, C., Clive, E. and Schulte-Nölke, H., (2006). Principles of European Law Study Group on a European Civil Code, Non-Contractual Liability Arising out of Damage Caused to Another [online]. Interim edition. [Viewed April 2022] Available from: https://​ www​ .ccbe​ .eu/​ fileadmin/​ speciality​ _distribution/​public/​documents/​EUROPEAN​_PRIVATE​_LAW/​EN​_EPL​_20100107​_Principles​_​ _definitions​_and​_model​_rules​_of​_European​_private​_law​_​-​_Draft​_Common​_Frame​_of​_Reference​ _​_DCFR​_​.pdf. von Bar, C., (2009). Non-Contractual Liability Arising out of Damage Caused to Another. München: Sellier European Law Publishers. von Bar, C. et al., eds., (2009). Principles, Definitions and Model Rules of European Private Law: Draft Common Frame of Reference. München: Sellier European Law Publishers. Basedow, J., (2021). EU Private Law, Anatomy of a Growing Legal Order. Cambridge: Intersentia. Büthe, T. and Mattli W., (2011). The New Global Rules, The Privatisation of Regulation in the World Economy. Princeton: Princeton University Press. Cafaggi, F., (2006). Rethinking Private Regulation in the European Regulatory Space. In: F. Cafaggi, ed. Reframing Self-Regulation in European Private Law. Alphen aan den Rijn: Kluwer Law International. pp. 3–75. Cantero Gamito, M., (2015). The private law dimension of the EU regulatory framework for electronic communications: evidence of the self-sufficiency of European regulatory private law. PhD thesis.

158  Research handbook on soft law European University Institute. [Viewed 13 April 2022]. Available from: http://​hdl​.handle​.net/​1814/​ 37647. Case C-415/93 Union royale belge des sociétés de football association and Others v Bosman and Others ECLI:EU:C:1995:463 Case C-171/11 Fra.bo ECLI:EU:C:2012:453. Case C-613/14 James Elliott Construction EU:C:2016:821. Case C-588/21 Public.Resource.Org and Right to Know v Commission and Others (case in progress). Case T-185/19 Public.Resource.Org and Right to Know v Commission ECLI:EU:T:2021:445. Judgment of Oberlandesgericht Düsseldorf, VI-2 U (Kart) 15/08 [online]. [Viewed 19 April 2022]. Available from: http://​www​.justiz​.nrw​.de/​nrwe/​olgs/​duesseldorf/​j2013/​VI​_2​_U​_Kart​_15​_08​_Grund​ _und​_Teilurteil​_20130814​.html. CISG Advisory Council., (2018). Standards and Conformity of the Goods under Article 35 CISG [online]. CISG Advisory Council Opinion No 19. [Viewed 14 April 2022]. Available at: https://​www​ .cisgac​.com/​file/​repository/​CISG​_Advisory​_Council​_Opinion​_No​_19b​.pdf. Contreras, L., ed., (2019). Cambridge Handbook on Technical Standardisation, Further Intersections of Public and Private Law. Cambridge: Cambridge University Press. Council Directive 85/374/EEC on the approximation of the laws, regulations and administrative provisions of the Member States concerning liability for defective products (1999). Consolidated version. Official Journal. L141, 4 June 1999, 1–7. Council Directive 92/59/EEC on general product safety (1992). Official Journal. L228, 11 August 1992, 64–71. Council Directive 93/13/EEC on unfair terms in consumer contracts (2011). Consolidated version. Official Journal. L95, 21 April 1993, 29–34. Davies, G., (2013). Freedom of Contract and the Horizontal Effect of Free Movement Law. In: D. Leczykiewicz and S. Weatherill, eds. The Involvement of EU Law in Private Law Relationships. Oxford: Hart Publishing. pp. 53–70. Della Negra, F., (2018). MiFID II and Private Law: Enforcing EU Conduct of Business Rules. Oxford: Hart Publishing. Directive 1999/44/EC of the European Parliament and of the Council on certain aspects of the sale of consumer goods and associated guarantees (1999). Official Journal. L171, 7 July 1999, 12–16. Directive 2001/95/EC of the European Parliament and of the Council on general product safety (2001). Official Journal. 2002. L011, 4–17. Directive 2006/123/EC of the European Parliament and of the Council on services in the internal market (2006). Official Journal. L376, 27 December 2006, 36–68. Directive 2011/83/EU of the European Parliament and of the Council on consumer rights (2011). Consolidated version. Official Journal. L304, 22 November 2015, 64–88. Directive 2014/29/EU of the European Parliament and of the Council on the harmonisation of the laws of the Member States relating to the making available on the market of simple pressure vessels (2014). Official Journal. L96, 29 March 2014, 45–78. Directive (EU) 2019/771 of the European Parliament and of the Council on certain aspects concerning contracts for the sale of goods (2019). Official Journal. L136, 22 May 2019, 28–50. Directive (EU) 2019/2161 of the European Parliament and of the Council as regards the better enforcement and modernisation of Union consumer protection rules (2019). Official Journal. L328, 18 December 2019, 7–28. Ebers, M., (2022). Standardizing AI – The Case of the European Commission’s Proposal for an Artificial Intelligence Act. In: L. A. DiMatteo, M. Cannarsa and C. Poncibò, eds. The Cambridge Handbook of Artificial Intelligence: Global Perspectives on Law and Ethics. Cambridge: Cambridge University Press. Eliantonio, M. and Cauffmann, C., eds, (2020). The Legitimacy of Standardisation as a Regulatory Technique, A Cross-Disciplinary Multi-level Analysis. Cheltenham: Edward Elgar Publishing. European Commission., (2021a). Roadmap for a Standardisation Strategy [online]. [Viewed 14 April 2022]. Available from: https://​ec​.europa​.eu/​info/​law/​better​-regulation/​have​-your​-say/​initiatives/​ 13099​-Standardisation​-strategy​_en. European Commission., (2021b). Proposal for a Regulation of the European Parliament and of the Council Laying Down Harmonised Rules on Artificial Intelligence (Artificial Intelligence Act)

Soft law, technical standards and European private law  159 [online]. COM(2021) 206 final. [Viewed 14 April 2022]. Available from: https://​eur​-lex​.europa​.eu/​ legal​-content/​EN/​TXT/​?uri​=​CELEX​%3A52021PC0206. European Commission., (2022a). An EU Strategy on Standardisation [online]. COM(2022) 31 final. [Viewed 14 April 2022]. Available at: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​?uri​=​CELEX​ %3A52022DC0031. Faber, W., (2019). Bereitstellung und Mängelbegriff. In J. Stabentheimer, C. Wendehorst and B. Zöchling-Jud, Hrsg. Das neue europäische Gewährleistungsrecht. Wien: Manz. pp. 63–110. Falke, J., (2000). Legal Aspects of standardisation in the Member States of the EC and EFTA, Country Report Germany (in German) Volume 3. Belgium: Office for Official Publications of the European Communities. van Gestel, R. and van Lochem, P., (2020). Private Standards as Replacement for Public Law Making. In: M. Cantero Gamito and H.-W. Micklitz, eds. The Role of the EU in Transnational Legal Ordering: Standards, Contracts and Codes. Cheltenham: Edward Elgar Publishing. pp. 27–54. van Gestel, R. and Micklitz, H.-W., (2013). European Integration through Standardisation: How Judicial Review is breaking down the Club House of Private Standardisation Bodies. Common Market Law Review. 50(1), 145–182. Gnes, M., (2017). Do Administrative Law Principles Apply to European Standardization: Agencification or Privatization? Legal Issues of Economic Integration. 44(4), 367–380. Helberger, N., Micklitz H.-W., Lynskey. O, Rott, P., Sax, P and Strycharz, J., (2021). EU Consumer Protection 2.0: Structural Asymmetries in Digital Consumer Markets. A Joint Report from research conducted under the EUCP2.0 Project. BEUC, March 2021, 207 pages. Hesselink, M. W., (2021). Justifying Contract in Europe, Political Philosophies of European Contract Law. Oxford: Oxford University Press. Hettne, J., (2017). Standards, Barriers to Trade and EU Internal Market Rules: Need for a Renewed Approach? Legal Issues of Economic Integration. 44(4), 409–421. Hondius, E. et al., (2009). Principles of European Law: Study Group on a European Civil Code. München: Sellier European Law Publishers. Joerges, C. et al., (1991). European Product Safety, Internal Market Policy and the New Approach to Technical Harmonisation and Standards. EUI Working Paper Law No. 10-14. [Viewed 13 April 2022]. Available at: http://​hdl​.handle​.net/​1814/​46244. Komisar, N., (1994). Imperfect Alternatives. Choosing Institutions in Law, Economics, and Public Policy. Chicago: University of Chicago Press. Knieper, R., (1981). Sachmängelgewährleistung und Verbraucherschutz. In: R. Knieper, ed. Zwang, Vernunft, Freiheit, Studien zur Juristischen Konstruktion der bürgerlichen Gesellschaft, Europäische Verlagsanstalt. Frankfurt a. M: Europäische Verlaganstalt. pp. 144–156. Ladeur, K.-H., (2011). The Evolution of General Administrative Law and the Emergence of Postmodern Administrative Law. Osgoode Hall Law School Legal Studies Research Paper Series No. 16/2011. [Viewed 13 April 2022]. Available from: dx​.doi​.org/​10​.2139/​ssrn​.1792062. Lee, K., (2018). The Legitimacy and Responsiveness of Industry Rule-Making. Oxford: Hart Publishing. van Leeuwen, B., (2017). European Standardisation of Services and its Impact on Private Law. Oxford: Hart Publishing. Lundqvist, B., (2014). Standardisation under EU Constitution Rules and US Antitrust Rules, the Rise and Limits of Self-Regulation. Cheltenham: Edward Elgar Publishing. Lundqvist, B., (2017). European Harmonized Standards as “Part of EU Law”: The Implications of the James Elliott Case for Copyright Protection and, Possibly, for EU Competition Law. Legal Issues of Economic Integration. 44(4), 421–437. Machnikowski, P., ed., (2016). European Product Liability – An Analysis of the State of the Art in the Era of New Technologies. Cambridge: Intersentia. Marburger, P., (1979). Die Regeln der Technik im Recht. Tübingen: Mohr Siebeck. Medzmariashvili, M., (2017). Delegation of Rulemaking Power to European Standards Organizations: Reconsidered. Legal Issues of Economic Integration. 44(4), 353–367. Michaels, R. and Jansen, N., (2006). Private Law Beyond the State? Europeanization, Globalization, Privatization. The American Journal of Comparative Law. 54(4), 843–890.

160  Research handbook on soft law Micklitz, H.-W., (2007). Services Standards: Defining the Core Elements and Their Minimum Requirements [online]. A study commissioned by ANEC. [Viewed 13 April 2022]. Available from: http://​www​.anec​.eu/​attachments/​ANEC​-R​&​T​-2006​-SERV​-004final​.pdf. Micklitz, H.-W., (2010)., Failures or Ideological Preconceptions? Thoughts on Two Grand Projects: The European Constitution and the European Civil Code. In: K. Tuori and S. Sankari, eds. The Many Constitutions of Europe. Farnham: Ashgate. pp. 109–142. Micklitz, H.-W., (2014). Liability for Products and Services. In: N. Reich et al., eds. European Consumer Law. 2nd edn. Cambridge: Intersentia. Micklitz, H.-W., (2023). The Role of Technical Standards in the EU Digital Policy Legislation. Available from: https://​www​.beuc​.eu/​sites/​default/​files/​publications/​BEUC​-X​-2023​-096​_The​_Role​ _of​_Standards​_in​_Future​_EU​_Digital​_Policy​_Legislation​.pdf. Micklitz, H.-W. and Kas, B., (2018). Rechtsprechungsübersicht zum Europäischen Vertrags-und Deliktsrecht (2014–2018). Europäisches Wirtschafts- und Steuerrecht 2018/4. pp. 181–219 and 241–300. Micklitz, H.-W., Helberger, N. and Rott, P., (2021). EU Consumer Protection 2.0: The Regulatory Gap: Consumer Protection in the Digital Economy [online]. Addendum to the report ‘Structural Asymmetries in Digital Consumer Markets’. [Viewed 13 April 2022]. Available from: https://​www​ .beuc​.eu/​publications/​beuc​-x​-2021​-116​_the​_regulatory​_gap​-consumer​_protection​_in​_the​_digital​ _economy​.pdf. Palka, P., (2018). Terms of Service are not Contracts – Beyond Contract Law in the Regulation of Online Platforms. In: S. Grundmann, ed. European Contract Law in the Digital Age. Cambridge: Intersentia. pp. 135–163. Regulation (EU) No 1025/2012 of the European Parliament and the Council on European standardization (2012). Official Journal. L316, 14 November 2012, 12–33. Regulation (EU) 2022/868 of the European Parliament and of the Council of 30 May 2022 on European data governance and amending Regulation (EU) 2018/1724 (Data Governance Act). L 152, 3 June 2022, 1–44. Regulation (EU) 2022/1925 of the European Parliament and of the Council (2022) on contestable and fair markets in the digital sector and amending Directives (EU) 2019/1937 and (EU) 2020/1828 (Digital Markets Act). L 2022, L 265, 10 October 2022, 1-66. Regulation (EU) 2022/2065 of the European Parliament and of the Council on a Single Market for Digital Services and amending Directive 2000/31/EC (Digital Services Act) L 277, 27 October 2022, 1–102. Regulation (EU) 2023/988 of the European Parliament and of the Council on general product safety, amending Regulation (EU) No 1025/2012 of the European Parliament and of the Council and Directive (EU) 2020/1828 of the European Parliament and the Council, and repealing Directive 2001/95/EC of the European Parliament and of the Council and Council Directive 87/357/EEC L 135, 23 May 2023, 1–51. Schepel, H. and Falke, J., (2000a). Legal Aspects of Standardisation in the Member States of the EC and EFTA, Volume 1, Comparative Report. Belgium: Office for Official Publications of the European Communities. Schepel, H. and Falke, J., (2000b). Legal Aspects of Standardisation in the Member States of the the EC and EFTA, Country Reports, Volume 2. Belgium: Office for Official Publications of the European Communities. Schepel, H., (2005). The Constitution of Private Governance. Oxford: Hart Publishing. Schwenzer, I., (2013). Art. 35 Rn. In: P. Schlechtriem and I. Schwenzer, Hrsg. Kommentar zum Einheitlichen UN-Kaufrecht – CISG. 6. Auflage. München: C.H. Beck. Sedano, E., (2022). Financialisation and the Vanishing of the Rule of Law, PhD European University Institute 2022. Stürmer, M., (2021). Europäisches Vertragsrecht, Instutionelle und methodische Grundlagen, materielles Recht, Kollisionsrecht. Berlin: Degruyter. United Nations Convention on Contracts for the International Sale of Goods (CISG), (1980). Vienna, signed 11 April 1980, in force 1 January 1988, 1489 UNTS 3. Vallejo, R., (2021). The Private Administrative Law of Technical Standardization. Yearbook of European Law. 40, 172–229.

Soft law, technical standards and European private law  161 Veale, M. and Borgesius, F., (2021). Demystifying the Draft EU Artificial Intelligence Act. Computer Law Review International. 22(4), 97–112. Vieweg, K., (2021). Technische Normung und Wirtschaftsrecht – Zu den ethischen Anforderungen beim Einsatz Künstlicher Intelligenz, Deutsches, Europäisches und Vergleichendes Wirtschaftsrecht. In: B.P. Paal et al., eds. Festschrift für W.F. Ebke zum 70 Geburtstag. München: C.H. Beck. pp. 1025–1035. Volpato, A., (2021). ‘Part of EU law’, but only partially: the issue of the accessibility of harmonised www​ .​ standards [blog post]. Maastrich University. 15 October 2021. Available from: https://​ maastricht​university​.nl/​blog/​2021/​10/​%E2​%80​%9Cpart​-eu​-law​%E2​%80​%9D​-only​-partially​-issue​ -accessibility​-harmonised​-standards. Wilhelmsson, T., (2008). The Contract Law Acquis: Towards More Coherence Through Generalisations? 4. Europäischer Juristentag. Wien, 3.– 5. Mai 2007. Wien: Manz. pp. 111–145.

11. Soft authority in global governance Jan Klabbers

1. INTRODUCTION The young Keynes, having attended the 1919 Versailles Conference, issued a stern warning about its economic ramifications, something other participants were much in agreement with (Nicolson [1933] 1964). On the final page of Economic Consequences of the Peace, he dedicated his book ‘to the formation of the general opinion of the future’ (Keynes 1920, p. 279). A decade and a half later, the same Keynes wrote his famous General Theory very much aware that his readership (economists and political philosophers mostly) would come to influence public opinion (Keynes [1936] 1970, pp. 383–384). At least one of his biographers accordingly credits him, somewhat loosely perhaps, with having been a forerunner of expertise-based governance (Davenport-Hines 2015, p. 107). Not only did Keynes as an individual aim to influence the hearts and minds of people and in this way aim to steer public debate in favour of some policies rather than others, he also advocated governance by experts more generally, on the basis of academic or scientific insights and understandings instead of ideology or vague intuition. In our day and age, and in international (and European Union (EU)) law, governance by experts and based on expertise is often, regrettably, classified under the label ‘soft law’ – although it seems that there is an increasingly widespread realization that the label does more harm than good.1 In what follows, I will discuss the relevance of labels and the use of language (section 2), why soft law was so popular and how it played out in international law since the 1970s (section 3), some of its pitfalls (section 4), and describe the move to epistemic authority (section 5), complete with examples thereof drawn from international organizations (section 6). Finally, section 7 concludes. Two caveats are in order. The first is that contrary to prima facie impressions perhaps, my earlier writings on soft law do not make radically different points. Those earlier writings, largely written 25 to 15 years ago,2 already question the wisdom of using the label soft law, but never doubt its ontology: it was clear then, and is clear now, that many of the individuals and institutions engaged in global governance engage in a practice they and others think of, and refer to, as soft law. In a sense, therewith, the current chapter picks up where those earlier writings trailed off: the earlier pieces criticized the use of the term soft law tout court; the current chapter aims to provide an alternative perspective, proposing to replace thinking of soft law with thinking of ‘epistemic governance’ or ‘epistemic authority’.3 The second caveat is that what follows will relate mostly to international organizations and their activities. These are not the only actors exercising epistemic authority or producing what some may call ‘soft law’, but they are among the more important ones. Some manifestations may stem from private associations (see Büthe and Mattli 2011); some may come from ad hoc conferences involving states or other actors;4 but for the better part, the sort of thing this chapter occupies itself with, stems from international organizations. 162

Soft authority in global governance  163

2.

ON LANGUAGE

In the various language games played by students of international affairs, labels carry considerable importance, substantively as well as in terms of the sociology of scholarship. Individual scholars flock to certain labels because of what those labels stand for, and lawyers will call themselves ‘legal realists’ or ‘neo-positivists’ or ‘rationalists’ or whatever may be in vogue. In turn, those labels derive further meaning from being adopted and adapted by those same scholars. While it is no longer true that each and every scholar in International Relations wants to be called a ‘Realist’, this is largely because ‘Realism’ (with capital R) has come to be associated with a set of assumptions and tropes that are too specific – and too unrealistic, really – to command universal attraction. A Realist is someone who believes that states are the central, indeed only relevant, actors in international politics. A Realist is someone who believes that states are into increasing their power; and a Realist is someone who believes that rules have no independent compliance push or compliance pull, unless backed up by force. And Realists tend to believe that force and power are largely measured in military terms, in terms of rockets and battalions. Not all International Relations scholars accept these assumptions, but the very term realism (with a small r, without the underlying assumptions) is still considered appealing: witness the emergence of ‘neo-realism’ and even, almost a contradiction in terms, ‘critical realism’ (on the latter, see Patomäki 2002). The main point to make, however, is that no one wants to be seen as the opposite of a realist (see Jütersonke 2016). Where once the world of International Relations scholars was divided into ‘realists’ (who would become Realists as the discipline matured) and idealists, the latter did not become Idealists with capital I, but instead grouped themselves under different labels: liberal institutionalists most prominently, but also, even if more marginally, Internationalists, and perhaps also the Constructivists – whose affinities with Realism are very limited indeed. No one wants to be seen as an Idealist; to be an Idealist is to be associated with fairy tales, with castles in the sky, with a lack of understanding of how the world actually works. Liberal Institutionalists, Internationalists, and Constructivists may all believe that there is more to global politics than the state; that there is an independent role to play for global and regional institutions; that international rules may have a signalling function or facilitative function even if not immediately backed up by power; and that power is not solely to be measured in military terms. They may share these assumptions, the counterpart to Realism’s assumptions, but are loath to be referred to by the term that in everyday language serves as realism’s opposite: idealism. In this way, Realism sets the tone and dictates the debate, despite being not very ‘realist’, really, as it should be clear to even the most casual and uneducated observer that some institutions play a major role (the United Nations (UN), the EU), and that some rules are of pivotal importance in propping up a system of states: think of the rules on statehood and recognition or the rules on jurisdiction, not to mention the right to self-defence. If idealism is the bête noire of International Relations scholarship, for international lawyers the worst insult possible is being labelled a ‘legalist’, and it comes to some extent from the same source. Legalism, in Shklar’s well-known definition, is ‘the ethical attitude that holds moral conduct to be a matter of rule following, and moral relationships to consist of duties and rights determined by rules’ (Shklar [1964] 1986, p. 1). The charge will be obvious: it shares with Idealism the idea that rules are supremely relevant in global politics, something Realists of varying ilk will tend to deny. Inasmuch as International Relations scholars are driven to

164  Research handbook on soft law avoid the label Idealist, so international lawyers want to avoid being seen as legalists, for legalists, it will be clear, are too attached to rules, ignoring the world of practical politics, ignoring the role of power. International lawyers do not particularly mind being referred to as Utopian: that is a charge they can defend themselves against by claiming that there is little wrong with having visions of a better world sometime in the future, tapping into Oscar Wilde’s classic quip that a map of the world without Utopia is a map not worth looking at (quoted in Righter 1995, p. ix). If this taps into the classic distinction between the lex lata (the law as it is) and lege ferenda (the law as it should be),5 legalism, by contrast, lacks this future-oriented aspect: to be a legalist is not to think of some future world, but of today’s world, and is so by definition: the legalism charge has to be connected to the lex lata in order to have any force. Hence, the Legalist believes that rules cause action, and the Legalist believes that this is already the case, rather than some undefined future possibility. Obviously then, the Legalist denies the core assumptions and tropes so warmly embraced by the International Relations Realist. This confronts the international lawyer with something of a dilemma. The lawyer is, after all, naturally drawn to rules, and tends to infuse them with some relevance and more or less independent force. The good lawyer knows, after all, that rules do play a social role: family life is facilitated by family law, whether backed up by power or not. The rules of contract law make all sorts of transactions possible. Tort law helps us to behave reasonably in most everyday actions, whereas the rules of the road help to make it possible to move from location A to location B in safe and more or less predictable manner (see e.g., Schauer 1991). Rules, any but the most self-loathing lawyer knows, serve a useful social function, no matter what the Realists say (but see Goldsmith and Posner 2005). Rules may signal desirable behaviour even without full enforcement (Edelman [1964] 1985), and rules help construct our world (Onuf [1989] 2013). But since Realism dictates the terms of the debate and frames what counts as acceptable knowledge and thought, lawyerly knowledge alone is insufficient. The international lawyer needs to meet the Realist on the Realist’s terms and terrain, explaining and exhorting the value of rules without lapsing into Legalism, for Legalism, so the International Relations scholar and the lawyer agree, is best avoided. Law needs to be backed by power, and law needs to reflect power. And what is more, the international lawyer has realized this ever since International Relations turned itself from a branch of international law into a self-standing academic discipline, quickly overshadowing academic international law precisely because of its deference to power and politics. As a result, international lawyers have since the early twentieth century aspired to sensitize international law to Realism’s demands, in the guise of different approaches. In the 1910s and 1920s, Max Huber’s (1928) sociological jurisprudence carried the promise of reconciliation, later followed by Charles de Visscher (1968). At Yale, a jurisprudence was developed that placed international law at the service of liberal United States power, in the late 1980s followed by a less demanding ‘incidents’ approach (Reisman and Willard 1988). In the meantime, individual international lawyers wrote textbooks on power in international affairs (e.g., Fawcett 1982). For all their individual variations, the central message was remarkably constant: international law is subordinate to international power politics. At the end of the day, international law must accept the primacy of the political. Yet, it seemed clear that establishing a methodology for international law, whether as sociological jurisprudence or anything else, was unlikely to work, for the good but oft-misunderstood reason that law is not merely (like International Relations) an academic discipline but is itself political practice in a way that does not apply to the social sciences at

Soft authority in global governance  165 large. Like all law, international law is Janus-faced: models solely geared to the academic study of international law tend to miss something of importance while, conversely, studies only looking at practice may miss something of relevance as well. As a result, the discipline looked for a way of thinking that would have both academic traction and resonate in practice – it found what became known as soft law.

3.

ON TROPES

Starting in the 1970s, international lawyers, a little later followed by EU lawyers, came to praise the virtues of what they refer to as ‘soft law’ (see the seminal piece Wellens and Borchardt 1989). The precise reasons for doing so have always remained somewhat implicit, but the popularity of soft law owes something to the situation described above: the dominant hold of Realist thinking, the understanding that rules are part of the world but also the desire to avoid the charge of legalism. It probably also owes something to the illusion of certainty the term offers. Confronted with legal instruments that were not clearly treaties and which defied categorization into one of the recognized sources of international law, some label was considered necessary. A phenomenon not properly understood cried out for further understanding, and for a while it seemed the phrase ‘soft law’ could offer the much-needed solace. If not treaties, these instruments could not be considered law ‘properly speaking’, as Austin might have put it two centuries ago (Austin [1832] 1995). And yet, these instruments (resolutions, declarations, codes of conduct, guidelines) all emanated from concord between the relevant actors (i.e., the actors with law-making authority in international law: states, whether meeting ad hoc or having been assembled in the form of an international organization)6 and seemed aimed at achieving some kind of result: setting out a normative proposition, aiming to change how people act. So, while it seemed obvious that this output could not be considered law but nonetheless tried to fulfil much the same function, then how to make sense of it? The answer was as clever as it was deceptive. If not law, yet meant to do the same thing, then it stood to reason to distinguish between gradations of law. This was considered to do justice to the twin demands of catering to politics and respecting the role of rules, all the time avoiding the charge of legalism. What is more, already Austin had done much the same: while he denied that international law would be law proper, he nonetheless did not doubt it was binding as, in his words, ‘positive morality’. The French, during the 1970s, spoke of droit vert (‘green law’), suggesting that many of these instruments were somehow embryonic but could someday mature into real law, but the more graphic term that rapidly became popular was the term ‘soft law’. The term could be successful because it seemed to offer an explanation in conjunction with a relevant categorization, and it tapped into a widely felt intuition: the intuition that morally, ethically, politically, and socially, some things are more important than others. Somehow a rule on trade in lightbulbs is less relevant than the prohibition of genocide. Somehow a rule on exchanging economic statistics by a certain date seems less pivotal for human existence than the prohibition of torture. So, obviously, the conclusion was drawn, without giving it much thought, that law could come in gradations – law could thus be ‘hard’, and law could also, so it seemed, be ‘soft’. The picture then emerges of a double deception: soft law tries to explain a mysterious phenomenon (the production of all sorts of instruments of uncertain provenance and status) by invoking a mysterious term (‘soft law’).

166  Research handbook on soft law Soft law, as a term, promises different things to different people. Perhaps the most important consideration is that it taps into several standard tropes prevailing in international legal discourse (for discussion, see Kratochwil 1989) – and this might also help explain why the term has considerably more traction in international law than among domestic lawyers.7 International lawyers traditionally picture themselves as heroically engaged in a struggle to overcome the Hobbesian world as a place where man is wolf to man, and where life is nasty, brutish, and short. Consequently, among international lawyers, cooperation is highly valued, as it is cooperation which helps to overcome the law of the jungle. International lawyers have accordingly not been shy to pronounce that some agreement is always preferable to no agreement at all; and that a soft agreement is likewise preferable to no agreement whatsoever (see the seminal piece Chinkin 1989). The veracity of the topos, incidentally, is by and large irrelevant, and others might be tempted to seek relief in the opposite topos, something to the effect that going it alone is superior to having to depend on others. But be that as it may, the topos about the positive role of cooperation and agreement is very prominent among international lawyers, so much so that some have observed that for international lawyers, ‘the international’ itself is their political project (Kennedy 1994) – and in this fertile soil, soft law could come to great fruition. And this in turn hitches a ride on another topos: for the internationalist, any international solution is bound to be superior to anything devised nationally or parochially. If this helps explain the academic popularity of soft law, another topos of tremendous importance is the claim that the end justifies the means. This should not sway independent academics very easily (although it has done so) but is obviously of great value to the very people creating what they think of as soft law. The average civil servant must have been delighted with having found an instrument doing the same thing as hard law, but without some of the problems sometimes associated with law: lengthy procedures, democratic control, interference by uninformed parliamentarians or government ministers, or judicial review. For, if soft law is not ‘really’ law, it typically need not pass parliamentary scrutiny (although contrary arguments have been presented, see Hollis and Newcomer 2009), it can be concluded or promulgated quickly, and is unlikely to be subjected to judicial review – courts, on this reading, are unlikely to apply instruments that are not considered to be law. And underlying all this was a still deeper trope: the trope according to which law and politics are separate spheres of activity. If soft law promises to bring law and politics in harmony (to have a law that accurately reflects politics and gives expression to political sentiments and configurations, in itself a realist trope, see Jütersonke 2016), this presupposes that law and politics are, in the original position, separate spheres of activity, in which politics reigns supreme. This informs most cocktail party descriptions of the role of the Foreign Office legal advisor: having heard, ideally, advice about the legality (vel non) of prospective action, the decision-makers gather and take their decision. Something like this may well happen, of course, but misconstrues the relationship between law and politics: the political decision will typically be made in legal terms (‘Our action does not constitute forbidden aggression, but instead constitutes an exercise of the inherent right of self-defence’), and is even unthinkable except in legal vocabularies: it is difficult, e.g., to even think of military action in terms that do not have any legal connotation, unless one wants to succumb to highly abstract descriptions. The term ‘military action’ evokes few legal connotations but is neither concrete nor precise: it can mean anything from troop exercises to a full-blown invasion. By contrast, to label one’s action an invasion is radically different from calling it an intervention, or self-defence. All

Soft authority in global governance  167 these terms, indeterminate as they are themselves, nonetheless come with expectations, connotations, and associations.

4.

ON PITFALLS

For all its popularity, both the term soft law and the concept of soft law come with serious pitfalls. Most remarkably perhaps, in the half century since the term was first used, it has never been further clarified. The meaning of soft law was highly obscure in the early 1970s, and it remains highly obscure in 2022. The term has defied any definition other than in opposition to what it is not: everything that is not hard law can be soft law, but beyond this there are no boundaries, no limitations, no points beyond which the concept cannot meaningfully be invoked. In practice, it turns out infinitely flexible, capable of accommodating any normative utterance, regardless of contents, pedigree, procedure. And that, in turn, is difficult to reconcile with any standard notion of law. The adjective ‘soft’ in soft law is a gentle misnomer; better adjectives would be blurry ‘law’, or mushy ‘law’, or squishy ‘law’, with the word ‘law’ between inverted commas. Law which has no beginning and no end cannot be law, really, whatever else it can be: authority perhaps, influence, maybe even governance. But if the essence of law is some notion of validity, as many lawyers would intuitively accept, something that helps distinguish law from non-law, then soft law is a giant misnomer: soft it may be, but law it ain’t. If this is a conceptual problem, it should not be dismissed as merely a conceptual problem. After all, we think in concepts, and we live through concepts. Embracing soft law means, in effect, giving up on law, and therewith also on any possible Rule of Law – when we cannot tell law from non-law, references to the Rule of Law become meaningless drivel. The anti-positivist, anti-legalist ambition underlying soft law carries the day, and over the last 50 years has managed to sweep away cherished notions of law. What causes all this remains a bit of a mystery. Surely, few will have thought about intentionally undermining the very idea of law: there is no legal equivalent of Blofeld or Dr Evil to point fingers at. As noted, soft law derives some appeal from tapping into standard tropes, and it further seems attractive due to its empiricist gloss. The most obvious epistemological norm of modernity is empiricism: we tend to think that things are true if we experience them. We may know, rationally, that this is at best only partly true: few of us ever spot any atoms or viruses, and still, we believe they exist. No one alive today has seen Caesar cross the Rubicon or experienced the Spanish Inquisition, and yet few of us doubt these historical occurrences (see Becker 1955). So, while we are willing to accept epistemic authority, the authority of experts based on knowledge, whether physics professors or historians, we are nonetheless rather keen on experiencing things for ourselves. It is this tension that pervades our attitudes to expert governance: the tension between sensory experience and rational abstraction (Klabbers 2014). This applies all the more so since expert advice tends to be culturally coloured: pregnant women in Finland are told to avoid ginger, something unthinkable in Asia, while people in the protestant part of Europe generally are told to stay clear from alcohol, which elsewhere is considered part of a healthy diet, when consumed in moderation. Plus, we realize that rational abstraction by definition presents a ‘one-size-fits-all’ picture, an abstraction shorn of any context, whereas our own lives are by definition full of context. So, it may be the case generally that smoking causes cancer, but it

168  Research handbook on soft law need not do so in my case. And we all know someone who smoked like a chimney without ever getting sick, as well as some who never smoked and still got lung cancer. This fundamental ambivalence is perfectly compatible with governance through recommendations and guidelines. These tell us what would generally be good for us (according to those who issue them – always an important caveat), without actually insisting that we comply. There might be little pushes or incentives, but no coercion in the name of the law – it is no accident that ‘nudging’ has become a popular tool of governance (Sunstein 2014). As long as the experts give us general advice without interfering with our context, we seem perfectly willing to abide by the recommendations and guidelines. But things are different when we feel the experts aim to overrule our personal context, as the Covid crisis has taught – the Covid crisis, especially the vaccination discussion, therewith provides an intriguing real time laboratory. When the pandemic first broke out, personal context was rapidly ignored: people were dying, sometimes in large numbers, so an emergency arose, and it was clear that this was best left to experts. Indeed, not leaving it to experts was considered irresponsible, or even downright obnoxious, and where political leaders refused to take expert advice seriously, they were quickly castigated. Or rather, it is perhaps more accurate to say that some of those political leaders employed different priorities, and those other priorities vied for prominence. On the one hand, it rapidly became clear that lengthy lockdowns would have serious economic ramifications. On the other hand, there were the mental health effects of strict measures: self-isolation, quarantine and lockdown all imply solitude, the absence of social life, the absence of tactile experiences, and none of that, so people realized, is good for the soul. So, fairly quickly, the matter came to be one of compromise between various imperatives: physical health, mental health, economic well-being. Singling out one of these at the expense of the others became difficult to defend, and this in turn taught a valuable lesson about expert governance: governance can only rarely, in times of deep crisis, afford to ignore all context altogether. In the real world, all things are related to each other, and any regulator will have to reckon with this. Here then lies the great advantage of soft law: by not carving anything in stone, it allows for context to seep in and influence decision-making; it allows for regulators to balance different imperatives, stemming from different bodies and organs promulgating different goals and suggesting different, and possibly conflicting, behaviours on the part of the governed. There are prices to pay, as noted: the absence of certainty, the absence of parliamentary control, the absence of judicial control – and there are always some who would wish to prioritize some aspects over others, and subject smooth governance to some set of legal rules or another, e.g., by suggesting that the management of complex crises is subject to human rights standards, or to the prescriptions of international investment law, the rules of transport law, or any other body of rules that can possibly be made to apply. Here, also, then lies the great conundrum: to accept soft law is to say goodbye to law, in return for (if all goes well) a reasonably smoothly running society – and such return is nothing to frown at. Conversely, to reject soft law is to make such a smooth society much harder to achieve; it is to invite delays and political posturing and grandstanding; it is to invite litigation and drawing lines in the sand, operating on the difficult assumption that it might be possible to hold experts legally accountable (Klabbers 2021a). Quite possibly, soft law was the inevitable response to a society growing ever-more complex, when not even the flight to principles could contain social forces any longer – rules were replaced by principles; principles in turn are replaced by standards, policies, and other soft manifestations of authority (see the seminal

Soft authority in global governance  169 piece Unger 1976). The latter may have always been around, but in the form of soft law have come to be actively and intentionally created by regulators at their wit’s end and are taking centre stage: no longer mostly informing and colouring rules and principles, but as active regulatory instruments in their own right.

5.

ON RESOLUTIONS

The term soft law started to become widely used in connection with resolutions adopted by the UN General Assembly. This is the plenary body of the most prominent international organization in the world, having near-universal membership and near-universal jurisdiction, and yet, according to the constitution of the UN, General Assembly resolutions are devoid of binding force. They are merely, so the UN Charter has it, recommendations. They can be taken note of and cast aside, but give rise to no concrete rights or obligations, other perhaps than the good faith Charter obligation that their addressees take them seriously. Harder legal effects, so it was clear, had to come from traditional law-making processes: a resolution could form the basis of a later treaty, or it could generate a general practice culminating in customary international law. But on its own, the resolution would remain non-binding, non-law (for an early explanation, see Tammes 1958). This was considered a curious and perhaps undesirable state of affairs, especially after the wave of decolonization of the 1950s and 1960s that resulted in a large increase in UN membership and therewith all but guaranteed that the Global South could almost automatically muster a majority in the General Assembly. Scholars hailing from the Global South pointed to a marked resemblance between the General Assembly and the political institution of parliament, suggesting that it would not at all be eccentric to think of the General Assembly as a global parliament and thus think of it as a legislative body (most prominently Elias 1972). This required some elastic thinking to overcome some loose ends (the General Assembly, unlike parliaments generally, is not elected; there is no notion of proportional representation other than ‘one state, one vote’; not all parliaments have legislative powers), and the argument eventually failed to convince, but it did tap into a broader sentiment: surely, it cannot be the case that the General Assembly just produces heaps of inconsequential paper – there must be something of legal value there. If not law properly speaking, then maybe it must be something quasi-law-like – perhaps it can be referred to as ‘soft law’. One can grasp how the label appealed to describe General Assembly resolutions. After all, these do stem from the plenary body of the politically most relevant international organization, both in terms of membership and coverage of policy domains – it is not that eccentric to claim that these reflect global public opinion, all the more as some of them had early on been recognized as being of huge political and normative significance, none more so than the 1948 Universal Declaration of Human Rights. But what could possibly apply to General Assembly resolutions (or at least some of them), need not apply to the normative output of other international organizations. Resolutions adopted by UNESCO, say, or the World Meteorological Organization, may well be of relevance to their respective policy domains, but will reverberate less loudly throughout the rest of the world. But since General Assembly resolutions were deemed to constitute soft law, the same label just had to be attached to the output of other international organizations. It has, after all, proven impossible to meaningfully distinguish between international organizations

170  Research handbook on soft law (Klabbers 2013), ever since Sayre’s overly broad conceptualization came to inform legal doctrine and legal theory. For Sayre, pretty much all forms of cooperation between states constituted an international organization; the only exception, he felt, was colonial possession (Sayre 1919). All forms of cooperation are international organizations; all organizations are alike; thus it follows that their normative output should be treated in the same way. Net result: what could have been a promising approach to General Assembly resolutions came to be applied to policy suggestions emanating from the Universal Postal Union as well; what could have had a meaningful role to play in explaining and justifying the pull of General Assembly resolutions got lost in the mushiness of all kinds of recommendations, policy papers, codes of conduct, and guidelines, emanating from all sorts of entities. This leaves us with the curious conclusion that it is clear, to all observers, that there is a political phenomenon that is sometimes referred to as soft law. International organizations adopt, promulgate, and produce all sorts of instruments, and try to influence the behaviour of their member states (and social actors within those member states) by all possible means. In other, perhaps somewhat swollen, words, few observers in their right mind would question the ontology of soft law. This is, after all, what happens in the real world, and it can readily be observed to be happening in the real world as well, without adopting a fancy theoretical framework or donning particular methodological spectacles. A cursory glance at the websites of most international organizations suffices: many, perhaps most, make use of guidelines, recommendations, and codes of conduct. Many, perhaps most, also resort to practices such as the production of handbooks and manuals. These set standards through processes of socialization: a domestic migration officer on Lesbos is unlikely to have any human rights treaties on her desk but will be far more likely to have within close reach the various handbooks and manuals produced by the International Organization for Migration (IOM). The IOM, after all, has experience in solving practical problems; it has over the years built up the sort of expertise that may provide practical guidance to harried and challenged immigration officers. And many, perhaps most, resort to the newest generation of instruments: not so much setting abstract standards in the form of recommendations, but rather setting concrete goals. The World Bank has become renowned for its use of indicators, a tool also favoured by others, and the OECD’s PISA program, assessing school performance in a number of states, does not even do that: it merely produces hugely influential rankings (von Bogdandy and Goldmann 2008). Best known of all, the UN has resorted to the same technique when promulgating, first, the Millennium Development Goals, followed by the Sustainable Development Goals (see Twining 2009). But if the ontology of soft law is clear, its epistemology is hopelessly fuzzy, for reasons discussed above: it may be called ‘law’ but is supposedly devoid of the normative pull and push of law, and usually does not meet legal standards concerning the valid creation of law. The term soft law does not explain anything, and merely creates the illusion of law without there actually being any law in the picture – hypocrisy with a capital H. It may have been a useful hypocrisy for some purposes, but even useful hypocrisy is still hypocrisy.

6.

ON AUTHORITY

It might be much preferable then to drop the label soft law and replace it by something less hypocritical. One term that comes to mind as a viable alternative is the term epistemic

Soft authority in global governance  171 authority, and it is useful for two broad reasons. One is that the noun, ‘authority’, has different connotations from the term law. Authority is less necessarily associated with legal thought and legal validity – influencers and even mafia bosses can exercise authority, without their injunctions being seen as law. Related, there is the advantage that while legal rules typically require a legal basis (if things are done by the book, that is), there are less strict demands on exercises of authority: it usually does not take a specific and concrete legal basis to produce a manual or collect and promulgate ‘best practices’. Second, the adjective ‘epistemic’ suggests appropriate links to knowledge and expertise. The attraction of international organizations has always been their ostensibly a-political character: with the limited exceptions of the League of Nations and the UN (usually classified as ‘political’ organizations), international organizations have traditionally been seen as a-political, as technical, merely engaged in harmonizing things, in sensible governance. It follows, that they can claim and utilize a certain technical expertise, which in turn helps justify exercises of authority. Epistemic authority can accordingly be divided in two related manners: active and passive (this builds on Klabbers 2019a). Passive epistemic authority, on this line of thinking, refers to the circumstance that the organization builds up expertise; insofar as this is the case, it is on a par with suggesting that Harvard University is a repository of expert knowledge. Active epistemic authority, however, is when the organization also tries to utilize this by setting standards. Come to think of it, perhaps a better way of putting this is to distinguish between internal and external epistemic authority, with ‘external’ being the part that aims to influence the outside world – although admittedly, as the analogy with Harvard University suggests, the boundaries will be porous. Even a cursory glance at some websites of international organizations suggests that epistemic authority, even of the active, external kind, is commonplace. A random list of examples might include the World Health Organization’s (WHO) Guidelines on Hepatitis B and C Testing (WHO 2017), the World Intellectual Property Organization’s (WIPO) Global Innovation Index (WIPO 2021), the Food and Agricultural Organization’s (FAO) FAO Framework for the Urban Food Agenda (FAO 2019), and the Postal Reform Guide published by the Universal Postal Union (UPU 2019). These documents all have in common that they contain strong suggestions on how to approach particular issues, either by suggesting concrete action or, in the case of the Global Innovation Index, suggesting which sort of activities may be rewarded. All this builds on, and manifests, the organization’s expertise. Put differently, guidelines on hepatitis testing emanating from the World Bank, or Manchester United FC, would not nearly be as authoritative as those from the WHO, not even if substantively they would be identical, and part of the reason for this is that the WHO can claim expertise in the medical field. By the same token, they would be less authoritative if we would suspect that the WHO is an overly politicized place, if the Guidelines would be devised by government bureaucrats with no specific expertise in the medical field (for this reason few connect UN General Assembly with expert governance, see e.g., Kennedy 2016). Likewise, we trust that the FAO knows what it is doing when it comes to food, that WIPO has some inkling about innovation, and that UPU knows how best to organize postal services. As a result, there is not even a need for these organizations to resort to formal legally binding instruments: regardless of whether a treaty on hepatitis testing would be appropriate, surely the treaty form would, substantively, not add much: test procedures for hepatitis are not improved if a legal form is used. Instead, what matters here is the substantive expertise.

172  Research handbook on soft law To be sure, this does not mean that the WHO Guidelines are completely innocent. Technical knowledge, as Jasanoff (1986) recognized several decades ago, may well reflect particular national cultures, traditions, or preferences, and Merry (2016) and others8 have drawn attention to the intense negotiation processes that precede the promulgation of standards and indicators, their ostensible neutrality notwithstanding. Moreover, it is (or should be) a truism that most decisions, those of international organizations included, tend to come with winners and losers (Klabbers (2022)): some medical companies may be better prepared to live up to standards contained in WHO Guidelines than others, and thus benefit from the approach promulgated by the WHO. But even so, the point remains that these Guidelines derive authority from substantive expertise, and in turn reflect and exercise authority based on substantive expertise. Much the same applies across the board as far as international organizations are concerned – few express law-making powers, but nonetheless great authority. This even leads to curious disclaimers. The International Civil Aviation Organization (ICAO) must have been subjected to criticisms by air passengers or others about its activities or abstentions: its website proclaims that ICAO is ‘not a global regulator’, further explaining that ICAO ‘cannot arbitrarily close or restrict a country’s airspace, shut down routes, or condemn airports or airlines for poor safety performance or customer service’ (ICAO 2022a). It is no doubt accurate to say that no powers should be exercised by ICAO arbitrarily; it is equally accurate, no doubt, that ICAO has few explicit regulatory powers; but beyond this, ICAO’s statement strikes as a little too defensive, as Shakespeare’s lady who ‘doth protest too much, methinks’. ICAO displays the same attitude elsewhere. For instance, one of its main policy tools is what it refers to as ‘performance-based navigation’, aiming to facilitate navigation and render flight operations more flexible, therewith reducing emissions and noise levels, enhancing safety, and improving financial returns (ICAO 2022b). This comes with a set of standards and comes with educational offerings, an app, and other paraphernalia, including a certificate that an organization is found in compliance with the conditions and criteria attached to performance-based navigation. This certificate however, so ICAO proclaims, ‘is simply a statement of support. It does not constitute an authorization, an approval or a certification…’ (ICAO 2022c). Again, this is somewhat beside the point: the statement of support suggests ICAO’s stamp of approval, manifesting an authority claim. ICAO is no doubt correct in stating that the power to grant authorization or approval rests with domestic authorities, but clearly, there is a strong feeling on the part of ICAO that performance-based navigation is the way forward and should be stimulated: it is not for nothing that ICAO publishes a list of entities that meet its criteria (ICAO 2022c). Much the same can be said about a multitude of other international organizations. The International Organization for Migration offers its expertise, in part, through handbooks and manuals, and in part through such activities as airport design and visa processing (Klabbers 2019b). The International Telecommunication Union (ITU), likewise, produces all sorts of handbooks, indicators and datasets (see e.g., ITU 2021), and perhaps even more than other organizations, the ITU forms an epistemic community comprising external actors: companies and scientific organizations can acquire – for a fee – a particular membership status, comparable to airline loyalty cards, which gives them access to the organization, some of its decision-making processes, and large-scale international business events in the field of telecommunication (Klabbers 2021b).

Soft authority in global governance  173

7.

CONCLUDING REMARKS

Soft law has pre-occupied the minds of international lawyers for half a century now, without it having become clear what the term stands for, which kinds of instruments it refers to, or the envisaged effects of those instruments. It is always the case that this kind of confusion and uncertainty benefits some, at the expense of others, and perhaps it is possible to argue that we are all both winners and losers, at different points in time and at different levels of profitability. As noted, soft law allows for smooth governance, but without putting too much stock on such staples of liberal political thought as transparency, democratic control, or judicial review. As his main biographer once said of Keynes, he ‘often gave the impression that, provided the state apparatus was equipped with the right theories and run by benevolent Old Etonians, it could be entrusted’ with considerable power over policy (Skidelsky 2010, p. 139). In these circumstances, it is not too far-fetched to suppose that the advent of soft law, useful as it is to regulators, has come to be responsible for precisely the sort of elite governance that has helped create a populist backlash, bringing a particular breed of politicians to power.9 Where there is no control over what our political leaders and our experts do when they assemble in the corridors of the United Nations, the World Health Organization, or the World Trade Organization, it should not come as a surprise that people get suspicious and anxious. Surely there is a salutary lesson there somewhere.

NOTES 1. This is impossible to verify or falsify: it is my impression that many colleagues are a little less inclined to speak of soft law, and a little less inclined to sing its praises, than they were a decade or perhaps even half a decade ago. 2. These earlier writings include the following articles: Klabbers 1996, 1998, 2007. 3. For useful sociological exploration, see Alasuutari and Qadir 2019. 4. A classic, if perhaps misguided, example is the 1941 Atlantic Charter, concluded between US president Roosevelt and UK prime minister Winston Churchill. This is often considered a good example of soft law, which ignores the circumstance that the UK actually registered it as a treaty, and that it contains few, if any, obligations for the US and UK: it is mostly a call on other states to abide by certain principles, and thus not a good example of soft law. 5. Note that this is only to some extent mirrored in the distinction between Ideal Theory and Non-Ideal Theory, leaving International Relations scholars without much of a weapon to fend off charges of Idealism. 6. The latter conception, that international organizations are in fact meeting places of states and their actions determined by state interests, is again another Realist conceit with a strong hold on the disciplines of both international law and International Relations. 7. Anecdotal: upon first engaging in discussions on soft law with Finnish domestic lawyers, I had difficulties understanding them (and vice versa, no doubt) until I realized that what they referred to as soft law, would constitute something like ‘general principles’ in my vocabulary, characterized by open-endedness perhaps rather than doubtful pedigree or dubious bindingness. 8. See also: Davis et al. 2012; Littoz-Monnet 2017; Sending 2017. 9. It may be observed that the soft law of global financial regulation (and most of it is soft law) has neither prevented financial crises nor impoverished the investment banking community. See Brummer 2011.

174  Research handbook on soft law

REFERENCES Alasuutari, P. and Qadir, A., (2019). Epistemic governance: social change in the modern world. Cham: Palgrave Macmillan. Austin, J., (1832). The province of jurisprudence determined. London: J. Murray. Reprint, W. E. Rumble, ed. Cambridge: Cambridge University Press, 1995. Becker, C. L., (1955). What are historical facts?. Western Political Quarterly. 8(3), 327–340. Bogdandy, A. von and Goldmann, M., (2008). The exercise of public authority through national policy assessment: the OECD's PISA policy as a paradigm for a new international standard instrument. International Organizations Law Review. 5(2), 241–298. Brummer, C., (2011). Soft law and the global financial system: rule making in the 21st century. 2nd ed. Cambridge: Cambridge University Press. Büthe, T. and Mattli, W., (2011). The new global rulers: the privatization of regulation in the world economy. Princeton, N.J.: Princeton University Press. Chinkin, C. M., (1989). The challenge of soft law: development and change in international law. International & Comparative Law Quarterly. 38(4), 850–866. Davenport-Hines, R. P. T., (2015). Universal man: the seven lives of John Maynard Keynes. London: William Collins. Davis, K. E., Fisher, A., Kingsbury, B. and Merry, S. E., eds., (2012). Governance by indicators: global power through quantification and rankings. Oxford: Oxford University Press. Edelman, M., (1964). The symbolic uses of politics. Reprint, Urbana, I.L.: University of Illinois Press, 1985. Elias, T. O., (1972). Modern sources of international law. In: W. Friedmann, L. Henkin and O. Lissitzyn, eds. Transnational law in a changing society: essays in honor of Philip C. Jessup. New York: Columbia University Press. pp. 34–69. FAO., (2019). FAO framework for the Urban Food Agenda: leveraging sub-national and local government action to ensure sustainable food systems and improved nutrition [online]. Rome: Food and Agricultural Organization of the United Nations. [Viewed 5 February 2022]. Available from: doi: 10.4060/ca3151en. Fawcett, J. E. S., (1982). Law and power in international relations. London: Faber and Faber. Goldsmith, J. L. and Posner, E. A., (2005). The limits of international law. New York: Oxford University Press. Hollis, D. B. and Newcomer, J. J., (2009). ‘Political’ commitments and the constitution. Virginia Journal of International Law. 49(3), 507–584. Huber, M., (1928). Die soziologischen grundlagen des völkerrechts. Berlin-Grunewald: W. Rothschild. ICAO., (2022a). About ICAO [online]. ICAO. [Viewed 5 February 2022]. Available from: https://​www​ .icao​.int/​about​-icao/​Pages/​default​.aspx. ICAO., (2022b). Overview [online]. ICAO. [Viewed 5 February 2022]. Available from: https://​www​ .icao​.int/​safety/​pbn/​Pages/​Overview​.aspx. ICAO., (2022c). PBN recognition of compliance [online]. ICAO. [Viewed 5 February 2022]. Available from: https://​www​.icao​.int/​safety/​pbn/​Pages/​PBN​-Recognition​.aspx. ITU., (2021). World telecommunication/ICT indicators database 2021 [online]. 25th ed. ITU Publications. [Viewed 5 February 2022]. Available from: https://​www​.itu​.int/​hub/​publication/​d​-ind​-wtid​-ol​-2021. Jasanoff, S., (1986). Risk management and political culture: a comparative analysis of science in the policy context. New York: Russell Sage Foundation. Jütersonke, O., (2016). Realist approaches to international law. In: A. Orford and F. Hoffmann, eds. Oxford handbook on the theory of international law. Oxford: Oxford University Press. pp. 327–343. Kennedy, D. W., (1994). A new world order: yesterday, today, and tomorrow. Transnational Law & Contemporary Problems. 4(2), 329–376. Kennedy, D., (2016). A world of struggle: how power, law, and expertise shape global political economy. Princeton: Princeton University Press. Keynes, J. M., (1920). The economic consequences of the peace. London: Macmillan. Keynes, J. M., (1936). The general theory of employment, interest and money. Reprint, London: Macmillan, 1970. Klabbers, J., (1996). The redundancy of soft law. Nordic Journal of International Law. 65(2), 167–182.

Soft authority in global governance  175 Klabbers, J., (1998). The undesirability of soft law. Nordic Journal of International Law. 67(4), 381–391. Klabbers, J., (2007). Reflections on soft international law in a privatized world. Finnish Yearbook of International Law. 16, 313–328. Klabbers, J., (2013). Unity, diversity, accountability: the ambivalent concept of international organization. Melbourne Journal of International Law. 14(1), 149–170. Klabbers, J., (2014). The virtues of expertise. In: M. Ambrus, K. Arts, E. Hey and H. Raulus, eds. The role of ‘experts’ in international and European decision-making processes: advisors, decision makers or irrelevant actors?. Cambridge: Cambridge University Press. pp. 82–102. Klabbers, J., (2019a). The normative gap in international organizations law: the case of the World Health Organization. International Organizations Law Review. 16(2), 272–298. Klabbers, J., (2019b). Notes on the ideology of international organizations law: the International Organization for Migration, state-making, and the market for migration. Leiden Journal of International Law. 32(3), 383–400. Klabbers, J., (2021a). Conceptualizing expert accountability: towards virtue. In: A. Arcuri and F. Coman-Kund, eds. Technocracy and the law: accountability, governance and expertise. Abingdon: Routledge. pp. 27–43. Klabbers, J., (2021b). Reflections on the International Telecommunication Union: international organizations as epistemic structures. In: A. Bianchi and M. Hirsch, eds. International law’s invisible frames: social cognition and knowledge production in international legal processes. Oxford: Oxford University Press. pp. 200–218. Klabbers, J., (2022). Rules, institutions and decisions: taking distribution seriously. In: G. Hellmann and J. Steffek, eds. Praxis as a perspective on international politics. Bristol: University of Bristol Press, pp. 127–142. Kratochwil, F. V., (1989). Rules, norms, and decisions: on the conditions of practical and legal reasoning in international relations and domestic affairs. Cambridge: Cambridge University Press. Littoz-Monnet, A., ed., (2017). The politics of expertise in international organizations: how international bureaucracies produce and mobilize knowledge. Abingdon: Routledge. Merry, S. E., (2016). The seductions of quantification: measuring human rights, gender violence, and sex trafficking. Chicago: University of Chicago Press. Nicolson, H., (1933). Peacemaking 1919. 3rd ed. Reprint, London: Methuen, 1964. Onuf, N. G., (1989). World of our making: rules and rule in social theory and international relations. Reprint, Abingdon: Routledge, 2013. Patomäki, H., (2002). After international relations: critical realism and the (re)construction of world politics. London: Routledge. Reisman, M. W. and Willard A. R., eds., (1988). International incidents: the law that counts in world politics. Princeton: Princeton University Press. Righter, R., (1995). Utopia lost: the United Nations and world order. New York: Twentieth Century Fund Press. Sayre, F. B., (1919). Experiments in international administration. New York: Harper. Schauer, F., (1991). Playing by the rules: a philosophical examination of rule-based decision-making in law and in life. Oxford: Clarendon. Sending, O. J., (2017). The politics of expertise: competing for authority in global governance. Ann Arbor, M.I.: University of Michigan Press. Shklar, J. N., (1964). Legalism: law, morals, and political trials. Reprint, Cambridge, M.A.: Harvard University Press, 1986. Skidelsky, R., (2010). Keynes: a very short introduction. New York: Oxford University Press. Sunstein, C. R., (2014). Nudging: A very short guide. Journal of Consumer Policy. 37(4), 583–588. Tammes, A. J. P., (1958). Decisions of international organs as a source of international law. In: Collected courses of the Hague Academy of International Law, vol. 94. pp. 265–363. Twining, W., (2009). General jurisprudence: understanding law from a global perspective. Cambridge: Cambridge University Press. Unger, R. M., (1976). Law in modern society: toward a criticism of social theory. New York: Free Press. UPU., (2019). Postal reform guide [online]. Bern: Universal Postal Union. [Viewed 5 February 2022]. Available from: https://​www​.upu​.int/​UPU/​media/​upu/​files/​postalSolutions/​dev​elopmentCo​operation/​ GuideReformPostalEn​.pdf.

176  Research handbook on soft law Visscher, C. de, (1968). Theory and reality in public international law. Translated by E. Corbett. Princeton, N.J.: Princeton University Press. Wellens, K. C. and Borchardt, G. M., (1989). Soft law in European Community law. European Law Review. 14(5), 267–321. WHO., (2017). Guidelines on hepatitis B and C testing [online]. Geneva: World Health Organization. [Viewed 4 February 2022]. Available from: https://​www​.who​.int/​publications/​i/​item/​9789241549981. WIPO., (2021). Global innovation index 2021: tracking innovation through the COVID-19 crisis [online]. 14th ed. Geneva: World Intellectual Property Organization. [Viewed 4 February 2022]. Available from: https://​www​.wipo​.int/​global​_innovation​_index/​en/​2021.

12. International standards and the dilution of responsibility Ingrid Gustafsson Nordin and Kristina Tamm Hallström

1.

WHY STANDARDS MATTER

The International Organization for Standardization (ISO) is the world’s biggest standardizer. It has produced over 23 000 standards, covering aspects of society from containers to sustainable tourism. Today, it would actually be hard to imagine a non-standardized world, as people have come to take the standardization of things for granted – standards tend to ‘have a way of sinking below the level of social visibility’ (Timmermans and Epstein 2010, p. 71). We only think about them when they are not applied, for example when we need an adapter to charge the phone in a country where the power outlet does not match the plug – because they are not standardized. Another plausible reason why we do not generally reflect upon standards is that, since the end of the 1980s, many standards are only indirectly linked to the products and services that we consume. The most famous and most widely spread standard produced by ISO is the management system standard ISO 9001, first published in 1987, which has been sold to over 1.5 million organizations worldwide. This standard and the large number of management standards developed in its aftermath target managerial and organizational aspects of organizations rather than the physical products themselves. As consumers or citizens, we only see the end result: a diploma hanging on the wall, showing that the dentist is certified according to the ISO 14001 standard for environmental management, a label on the chocolate bar indicating that the cocoa is grown organically or that the workers have fair work conditions, or the CE mark on a teddy bear signaling that it is in compliance with standards for toy safety. Yet we seldom stop to think about what might be behind those labels, who might be in charge for whatever the rules stipulate or who might be responsible if something goes wrong. Although we live in a world of standards and we tend to take their ubiquitous presence as given in the modern, rationalized, Western world, the role of standards has not always been the same. In today’s globalized society, standards play a crucial role for transnational governance, often combined with market-based monitoring in terms of certification and accreditation that replace national regulatory regimes (Gustafsson 2020). The work of developing, deciding about, legitimizing and diffusing standards around the world is a highly political exercise involving complex networks of public organizations, corporations, civil society organizations and meta-organizations of various sorts (Loconto and Busch 2010; Brunsson, Gustafsson and Tamm Hallström 2018; Yates and Murphy 2019). Standards and the business structures surrounding them are the result of negotiations and power struggles and their impact has severe implications for almost all parts of society – standards matter. These are all important reasons to study standards from a multitude of perspectives, and to ask questions about who writes them, who promotes them, who follows them and why. 177

178  Research handbook on soft law There are, however, some essential characteristics of standards that make them especially important as a topic for studies of globalization and soft law. Standards are private rules that are not issued by parliaments but by private standardization organizations; unlike the case of public rules, the user of a standard is often required to pay for the standard; they are often supported and acknowledged by public organizations with reference to the public interest; and most importantly and like many other examples of soft law, standards are voluntary from the perspective of the standard setters. Standard setters cannot force others to follow their standards. Given their voluntariness, seminal work in the field of organization studies focusing on standards specifically suggests that standards have a particular way of creating grey zones when it comes to responsibility (Brunsson and Jacobsson 2000) – since they are voluntary, the responsibility for a standard becomes unclear or divided between the standard setter and the standard follower. Given these grey zones of responsibility inherited in standards, and given standards’ huge impact on society, we focus in our chapter on the question of how to understand responsibility in relation to standards. Who is responsible in the world of standards? We propose that standards can be understood from an organization theory perspective. This allows us to discuss standards and responsibility in a rather specific way. Because standards are a particular example of the concept ‘soft law’, we believe our discussions can be generalized to the field of soft law more broadly. Hence, our aim by using this particular frame of reference is to contribute to the soft law research in two ways: by putting standards and responsibility at the forefront and by merging the scholarly field of soft law with the field of organization studies. We have structured our chapter in the following way: first, we will give a brief history of standards as well as an overview of their basic characteristics. Then, we go on to describe what the global world of standards looks like today – various complex, dense systems of organizations organizing each other. We illustrate this complexity by accounting for one such system of organizations. The subsequent section presents a theoretical lens through which we can understand standards as examples of decided orders, that is, social orders that are organized in the sense of being based on decisions. The last section will combine the notion of standards as a specific kind of decided order (voluntary rules) with theories about responsibility, originated from organization theory. We will conclude with what we think might be lessons to be learned for the field of soft law studies.

2.

STANDARDS – THEIR HISTORY AND THEIR CHARACTERISTICS

In view of the history of standardization, different waves of standardization can be distinguished (Yates and Murphy 2019). The first wave (around 1900–1939) and the second (around 1945–1980) share the feature of having been built on a structure of national standard setters that over time became members of international standardization organizations. Standards were justified as a means of enabling industrial coordination and growth to the benefit of society as a whole. They were typically developed within private standard-setting organizations in which engineers gathered in the form of an elite movement in the service of the public good. Areas of standardization were for example industrial components, telecommunications, electrotechnical equipment, containers, to mention a few (Yates and Murphy 2019, p. 109). It was mainly during the second wave that the consumer, as an actor category, was established within stand-

International standards and the dilution of responsibility  179 ardization (Cochoy 2005; Yates and Murphy 2019). A third wave of standardization started in the 1980s, which changed the standardization landscape considerably. First, several global standard setters were established, which challenged the traditional ones, for example organizations setting standards for the architecture and operations of the Internet. Second, there was a breakthrough of management system standards following a Total Quality Management approach emphasizing quality in manufacturing systems and processes rather than the final products (Higgins 2005, p. 134; also Tamm Hallström 2004; Arnberg, Gustavsson and Tamm Hallström 2022). Known examples are the ISO 9001 and ISO 14001 standards, but there are also standards targeting various sustainability issues coming from both traditional standard setters such as ISO and numerous new organizations such as Fairtrade International, Social Accountability International, Forest Stewardship Council, and many more. These management standards specify systems for how to structure and manage processes in order to obtain set goals for the organization in question. The underlying assumption is that organizations with a rationally developed and documented management system will use it in their daily work and thereby generate products and services in line with the management system and relevant product standards and regulations included in it. Unlike the many standards for components and products which specify physical dimensions such as height, weight and volume in order to enable coordination and functionality, the articulated motive behind management system standards is more often to create trust in and among organizations in global markets, thus a more abstract purpose directed towards organizations (Brunsson et al. 2018). As noted, another change that came with the third wave was that markets of verification expanded rapidly. Such markets became lucrative standards-related businesses, including certification and accreditation as a ‘certification of the certifiers’, and a large number of member-organizations – for standard users, standard setters, certifiers and accreditors – with a purpose to coordinate and establish trust in these, often private, regulatory activities. Together these organizations and the standards-related businesses they engage in, construct complex networks of organizations involved globally (Brunsson et al. 2018; Yates and Murphy 2019). With these developments, the language as well as the drivers for standardization changed to be less about the mission to improve the world of the two first waves, toward more commercialism, branding and customer-orientation of a global ‘standardization business’ (Yates and Murphy 2019, p. 330).

3.

A COMPLEX WORLD OF STANDARDS

Standards are widespread today. They are used in many different areas, and oftentimes standards seem to be the ‘go to’-solution if one wants to accomplish something, both internationally and domestically. The fact that they are voluntary, combined with an aura of being expert-based advice, makes them attractive in terms of rational-voluntaristic authority (Boli 1999). However, their voluntary status also means that they are weak, in need of support; the standardizers cannot enforce the use of their rules themselves. Therefore, intricate and complex systems of monitoring of standard compliance have emerged, starting off in the 1980s. In this section, we account for one such system, the largest and most complex one worldwide, where the standard setter ISO together with thousands of auditing organizations play key roles.

180  Research handbook on soft law In 1985, the EU launched its program for the internal market, called the New Approach (see Micklitz in this volume). The ‘new’ in the New Approach was the idea that EU harmonization legislation would be complemented by so-called ‘harmonized standards’. In some areas, this meant that national legislation would be substituted, or rather complemented, by private law (standards) providing ‘technical specifications’ of the more general requirements of EU Directives and national legislation. One of the main arguments was that the differences in national legislation slowed down the circulation of goods on the internal market: the same product could be subject to several diverse national legislations regarding production and testing, which inevitably hampered the trade among EU countries (Gustafsson 2020; Arnberg et al. 2022). According to the new regulatory regime, safety and environmental requirements were to be stated in overarching EU directives which then were complemented by standards that would be the same for everyone across member states. This way, standards were used as a way to lower technical barriers to trade. The ideas for the management system standard ISO 9001 were incorporated into the standards making up the New Approach. This meant that from the start, even though the idea was to set rules for various products, the standards used were targeting organizations and organizational routines. The main idea was that the producing organization was not expected to show that it monitored its products, but to show that it organized for ‘self-monitoring’ of its production routines, sometimes referred to as ‘self-declaration’. A ‘self-declaration’ is a statement of conformity provided by the manufacturer of a particular product, ensuring that the requirements issued in a directive, are lived up to. Only the manufacturer can set up such a declaration, not those firms importing or distributing the particular product. In many cases, the three roles of manufacturer, importer and distributer are distributed among different organizations, hence not held by the same organization. Another important aspect of the New Approach is the so-called presumption of conformity, meaning that if an organization followed a harmonized standard to which a product directive for, say, toys, referred, it was presumed that the organization lived up to the requirements in the directive as well. This way, the EU made standards an essential element in its way of regulating the internal market. Rules are easy to decouple. It is easy for organizations to state to their environment that a rule has been implemented while in practice, no real changes are done. This is indeed true for standards, too. Certification is a specific kind of monitoring that is tailored for standard compliance and exists in different sectors with slightly different meanings. In the New Approach, certification means that an organization can state that it conforms with specific directives. It is typically voluntary, but for certain product categories it may be mandatory for organizations to have their standard compliance monitored by an external certification firm. Certification firms are for profit, receiving their income through sales of certification services to the organizations to be certified, and work in all sectors of society. When the EU presented its New Approach, it was explicitly made clear that certification was to be conducted by private firms – not public authorities – which led to a rapid expansion of certification markets around the world (Galland 2017; Gustafsson 2020; Arnberg et al. 2022). Many firms in these markets are multinational such as Bureau Veritas, Intertek and Lloyd’s Register. Certification firms are tasked with controlling whether the organization in question – say, a toy producer – has their documentation in order, that is, if they have documented their routines and structures according to a standard (the self-monitoring mentioned earlier). The majority of certification audits today are ‘desk audits’, meaning that they monitor documentation of an organization’s management system

International standards and the dilution of responsibility  181 and not the actual physical products through a testing procedure in a laboratory, for example. These audits are based on the assumption that is core to management system standards such as the ISO 9001: that a documented management system specifying organizational routines and structures developed in accordance with mandatory requirements is a valid mechanism for quality assurance of the products and services produced. The work of certification firms is guided by a standard too. For most certifiers, the ISO 17021 is the dominant standard that specifies how firms in the certification business working with management system certification, should organize and manage their business activities of auditing the standard compliance of another organization. Put simply, certifiers are assumed to follow a standard for how they work when they audit that other organizations follow standards. Because certification organizations were not public authorities or non-profit organizations but profit-making firms, the EU needed to make sure that certification firms were not failing to fulfil their purpose to serve the public interest, for example by prioritizing economic gain at the expense of thorough, independent auditing. For this reason, accreditation was introduced. Accreditation is a kind of monitoring aimed at other forms of monitoring organizations, such as certification organizations, laboratories and calibration firms, for example. Accreditation is best described as a certification of certifiers. Within the EU, it is sometimes conducted by public authorities. For example in Sweden, the Swedish accreditation organization Swedac is a public authority that is mainly funded by sales of accreditation services and which answers to the Ministry of Foreign Affairs. In other countries, a private organization acts as the national accreditor, such as UKAS in the case of the United Kingdom. In 2008, the EU launched the Goods Package that included directives that specified and developed the New Approach. The Goods Package paid special attention to accreditation through Regulation (EC) 765/2008, justified as a way to strengthen the trust in the New Approach: Through the establishment of a new, European accreditation infrastructure, trust for the entire chain of control would be installed. Specific ideas about how to organize accreditation were also presented. Accreditation organizations were not allowed to compete with one another and there could only be one per member country. Those countries that did not have an accreditation organization had to create one and those countries who had many of them, like Germany, the United Kingdom and the Netherlands, had to merge them into one National Accreditation Body. Also, the accreditation organizations could not be for-profit. In many ways, the ideas about accreditation were described as an inverted version of the ideas surrounding certification. But just like certification firms, accreditation organizations are guided by a standard in their work, the ISO 17011 for accreditors. This standard specifies how accreditors should work when conducting auditing of certification firms. Their mission is to make sure that certifiers follow a standard such as the ISO 17021 correctly, that is, that they have proper routines, documented in a management system, for how to certify that other organizations have proper routines for how to follow other standards correctly. Just like the ISO 17021 and ISO 9001 standards, the ISO 17011 is mainly directed towards organizational routines and emphasizes values such as ‘impartiality’ and ‘independence’. Through Regulation (EC) 765/2008, all national accreditors within the EU need to be members of the meta-organization European Accreditation (EA). The EA has a formal mandate to coordinate and monitor its members. The auditing of accreditors is done on a peer basis, meaning that the members monitor each other. The purpose of the monitoring is to ensure that the ISO 17011 standard is implemented correctly. There are regional versions of

182  Research handbook on soft law the EA around the world – in Oceania, Northern Africa, Asia and North America for example – all having their member organizations for accreditors, where the members monitor each other. On a global level, the equivalent is called the International Accreditation Forum (IAF) assembling all the regional meta-organizations as well as their individual member organizations. The IAF is constituted by its members which are organizations themselves. As common for many other meta-organizations, the IAF is a weak organization in terms of employees and resources (Ahrne and Brunsson 2008) – a rather ‘empty’ organization in terms of employees and physical office, although it has some administrative staff. The purpose of the IAF is to coordinate accreditation organizations globally and to promote the value of standards, certification and accreditation to the world. The coordination is done through the system of Multi-Lateral Agreements (MLA), where the member organizations agree that they will treat all accredited certifiers around the world in the same way as if they had done the accreditation themselves. The idea being that accredited certificates under the MLA should be treated as equal as they are conducted in accordance with an international standard. In brochures and policy documents issued by the EA and the IAF, the MLA is sometimes called ‘the passport to trade’.1 Built on standards, certifications and accreditations, the IAF enables a global system of organizations organizing each other. It is a system based on standards where all organizations are standardized, and where the majority of organizations somehow audit the standard compliance of another organization. To sum up, starting off with the launch of the New Approach in 1985, and with the parallel proliferated use of management systems standards such as the ISO 9001, the system of standards, certification and accreditation we have briefly described above, incorporates many of the greatest societal trends and reforms that we have seen during the last four decades. The system represents a shift from ‘bureaucracy’ and hard law to voluntary, private rules where state-run monitoring systems have been replaced by market-based solutions. It also represents the overall turn from domestic and small scale markets to global and transnational systems of trade. Given how the system described in this chapter is organized, it also raises several important research questions. Who is in charge of this massive, global, complex set of organizations organizing each other? There seems to exist thousands of organizations, where everyone is in charge of monitoring or regulating someone. Standards, certification, accreditation and peer reviews may in this context be seen as tools for the articulation and assurance of individual organizations’ responsibility. But no one is in charge for everyone (Gustafsson 2020). Neither ISO nor the EU or the IAF has the mandate to govern the entire system and there is no equivalent to a global government or a CEO. And if there is no one in charge, how can we understand responsibility in a global system based on voluntary rules?

4.

UNDERSTANDING STANDARDS AS DECIDED ORDER

In order to understand issues regarding responsibility on a global level, we propose using organization theory. Organization theory enables us to grasp the complexity of organization, the mechanisms that drives the escalating organization of organizations, and to understand its implications.

International standards and the dilution of responsibility  183 In traditional organization theory, formal organizations are commonly understood as systems of decisions (Luhmann 2000; also March and Simon 1958). From this perspective, decisions are key in understanding organizations because organizations are assumed to be constituted by decisions and decided structures. It also singles out organizations, constituted by decisions, as different from other types of social orders such as institutions or networks. Institutions are typically built through social behaviours that have emerged over time, and that have become institutionalized in the sense that they become taken for granted (Jepperson 1991). Networks are social orders based on reciprocity and trust (Powell 1990) and not on decisions. A ‘pure’ network is not decided, has no hierarchy and no formal membership. Organizations, on the other hand, start with a decision to create an organization, and organizations then decide on rules, membership, sanctions, hierarchy and monitoring of its members. Recent developments in organization theory going under the label of partial organization (Ahrne and Brunsson 2011, 2019) suggest that organization as decided order exists not only inside formal organizations but also outside and among them. Organizations make decisions about other organizations all the time, meaning that decision-making in organizations is not only directed towards the internal structures of the decision-making organization, but also to other, external organizations. Organizations are recipients of many external decisions, oftentimes in the form of various rules. Standards would be the prime example: ISO is a formal organization making decisions about rules, that is, the standards they issue, and these standards are not directed for use within ISO itself but for other organizations around the world in various sectors and of different sizes. The complex and globally spanning constellation of organizations that we have described above illustrates this very well: it is made up of organizations making decisions about others rather than themselves. In addition to ISO deciding about standards for others to follow, certifiers make decisions about the standard compliance of their client organizations; accreditors make decisions about the standard compliance of certifiers; member organizations such as the EA and the IAF, make decisions about standard compliance of their members, etcetera. From this perspective, it is possible to understand our global world not only as a world with increasingly ‘blurred boundaries’ or hybrid forms of governance, as is often suggested (Bache, et al. 2015; Benish and Mattei 2020). It is also possible to understand it as an increasingly organized world – more and more organizations exist to make decisions – to set rules, to decide on membership or monitoring activities – that target other organizations. In a globalized world where there is no world state with authority to enforce regulation, there is room for more organizations to make decisions about others – to come up with suggestions and recommendations that are voluntary for others to decide whether to follow or not, spanning larger areas than the jurisdiction of the nation state. This is what much of the transnational governance literature points out (Djelic and Sahlin Andersson 2006; Bexell and Mörth 2010). These rules are not legislation and many of them are at least in theory, voluntary to follow, that is, ‘soft law’. As noted, standards are prime examples of such a tool that has become attractive for the governance of global markets. But this development also means that there are proliferating markets for standards and other soft law tools, which in turn opens up for yet more markets for organizations such as certifiers, accreditors and member organizations conducting various forms of monitoring to strengthen the perceived value of standards. The adjective ‘soft’ suggests that something in soft law is less formalized and more porous in comparison to ‘hard’ law which is a legally binding obligation. From our perspective, however, typical examples of soft law, such as guidelines, recommendations or standards,

184  Research handbook on soft law share one fundamental characteristic with hard law: they are decided rules. Unlike institutional norms that are emergent patterns of behaviour that become taken for granted, soft law takes the form of specified, explicit criteria that have been decided by someone. Both soft and hard law are rules with names, they originate from somewhere and are created by someone, either a natural or a legal person. In other words, they are based on decisions. Someone decided to make the rule and decided about the specific content. Given their voluntary nature, there are also decisions behind the compliance with soft law; someone makes a decision to follow the rule. The difference between hard and soft law, very simplified, is whether people are forced to follow such rules, which then would not count as a decision, or if they can make the voluntary decision by their own to follow them, i.e., rules that contain an element of voluntariness (see Snyder, this volume; also Abbott and Snidal 2009). Viewing soft law and standards in this way has important implications for responsibility, which we will elaborate on in the next section.

5.

STANDARDS AND RESPONSIBILITY

Responsibility is one of the most debated, theorized and studied phenomenon in social sciences (Miller 2001; Bovens 2010; Brunsson, et al. 2022). There are various definitions and discussions about responsibility in law, philosophy, accounting and political science. Many times, not least in public debates and in media, issues about responsibility are discussed in normative terms, about who should be responsible in a given situation. Our take on responsibility is somewhat different. We do not seek to identify, evaluate or judge who should or should not be responsible. Our assumption is rather that responsibility is socially constructed. It becomes an empirical question of how responsibility is created in different situations and the ways in which it can be tied to physical or legal persons convincingly. With the analytical framework that we offer, we provide clarity on how such responsibility creation can be explained. Key to our understanding is the idea of choice and cause as essential for responsibility. Responsibility distribution is a combination of choice and cause. This line of thinking runs back even to Aristotle who stated that whoever caused something by free will is to be responsible for whatever caused. Choice is commonly communicated in the form of decisions (Brunsson 1990). According to traditional decision theory, decisions are even equated with choice, the idea being that whoever makes a decision has chosen between at least two alternatives. Decisions also indicate cause: one decides in order to make something happen, to cause something. Because decisions are crucial in (formal) organizations, organizations are usually depicted as something that concentrates responsibility. Organizations concentrate responsibility in two ways: (a) as entities, that is, as legal persons to which responsibility can be assigned, and (b) to the hierarchical centre, that is, to the top management of the organization with the power to make decisions for the rest of the organization and thereby also ascribed the overall responsibility of the organization (as noted by Fayol 1949/1999). Although decisions are communicated as an outcome of choice in view of a cause – thus implying a direct link between choice, cause and decision – in practice they are not necessarily connected as tightly as communicated. The real choice in a matter might be made somewhere else than where the decisions are made. For example, ‘experts’ or consultants are often brought in to help politicians or managers navigate their decision processes, providing data or sug-

International standards and the dilution of responsibility  185 gestions that in practice leave no room for choice by the formal decision-maker in the actual decision process. The formal decision-maker may still be presented and communicated as the ‘real’ decision-maker as the decision would not be perceived legitimate otherwise. Also, decisions do not always cause what many decision-makers hope they will. In organization theory, the disconnection between decisions made and the impact the decision has in real life is called decoupling (Meyer and Rowan 1977; Brunsson 1989 for ‘hypocrisy’). Viewing decisions this way enables us to problematize responsibility. If one wants to escape responsibility, one can admit that one made a decision while claiming that one really did not have a choice whether to make that decision or not. Or one might say that the decision did not cause whatever is being discussed. Or, one might say that one did not make a decision at all. It works the other way around too. Someone who wants to take credit for something and appear responsible might stage a decision-making process for something that is just routine, or might tie a decision to a cause that in reality was unrelated to the decision. Against this backdrop, the idea that formal organizations are entities to which responsibility can be allocated, needs to be reconsidered if one takes into account that decisions are being made also outside and among organizations. As noted, the number of organizations making decisions about other organizations has increased over time and is part of globalization. As a phenomenon, however, it is not new. Organizations are being told what to do through external rules all the time; all organizations in society have at least some legislation to adhere to. But in such cases, the issue of responsibility is not complicated: since following legislation is mandatory, responsibility for the rule lies with the rule setter, that is, the legislative organization. There is no choice involved in following legislation as it is mandatory. On the contrary, it would be a choice to disobey mandatory rules which then would lead to responsibility. When it comes to voluntary rules, however, things get more complicated. As we stated above, standards are one prime example. Standards are decided rules directing the work in organizations, but the decision about the content of the standard comes from an external organization, from the standard setter. Standards are also, at least in theory, to cause something; like all rules, standards are meant to change or restructure something – to coordinate, to improve quality, to make more sustainable products, to assure occupational health and safety and fair work conditions of employees, etcetera. Contrary to legislation that in principle only involves one decision about the rule which is the decision to create the rule, standards give rise to two decisions: the decision made by the standard setter about the content and publication of the standard, and the decision to follow the standard by any kind of organization that wishes to follow a standard. Although there may be agreement on the causes generated by a standard, the responsibility for such consequences is less clear, as the responsibility that usually comes with a decision, here is split in two through the two decisions made – the decision about the standard itself and the decision to follow this particular standard. The standard setter can blame the standardized: ‘We just wrote the rule and it’s voluntary to follow – it’s not up to us what the standard causes after it leaves “the nest”’. And the standard-follower may blame the standard setter: ‘We just followed the rule, whatever is in the rule is not our fault’. What we argue here is not that it will never be possible to successfully make a standard setter accountable for faulty standards, or that standard followers can never be attributed responsibility for causing damage when following a standard developed by someone else. Rather, our point is that the outcome of such argumentation is an empirical question. And when organizations are organized voluntarily from the outside, for example through standards,

186  Research handbook on soft law certifications, accreditations, rankings, etcetera, more decisions are involved and thus more candidates to blame, which in turn affects the social construction of responsibility linked to those decisions. Every part of society that is governed by voluntary rules has to deal with this kind of responsibility division. In the world of standards that we have described in this chapter, with thousands of organizations involved in organizing each other through standards combined with various forms of auditing and membership in global or regional organizations, responsibility is cut into such small pieces that it becomes extremely diluted and eventually evaporates. There are studies on how to find someone accountable (for breast implants, see Galland 2017 or Van Leeuwen 2014) but even though there is a Product Liability Directive in the EU and even though people may eventually (in some cases) claim damages (see Micklitz in this volume), it still proves our point: the way these standards systems are organized makes it really hard to find someone responsible. For pedagogical purposes, we will illustrate this with a fictional situation. Let us pretend a toy teddy bear lost an eye and nearly choked a child. Who is responsible? The organization that produced the teddy bear made a decision to follow a standard for toys. Here, the toy producer can blame the standard setter, and the standard setter can blame the toy producer (and normally several more organizations are involved in the supply chain as the manufacturer is usually not the same as the organization selling the teddy bear – reality is messier than what we illustrate here). The toy producer was moreover certified according to that standard, implying that the toy producer can blame the certifier while the certifier can blame the toy producer, or it can blame either ISO for the ISO 17021 standard that the certifier is following, or the accreditor that monitors the certifier for its compliance with this standard. Already here we are losing sight of responsibility. Moreover, the accreditor may blame ISO for the ISO 17011 standard it is following in its work of accrediting certifiers. As mentioned, the accreditors are monitored too, implying that the accreditor of the certifier of the organization that produced a teddy bear that almost choked a child, might blame the EU-mandated organization EA that performs peer evaluations of its members, the accreditation organizations. They may also ascribe the responsibility to the IAF, the international organization for all accreditors around the world. And since the work in these two international member organizations is based on the members peer evaluating each other according to standards, responsibility finally disappears. The teddy bear is lost somewhere in the system. It is lost for two reasons: partly because since the 1980s, standardizing and monitoring focused more and more towards organizational routines and less and less on physical things. The teddy bear, the actual physical toy, was never in sight, only the idea that with documented routines and safety protocols targeting the manufacturing organization, a teddy bear will not lose an eye. The responsibility for the teddy bear is lost also partly because the system of which the teddy bear is a part, involves so many decisions about other organizations, that responsibility is virtually impossible to find. Every decision about a standard or a certificate or an accreditation, is a transfer of responsibility. These two reasons are probably related. If one trusted organizations and their routines, one would not need yet another standard to audit the prior one. Maybe, if standards and the monitoring that supports them were focused on physical things and not organizational routines, then the escalating system of organization would stop. From another perspective, if one wants to organize as much as possible, yet allocate as little responsibility as possible, one should take a close look at the system we have described.

International standards and the dilution of responsibility  187

6. CONCLUSIONS International standards are voluntary, private rules that often work on a global level. In other words, standards are perfect examples of soft law, but as such they have not received attention from scholars within the soft law field. Our aim with this chapter was therefore twofold: to direct attention to international standards, but also to demonstrate and discuss the relevance of organization theory in relation to the field of soft law. Standards can, and we argue, should, be seen as tools for rationalization. The standards we have discussed target organizational routines and processes, with the assumption that if only an organization’s structures can be designed in a specific way, certain favourable outcomes will emerge. The decided order that over time has been built up around standards and complemented with monitoring such as certification and accreditation and other instances of organization, is a system where the belief in what organizations can accomplish is extraordinarily strong: the basic, underlying idea running through the entire system is that if one can standardize organizations in a specific way, the desirable outcome will arise. This assumption, however, is not one that is shared among scholars in organization theory (e.g. Meyer and Rowan 1977; Brunsson 1989; Bromley and Meyer 2021). Organization theory started gaining ground in the 1950s, though there is no common agreement on an exact date or period of time (see Starbuck 2003 and Czarniawska 2013). Although it would be very misleading to refer to the field of organization studies as one theory (organization theory), there are threads of thoughts that flow as an undercurrent in many works. The studies we have referred to are all devoted to questioning the myth and ideal of the rational organization. International standards have made globalization possible, and it is hard to imagine a global market economy or the EU functioning without them. But we want to draw attention to the underlying assumptions about rationality that these standards are based on. It seems that there is a very strong belief in organization in terms of decided order as a mechanism of coordination – not the least on global markets, which might be a bit counterintuitive. From an organization theory perspective, markets require and produce a lot of organization (Brunsson and Jutterström 2018), both formal organizations as decision-making constructs and organization as decided order within and among formal organizations. Standards can be seen as part of such an analytical framework. In other words, international standards are often presented as ‘market solutions’ but in practice, these rules produce escalating organization. We analysed this using a specific framework at the forefront in contemporary organization studies, namely partial organization. This way, we could also understand and explain the implications of using standards to an increasing extent in society: standards tend to dilute responsibility. From our perspective, the more standards there are, the more auditing and rules for auditing, there will be – organizational practices that become more and more intermingled. This means more partial organization which in turn means weaker conditions for attributing clear responsibility, although many standards set out to design ‘responsible organizations’. Standards for Corporate Social Responsibility, CSR, would be the prime example (see Arnold 2022). To conclude, more organization does not result in more responsibility, but rather the opposite. This insight is acknowledged by scholars in transnational and multi-level governance as well, but using organization theory, this phenomenon can be explained rather than just described.

188  Research handbook on soft law

ACKNOWLEDGEMENTS The authors would like to acknowledge the support of the Swedish Research Council project 2017–01284 and of Handelsbankens forskningsstiftelser project P17–0140.

NOTE 1. For the term ‘passport to trade’, see e.g., https://​european​-accreditation​.org/​mutual​-recognition/​iaf​ -ilac​-recognition/​[viewed 14 June 2023].

REFERENCES Abbott, K. W. and Snidal, D., (2009). The governance triangle: regulatory standards institutions and the shadow of the state. In: W. Mattli and N. Woods, eds. The Politics of Global Regulation. Princeton: Princeton University Press. Ahrne, G. and Brunsson, N., (2008). Meta-organizations. Cheltenham: Edward Elgar. Ahrne, G. and Brunsson, N., (2011). Organization outside organizations: the significance of partial organization. Organization. 18(1), 83–104. Ahrne, G. and Brunsson, N., (2019). Organization outside organizations. The abundance of partial organization in social life. Cambridge: Cambridge University Press. Arnberg, K., Gustavsson, M. and Tamm Hallström, K., (2022). Neoliberalized business-consumer relations in the Swedish certification market, 1988–2018. Enterprise & Society. 23(2), 1–29. Available from: doi:10.1017/eso.2022.3. Arnold, N., (2022). Accountability in transnational governance: The partial organization of voluntary sustainability standards in long-term account-giving. Regulation and Governance. 16(2), 1–17. Bache, I., Bartle, I., Flinders, M. and Marsden, G., (2015). Blame games and climate change: Accountability, multi-level governance and carbon management. The British Journal of Politics and International Relations. 17(1), 64–88. Benish, A. and Mattei, P., (2020). Accountability and hybridity in welfare governance. Public Administration. 98(2), 281–290. Bexell, M. and Mörth, U., eds., (2010). Democracy and public-private partnerships in global governance. London: Palgrave Macmillan. Boli, J., (1999). Conclusion: World authority structures and legitimations. In: J. Boli and G. Thomas, eds. Constructing world culture – International non-governmental organizations since 1875. Stanford: Stanford University Press, pp. 267–300. Bovens, M., (2010). Two concepts of accountability: Accountability as a virtue and as a mechanism. West European Politics. 33(5), 946–967. Bromley, P. and Meyer, J. W., (2021). Hyper-management: Neoliberal expansions of purpose and leadership. Organization Theory. 2(3), 1–20. Brunsson, N., (1989). The Organization of Hypocrisy: Talk, Decisions and Actions in Organizations. Chichester: Wiley. Brunsson, N., (1990). Deciding for responsibility and legitimation: Alternative interpretations of organizational decision-making. Accounting Organization and Society. 15(l–2), 47–59. Brunsson, N., Gustafsson, I. and Tamm Hallström, K., (2018). Markets, trust and complex organizations. In: N. Brunsson and M. Jutterström, eds. Organizing and Re-organizing Markets. Oxford: Oxford University Press. pp. 136–152. Brunsson, N., Gustafsson, I, and Tamm Hallström, K., (2022). Un-responsible organization. How more organization produces less responsibility. Organization Theory. 3(4). Brunsson, N. and Jacobsson, B., (2000). A world of standards. Oxford: Oxford University Press. Brunsson, N. and Jutterström, M., eds., (2018). Organizing and Re-organizing Markets. Oxford: Oxford University Press.

International standards and the dilution of responsibility  189 Cochoy, F., (2005). A brief history of ‘customer’, or the gradual standardization of markets and organizations. Sociologie du travail (English supplement). 47(1), 36–56. Czarniawska, B., (2013). Organizations as obstacles to organizing. In: D. Robichaud and F. Cooren, eds. Organization and organizing. Materiality, agency, and discourse. New York: Routledge. pp. 3–22. Djelic, M.-L. and Sahlin-Andersson, K., eds., (2006). Transnational governance: Institutional dynamics of regulation. Cambridge: Cambridge University Press. Fayol, H., (1949) (1918; 1999). Administration industrielle et générale. Paris: Dunod. Galland, J.-P., (2017). Big third-party certifiers and the construction of transnational regulation. ANNALS of the American Academy of Political and Social Science. 670(1), 263–279. Gustafsson, I., (2020). How Standards Rule the World: The Construction of a Global Control Regime. Cheltenham: Edward Elgar. Higgins, W., (2005). Engine of Change. Standards Australia Since 1922. Blackheath: Brandl & Schlesinger. Jepperson, R., (1991). Institutions, institutional effects and institutionalism, In: W. W. Powell and P. J. DiMaggio, eds. The New institutionalism in organizational analysis, Chicago: University of Chicago Press. pp. 143–163. Loconto, A. and Busch, L., (2010). Standards, Techno-economic Networks and Playing Fields: Performing the Global Market Economy. Review of International Political Economy. 17(3), 507–536. Luhmann, N., (2000). The paradox of decision making. In: D. Seidl, and K. H. Becker, eds. Niklas Luhmann and organization studies, Copenhagen: Copenhagen Business School Press. March, J.G. and Simon, H.A., (1958). Organizations. New York: Wiley. Meyer, J. and Rowan, B., (1977). Institutionalized organizations: formal structure as myth and ceremony. American Journal of Sociology. 83(2), 340–363. Miller, D., (2001). Distributing responsibilities. Journal of Political Philosophy. 9(4), 453–471. Powell, W.W., (1990). Neither market nor hierarchy: network forms of organization. Research in organizational behavior. 12, 295–336. Regulation (EC) No 765/2008 of the European Parliament and of the Council of 9 July 2008 setting out the requirements for accreditation and market surveillance relating to the marketing of products and repealing Regulation (EEC) No 339/93 (2008). Official Journal. L218, 13 August 2008, 30–47. Starbuck, W. H., (2003). The origins of organization theory. In: H. Tsoukas, and C. Knudsen, eds. The Oxford handbook of organization theory. meta-theoretical perspectives. Oxford: Oxford University Press. Tamm Hallström, K., (2004). Organizing international standardization – ISO and the IASC in quest of authority. Cheltenham: Edward Elgar. Timmermans, S. and Epstein, S., (2010). A world of standards but not a standard world: toward a sociology of standards and standardization. Annual Review of Sociology. 36, 69–89. Van Leeuwen, B., (2014). PIP breast implants, the EU’s New Approach for goods and market surveillance by notified bodies. European Journal of Risk Regulation. 5(3), 338–350. Yates, J.-A. and Murphy, C. N., (2019). Engineering rules: global standard settings since 1880. Baltimore: Johns Hopkins University Press.

PART III ACTORS, INSTITUTIONS AND MAKING OF SOFT LAW

13. Soft law and courts: saviours or saboteurs of the rule of (soft) law? Mariolina Eliantonio and Emilia Korkea-aho

1. INTRODUCTION The research on soft law, understood here as norms that are not binding but which have practical and legal effects, has been abundant in the last two decades. The scholarly consensus on soft law as a tool of governance is that soft law has many important advantages, in that it can be quickly adopted and amended, it promotes flexibility, and accommodates variation in implementation and enforcement taking diverse regulatory preferences and local contexts into account (Ştefan 2012; Korkea-aho 2015b). Often in almost the same breath, the literature adds that soft law also has numerous disadvantages, because the ways in which soft law is adopted are not always legitimate or transparent and because the use of soft law by the executive arm might push legislatures out of the way and disregard citizens’ concerns (Ştefan 2014; Senden 2004; Senden 2013; Korkea-aho 2015a). Against this background, it is then hardly surprising that the role of courts vis-à-vis soft law has equally led to a range of conflicting views. In the EU policy-making context, the oft-repeated view is that soft law weakens the role of courts by regulating in a manner that evades judicial control. For instance, as a policy instrument, the Open Method of Coordination (OMC) is clearly outside judicial control, with Hervey (2010) and Smismans (2010) arguing that while the ideational and organizational components of the strategy did have an effect on regulatory output, they had a very modest impact on the case law (on OMC, see van Gerven and Stiller in this volume). Many have even warned that courts should not engage with soft law, as this might be interpreted as tacit acceptance of illegitimate ways of law-making and open the way for ‘backdoor legislation’ (Ganz 1987 cited in Ştefan et al. 2019). International law scholars have argued that, because there is a certain degree of confusion about the effects of soft law, judiciaries should keep their distance from such measures (Chinkin 1989, pp. 862–865; Klabbers 1998). Equally forcefully, others have noted that rather than erecting a firewall between courts and soft law instruments, effective judicial protection instead requires that courts engage with soft law instruments (Eliantonio 2018). In the same vein, it is argued that the interpretation of the law is not possible if the law is understood only as comprising rules adopted by legislatures (Scott 2011; Korkea-aho 2014). Why are there so many different views on the relationship between courts and soft law and why do many of them sceptically view courts’ engagement with soft law? The first reason has to do with the fact that despite the best intellectual efforts of scholarship, there is no conceptual clarity about the effects of soft law, which muddies the water for efforts to outline the role of courts. The second reason relates to the absence of clear procedural requirements, which makes it hard for the courts to engage with, and establish clear legal parameters for, the control of soft law instruments. Third, we posit that engaging with soft law, the courts are put in a position to think about how policy-making process works, which is something that lawyers 191

192  Research handbook on soft law (and judges in particular) are not used to doing. Fourth, a variety of views is a consequence of a multitude of expectations which scholars and policy-makers thrust on courts. But while there may be confusion about the role of courts, the courts do engage with soft law instruments, and do so frequently. The analysis of case law at national and EU levels shows that courts are aware that by engaging with soft law they subject themselves to criticism from all possible corners. When courts consider soft law for interpretative purposes or review the exercise of administrative discretion, they can be either regarded as ‘taming’ soft law’s experimentalist nature, by grounding it into rule of law principles or seen as jeopardizing the same rule of law values, by endowing soft law with legal and practical effects it is not meant to have. Similarly, when admitting claims against soft law measures, courts may be perceived as unduly enlarging the scope of the applicable legislative framework, but when they reject the admissibility of such claims, they are accused of being blinded by the strict binary view of the hard/soft law divide. It is almost impossible for the courts to get it right. For this reason, they have proceeded with caution, yet we note that in recent years they have relaxed the (admissibility) criteria for challenges against soft law measures to ensure effective judicial protection and increasingly engaged with soft law measures in cases brought before them. Our aim in this chapter is to ‘deconstruct’ what we call the complex pull between views for ‘more’ and ‘less’ courts at the national and EU level, by introducing the arguments in favour of, and against, courts and analyzing them in the normative environment in which the courts both use and control soft law. By normative environment we mean that we limit our analysis to case law and legal rules, while leaving aside political, cultural or social reasons that may explain the courts’ engagement (or the lack of it) with soft law. While we consider some national case law, the focus will be on the Court of Justice of the European Union (CJEU). In this way, we do not mean to undermine the importance of, and the need for, comparative research on the way in which national courts engage with national and EU soft law (see, in this respect, Eliantonio et al. 2021 and the working papers of the SoLaR project).1 We also do not focus on international courts and how they deal with soft law (see, however, briefly Urueña and Tamayo-Álvarez in this volume), a topic which would also merit further research as well as comparison with courts within the EU as well as with courts in other national systems. We argue in this chapter that the courts’ cautionary approach, which has probably been due to a survival instinct, might result in ‘the best of both worlds’, i.e., something which allows courts to consider soft law as an integral part of the normative framework while, at the same time, controlling its potentially erratic energy and properties against the rule of law parameters. The first step is to be clear about the various arguments that have been brought forward for and against the courts’ engagement with soft law. The chapter proceeds as follows. We first discuss what constitutes the rule of (soft) law by identifying expectations that arise from the rule of law for soft law. We then begin ‘deconstructing’ the pull between views for ‘more’ and ‘less’ courts by first introducing the arguments in favour of courts. We divide these arguments and their proponents into two camps: those who prefer more engagement by courts because they believe that courts are necessary to control soft law to make sure that it respects higher constitutional parameters, and those who think that courts cannot avoid considering soft law, because soft law is such an integral part of legislation and legislative frameworks. To say that courts should disregard soft law would be akin to saying that courts should read only every second article in a directive! Then we move on to those arguments in favour of less engagement with soft law by courts. The first group

Soft law and courts  193 of scholars think that courts’ involvement in the application or control of soft law unhelpfully blurs the distinction between hard law and soft law, creating all sorts of problems with regard to law’s authority and legitimacy. The second group of arguments is based on the idea that it is not courts’ role to fix whatever problems there are with soft law, and that other mechanisms of control are better placed to repair any legal violations to which soft law might give rise. These four groups or sets of arguments are not clearly separated, and many scholars can be identified as belonging to more than one group, us as authors included. In our overview of different arguments, we have attempted to be as fair and inclusive as possible, also noting that we as authors belong to different groups and have emphasized different views in our own research. We, however, believe that distinguishing these four broad arguments is helpful in organizing the discussion around courts and soft law and envisaging what the proper role of courts should be.

2.

THE RULE OF (SOFT) LAW: WHAT DOES IT ENTAIL?

Lon L. Fuller could be described as an early rule of law scholar. In his 1964 book The Morality of Law, he articulated the principles of what he called ‘the inner morality of law’ – principles requiring that laws be general, public, prospective, coherent, clear, stable, and practicable – and argued that these were indispensable to law-making. Mark Dawson, drawing on the Fullerian conception, described EU law-making that is committed to respecting the principle of legality as ‘one that sees the function of law as the provision of stable normative expectations’, where legal rules are ‘prospective and stable, they apply equally to all and they are to be formulated independently of the circumstances in which they are applied’ (Dawson 2009, pp. 2–3). It is easy to see why and how the development of soft law, if one adheres to this conception of law-making, may pose a threat to the principle of legality and more generally to rule of law principles. As a concept, the rule of law is one of the foundational values to which both the EU and the Member States must adhere under Article 2 of the Treaty on the European Union (TEU). The Council of Europe’s Venice Commission has positioned itself as a contemporary rule of law reformer and identified the core dimensions of the rule of law, which are used as general benchmarks in European and EU discussions on the rule of law (Venice Commission 2016). The first dimension is legal certainty, which involves the accessibility of the law. The law must be certain, foreseeable and easy to understand.2 This relates to soft law in that there are no clear publication requirements for soft law, and the research has shown that translation and publication patterns for soft law vary considerably between different instruments, if they exist at all (Eliantonio 2018). In this context, the Venice Commission (2016) also reminds that basic principles such as nullum crimen sine lege, or the prohibition of retroactivity in criminal law, are bulwarks of the legal certainty. It is, therefore, easy to see why there is little soft law in the area of criminal law. The second dimension of the rule of law is the prevention of abuse/misuse of powers. This means having safeguards against arbitrariness, ‘providing that the discretionary power of the officials is not unlimited, and it is regulated by law’.3 In the context of soft law, this would seem to imply that powers to adopt soft law are not unlimited, but should – at a minimum – require a legal basis in primary or secondary EU law, as well as the respect for hierarchically superior rules of primary and secondary EU law. This would also call for courts as controllers of possible abuses of administrative discretion associated with the use of law.

194  Research handbook on soft law The third dimension is equality before the law and non-discrimination. It is paramount that law guarantees the absence of any discrimination on grounds such as race, colour, sex, language, religion, political opinion, national or social origin, birth and so on. Also important is that similar situations must be treated similarly and different situations differently. Positive measures could be allowed as long as they are proportionate and necessary (Venice Commission 2016). Considered in the context of soft law making, this would imply that soft law can only be used to the extent that it respects or serves to protect the prohibition of discrimination. Finally, the fourth dimension is access to justice, which implies the presence of an independent and impartial judiciary and the right to a fair trial. The independence and the impartiality of the judiciary are central to the public perception of justice (Venice Commission 2016, p. 20). In the context of soft law, this requirement points to the need to establish a judicial forum to review possible violations by soft law of the rule of law parameters. In similar fashion, a number of foundational principles of EU constitutional and administrative law have been rooted by the CJEU as elements of the rule of law, relevant for EU and national authorities alike. These involve principles such as the principle of legality (Case C-496/99 P, Commission v CAS Succhi di Frutta), the principle of legal certainty (Joined Cases C-212 to 217/80, Amministrazione della finanze dello Stato v Salumi), the prohibition of arbitrariness (Joined Cases C-46/87 and 227/88, Hoechst v Commission), the right to a fair trial before an independent and impartial court (Case C-550/09, E and F) and the principle of equality (Case C-550/07 P, Akzo Nobel Chemicals and Akcros Chemicals v Commission).4 The rooting of the rule of law in these principles would seem to imply that soft law, as part of the EU acquis, would also have to respect (and consequently be reviewed against) these principles. These core dimensions of the rule of law have been mostly used to evaluate Member State compliance with EU hard law (see Grogan and van Dam in this volume for the rule of law crisis). But by grounding the assessment of the courts’ role with respect to these dimensions in this chapter, we also argue that the four dimensions of the notion of the rule of law provide a way to think about soft law and its judicial control by illuminating which particular aspects of soft law may prove problematic when evaluated from the perspective of courts.

3.

ARGUMENTS FOR ‘MORE’ COURTS

In sections 3 and 4 we will discuss the arguments for more and less courts, also highlighting how, if at all, the arguments refer to the above-mentioned rule of law dimensions. 3.1

Judicial Review of Soft Law to Control the Exercise of Administrative Discretion

One of the most classical tasks for courts is to control the way in which administrative authorities exercise their discretionary powers. Specifically with respect to EU (soft) law, Article 19(1) TEU states that the task of the CJEU is to ‘ensure that in the interpretation and application of the Treaties the law is observed’. And it is in this context that both national and EU courts have come across soft law. Competition and State aid provided the first policy area for the courts to develop soft law specific case law. Ştefan (2014, see also Ştefan in this volume) has shown how the Commission has since 1960s adopted numerous soft law instruments in

Soft law and courts  195 competition and State aid. Following the turn of the millennium, the use of soft law has blossomed in virtually every EU policy field. This blossoming also quickly showed in EU courts’ case law which has evolved over the years in two parallel directions. First, with respect to soft law adopted by the EU institutions, the CJEU established that soft law instruments are binding on the enacting institution because, by publishing guidelines, the Commission imposes a limit on the exercise of its discretion and cannot depart from those rules or else risk being found in breach of the general principles of law, such as equal treatment or the protection of legitimate expectations, where appropriate (Case C-189/02 P, Dansk Rørindustri and Others v Commission). In other words, there can and should be active judicial engagement with these soft law instruments, which should also function as a reminder to the Commission to do its job as a ‘soft law regulator’ properly in the first place or else risk judicial control. Have the EU courts found also national authorities and courts bound by EU soft law? With regard to the effects that soft law can have on national authorities, it should be noted first that there are no straightforward grounds on which to establish binding legal effects for Member States. Such effects for Member States have traditionally been connected by the EU courts to the duty of cooperation provided for in Article 4(3) TEU; this connection is not however solid. It is not certain whether a general obligation to comply with soft law instruments can be drawn solely from the duty to cooperate. For example, in the Kotnik and others case (C-526/14, para. 38), Advocate General Wahl argued that: the provisions of such acts of ‘soft law’ are, by virtue of the duty of sincere co-operation enshrined in Article 4(3) TEU, to be taken into due account by the Member States’ authorities, that duty cannot be understood as making those rules binding – not even de facto – on pain of eluding the legislative procedure set out in the TFEU.

The EU courts thus seem to attach importance to the principle of sincere cooperation, but carefully avoid establishing or giving the impression that such principle would make soft law either de facto or formally binding in a bid to respect the legislative procedure. The EU courts have also so far rejected arguments that national authorities should consider soft law on the basis of principles such as legal certainty and legitimate expectations. This means that, while the CJEU is ready to accept the argument as far as the Commission (as the prominent soft law regulator) is concerned, it is clearly less eager to apply the same argument to national authorities. While the CJEU has not unequivocally spelt out the reasons, it is national procedural autonomy that seems to matter to the Court here. In the Pfleiderer case (C-360/09), the Court held that soft law can produce effects on the activities of national authorities, but such effects could not be given legal weight in judicial proceedings. In another important judgment defining the boundaries of national discretion (C-226/11 Expedia), the Court of Justice argued that the Notice on agreements of minor importance that do not fall under EU antitrust rules (a piece of soft law in the field of EU competition policy) does not bind national authorities or courts, and that they have complete discretion to take the Notice into consideration when deciding cases. This judgment has been criticized on the grounds that if national authorities disregard the Notice, this can ultimately negatively affect compliance with principles such as legal certainty and have serious consequences from the point of view of individual rights, as legitimate expectations and legal certainty appear to have variable content in the multi-layered system of EU competition law enforcement (Ştefan 2014). The binding nature of soft law on Member State authorities remains a complex issue and one where policy

196  Research handbook on soft law sector differences continue to matter. For instance, in competition and State aid, the Court has hold that if Member States have agreed to and accepted soft law (the case of ‘negotiated soft law’), it is binding on those Member States (Case 311/94 Ijssel-Vliet, Ştefan 2017). Advocate General Kokott (whose line of argument the CJEU did not follow in Expedia) foresaw the risks of allowing national authorities to decide whether or not to take the Notice seriously, arguing that: The Commission’s leading role … would be undermined if the authorities and courts of the Member States simply ignored a competition policy notice issued by the Commission. It therefore follows from the duty of sincere cooperation which applies to all the Member States (Article 10 EC, now Article 4(3) TEU) that the national authorities and courts must take due account of the Commission’s competition policy notices, when exercising their powers under Regulation No 1/2003 (para. 38).

The Expedia judgment is not the first time the CJEU has struggled with national courts’ discretion in terms of soft law. In Grimaldi (C-322/1989), the Court of Justice required national judges to ‘take into consideration’ soft law whenever deciding cases. No further explanation was offered on what exactly this statement was supposed to entail for national courts. Some have argued that it introduces a duty of consistent interpretation in relation to soft law instruments (Arnull 1990). National courts would thus be under a duty to interpret national law in the light of the wording of soft law instruments issued at the European level. Conversely, it has been argued that the reading of this judgment should be less strict, and that national courts are required to take soft law into consideration only when it helps to clarify the meaning of EU or national law (Senden 2004). Performing a longitudinal analysis of the Grimaldi case law, Korkea-aho (2018) shows that, since the 1989 Grimaldi ruling, the CJEU has referred to Grimaldi only seven times. Whilst the CJEU has not reversed the precedent set by Grimaldi, nearly three decades of EU soft law-making have eroded the foundations of the doctrine to the extent that the obligation has become heavily nuanced. First, the Grimaldi obligation has in certain circumstances become stricter. In Case C-28/15 Koninklijke KPN, the Court of Justice seems to have made a 180-degree turn (without even referring to Grimaldi!) by holding that national courts should take soft law into account unless ignoring it is justified by the particular circumstances of the case. This means that, to the extent that the soft law measure is issued by the EU institution and its development is foreseen in EU primary or secondary law, national courts can depart from the interpretation offered in the measure only if they can provide detailed and substantively valid reasons why it should not apply. Second, the more traditional, softer version of the Grimaldi obligation remains for those cases where the soft law instrument is free-standing, that is, it is not derived from EU primary or secondary law, or where non-binding guidance is given by actors other than the EU institutions. In these specific circumstances, the national court has more leeway to decide whether or not to take soft law instruments into account. The third noteworthy feature that emerges from the analyzed jurisprudence is that Member States have become more proactive in challenging EU soft law (see especially Kotnik and others in section 3.2 below). In brief, the EU courts have used soft law to remind the Commission that it is bound by whatever soft law it decides to adopt. In other words, they will review the use of discretion by the Commission against the rules that the Commission itself has adopted. For the reasons of national procedural autonomy, the EU courts have not extended this same obligation to the national level, and they have been clearly reluctant to establish that national courts should

Soft law and courts  197 use soft law (EU or national) to review national administrative action. The combination of Grimaldi and Koninklijke KPN has been criticized for not spelling out the ‘ultimate’ duties of national courts and putting courts ‘in a tunnel with no apparent exit’ (Gentile 2021, p. 999). The tunnel effect is exacerbated by the fact that there is uncertainty how the CILFIT doctrine (which establishes that if a judgment or rule of law is clear enough, then a Member State court has no duty to refer) should be applied to soft law instruments (Cuyvers and van Dam 2021). If the first direction concerned the EU institutions and soft law, the second direction in which EU case law developed concerns citizens, companies, and non-governmental organizations (NGOs) resorting to soft law, arguing that EU courts should review soft law which they deem is in breach of higher EU law provisions. Such hopeful applicants need to overcome two hurdles. First, for a measure to be reviewable by way of judicial review, it must have been intended to produce legal effects, as stipulated in Article 263 of the Treaty on the Functioning of the European Union (TFEU). The first case that specifically addressed this was C-22/70 ERTA, which concerned minutes of the Council regarding the negotiation and conclusion of an international agreement. In this case, the Court of Justice held that ‘an action for annulment must be available in the case of all measures adopted by the institutions, whatever their nature or form, which are intended to have legal effects’ (para. 42). Traditionally, the EU courts have been very restrictive in the assessment of whether a soft law measure produces legal effects, with the consequence that claims against such measures have often been unsuccessful (for the resulting ‘asymmetric’ relationship between Articles 263 and 267, see de Witte in this volume). Nearly three decades later, a revolution nearly happened. In Case T‑721/14, Belgium v Commission, the question with which the General Court was confronted was whether a Commission recommendation is capable of having legal effects (on the effects of Commission recommendations see van Dam 2017). The General Court examined its wording and context, its substance and the intention of its author, and held that it did not have binding legal effects on its addressees by giving more weight to the wording than to other factors (see paras 21, 22, 26). The outcome meant that recommendations could not be subject to annulment actions. Belgium appealed, and found a friend in Advocate General Bobek, for whom, ‘the General Court erred in law: it incorrectly interpreted the effects of the Recommendation at issue, and thus it incorrectly assessed the admissibility of the application’ (Case C‑16/16 P, para. 54). Bobek argued that even though soft law is not binding in the traditional sense, it can be seen as a type of imperfect norm that can contain ‘mild obligations’ (para. 86). In the end, the Court of Justice disagreed with Bobek, and sided with the General Court (Arnull 2018), stopping the revolution in its tracks. At the moment, on the basis of the applicable case law, the Court’s understanding of the conditions under which a soft law measure can be deemed to produce legal effects remains very restrictive and somewhat erratic, a matter which has been criticized by several scholars (Eliantonio 2021; Xanthoulis 2021; Türk and Xanthoulis 2019), who have also pleaded for a ‘liberal-constitutional’ reading of the Treaties, requiring ‘the creation of check and balances among the traditional “state” powers and the judicial protection of individual liberties and rights’ (Gentile 2020, p. 468). Apart from situations where soft law binds the institution which has adopted it, the production of legal effects has been recognized only in a handful of situations, such as where the soft law measure at stake has been agreed to by Member States (so-called negotiated soft law, Case 311/94 Ijssel-Vliet) or where the measure can be construed

198  Research handbook on soft law as introducing a new obligation (Case C-366/88 France v Commission, see along the same lines, also Case C-325/91 France v Commission). Establishing legal effects is just the first hurdle. The second hurdle is the so-called ‘authorship criterion’. Article 263 TFEU annulments exclude authors other than those expressly mentioned in that provision, namely acts of the EU institutions and those of ‘bodies, offices or agencies of the Union’. Many soft law documents are ‘jointly’ produced by the Commission and various networks (including national authorities but also experts and NGOs), which raises the question of whether they are ‘authored’ by the EU institutions in the meaning of Article 263 or not (Scott 2011). The increasing body of case law on technical standards illustrates these difficulties but offers also some possible ‘inroads’ to go beyond a strict interpretation of the authorship criterion (Eliantonio 2018). In Case C-613/14 James Elliott Construction, the Court argued that, while not being ascribable to EU ‘institutions, bodies, offices or agencies’, harmonized technical standards are ‘part of EU law’ as ‘measures implementing or applying an act of EU law’. Despite more voices in an ever-growing choir of scholarship and the court’s own Advocates General calling for judicial review of soft law instruments, the EU courts have been careful when engaging with soft law and have only recognized certain legal effects stemming from soft law measures in limited circumstances. They have, however, imposed on national courts some duties to take soft law into account. A large multi-country and multi-policy study SoLaR (see Eliantonio et al. 2021) shows that national courts do indeed take EU soft law into account, by using it to interpret certain terms in the underlying EU legislation, reviewing the exercise of discretion by national authorities when acting within the scope of application of EU law and ensuring fair and equal treatment. However, there is a certain degree of ‘unease’ with respect to national courts and EU soft law, demonstrated by the fact that the non-binding nature of soft law is continuously and relentlessly restated. Recently, strong calls for ‘more courts’ have been made with references to the imperative to ensure effective judicial protection as required by Article 47 of the EU Charter of Fundamental Rights. The need to ensure effective judicial protection seems particularly pressing given the legal and practical effects that soft law generates in the EU legal order and in light of a growing trend at the national level to open the judicial doors to soft law, a trend also explicitly highlighted by the opinion of Advocate General Bobek in the Belgium v Commission case mentioned above. In a comparative study, Eliantonio (2021) shows how the national courts of several Member States have traditionally admitted (or have recently moved towards admitting) claims against national soft law measures in situations where claims in EU courts would not have been successful (for the impact of EU competition soft law on national courts, see also Georgieva 2014). For example, English courts are traditionally open towards claims against national non-binding measures; Italian and German courts also admit such measures, provided that certain ‘external effects’ are generated (see also Hartlapp and Hofmann 2021); French courts have instead evolved from a very restrictive approach towards judicial review of soft law to a much more open one, which also has kept pace with the increasing use of non-binding measures as regulatory instruments of administrative governance (Rubio and Ştefan 2021). This suggests that national courts are growing increasingly sensitive to performing their roles as protectors of the rule of (soft) law. However, given the general gist of the case law, it is unlikely that the CJEU will follow suit any time soon.

Soft law and courts  199 3.2

Judicial Engagement of Soft Law as an Acknowledgement of the Hybridity of the Legislative Framework

Another group of scholars think – like those referred to above – that courts need to engage (more) with soft law. While they also call for ‘more’ courts in respect of soft law, their argumentation to call for more significant judicial engagement is different. The scholars, whose ideas we address here, think that the courts cannot avoid considering soft law, because soft law is such an integral part of the EU legislative framework in several policy areas. At the moderate end of this scholarship are those views according to which, on the basis of the CJEU case law (in particular the Grimaldi and Koninklijke KPN rulings mentioned above) European soft law is used as an interpretation aid, voluntarily in the case of the Court of Justice, but de facto mandatorily for national courts. For this reason, so the argument goes, the national courts in particular have received an ‘EU mandate’ to engage with soft law and cannot ignore its existence and role in the relevant EU legislative framework. In relation to environmental protection (a field where EU soft law is abundant and a necessary complement to EU hard law), Scott (2011) positively comments on the fact that the EU courts look beyond the lack of legally binding force and admit that soft law can produce certain legal effects. However, she considers that the CJEU does not go far enough, and disregards many of the legal effects of soft law, thus, often, depriving such instruments of judicial scrutiny. In this sense, the argument here is that EU courts should control EU soft law, as avoiding doing so would imply a gap of not only protection but also understanding vis-à-vis the applicable regulatory framework as a whole. In another study on environmental protection, Korkea-aho (2015a) argues that especially soft law produced by actors associated with the implementation and application of the Water Framework Directive and the EU’s Chemicals Regulation REACH is important to understand the legislative framework. It also has relevance for courts, because soft law can shape judicial decision-making, and, at best, even positively support the functioning of the courts by increasing the information available to them when interpreting the legislative framework and reviewing administrative action. These scholars who call for ‘more’ courts all highlight how legislation is evolving towards more open-ended legislation, with soft law consequently needed to fill in gaps, thereby embedding itself in the normative framework. For example, framework directives – an instrument used in the area of environmental law in particular – are an example of where soft law is needed to implement them. This kind of implementing soft law, which Korkea-aho (2015b) has called ‘administrative reasoning’ embodies the work taken up by multiple actors in the form of soft law guidance. Its use simplifies the work done by courts, but also acknowledges the important work done by several stakeholders at different stages of the policy-making process (see also Komesar and Wagner 2017 in the US context). The EU courts have also endorsed this position. In Case C-106/14, FCD and FMB, para 28, the Court of Justice argued that: a document such as the ECHA Guidance document may be one of the factors to be taken into consideration in interpreting the REACH Regulation. However, despite the scientific and technical nature of the aspects relating to the chemical substances regulated by the REACH Regulation, a document of that nature remains purely explanatory. The interpretation it gives of that regulation’s provisions is of no legislative effect whatsoever. It is a document drawn up by the ECHA and is not among the legal acts of the European Union referred to in Article 288 TFEU; accordingly it cannot be of a legally binding nature [emphasis added].

200  Research handbook on soft law Either implicitly or more openly all the above arguments are premised on the idea that the use of soft law by courts is not simply an ideological choice for the courts to make (to fulfil some normative ideal of the courts’ role in a democratic society). But rather, it is a pragmatic move to make sense of legislation they are called on to review. If we accept the role of soft law as providing interpretative support and adding to the normative framework, it is necessary to take note of preliminary rulings. According to established case law, Member States can refer questions about soft law to the EU courts. The judgment in Case C‑526/14 Kotnik and Others is an interesting recent example. The case – the first reference for a preliminary ruling from the Slovenian Ustavno sodišče (Constitutional Court) – concerns a Banking Communication issued by the Commission to provide a framework for the assessment of the compatibility of State aid to banks during the financial crisis. The referring court sought guidance from the Court on, in particular, the validity and interpretation of the provisions of that communication. The Court does not comment on the formulation of the referring court and proceeds to discuss the recommendation, as requested by the referring court. This suggests that the door is open for national courts to engage EU courts via the preliminary ruling procedure and in this way to contribute to the normalization of soft law as a governance mechanism (see also C-911/19 Fédération bancaire française). As Marjosola et al. 2022 (see also de Witte in this volume) argue, the national route also offers non-privileged applicants a way to challenge soft law of, in the latter case, EU agencies. The jury is still out on whether judicial engagement with soft law would nurture rule of law compliant soft law, with Scott and Sturm (2007) arguing for the potential of the courts to emphasize the observance of rule of law values. In their view, courts are in a good position to influence the deliberation processes, by determining the standards for the review of soft law and enforcing them. Ştefan (2014) adopts a more cautious approach. What is essential to her is that the EU courts rely on a mechanism based on good administration principles such as legitimate expectations, legal certainty, and transparency, in order to recognize the legal effects of soft law. This way the EU courts re-institutionalize soft law that lacks formal guarantees of legitimacy, while grounding it in rule of law principles that form part of the constitutional values common to the Member States. However, the grounds and the intensity of effects that the courts recognize to soft law vary in accordance with the level within which they are invoked – European or national. Effective judicial protection – an imperative protected by Article 47 of the Charter of Fundamental Rights – should not lead to different results depending on the system in which protection from courts is sought in the first place.

4.

ARGUMENTS FOR ‘LESS’ COURTS

4.1

The Dangers of Blurring the Distinction Between Hard Law and Soft Law

The most important criticism of the use of soft law in court is that by engaging with soft law, courts can, without the intention to do so, transform soft law into hard law or judicialize the political aspect of soft law (for judicialization of governance, see Stone Sweet 1999). How exactly this transformation occurs is explained differently by different scholars of soft law. They also assess potential risks associated with this transformation differently. As a long-time critic of soft law, Klabbers has pointed out that use of soft law instruments in court is simply undesirable. He argues that, whenever engaging with soft law, the domestic

Soft law and courts  201 and international courts try to ‘recast it into the more accepted sources of international law: treaties and custom’ (1998, p. 174) and that soft law becomes ‘completely indistinguishable from hard law’ (1998, p. 177) whenever applied, complied with or violated. For Klabbers, the risk of courts engaging with soft law seems to relate to the authority and legitimacy of hard law, which in the area of international law is in short supply to begin with. As explained by Wellens and Borchardt (1989, p. 291), it is not enough that the soft law argument is part of the obiter dicta of the judgment: it must be its ratio decidendi in order to become hard by judicial intervention. However, transformation from soft law to hard law could occur solely by the mention of soft law, i.e., by individuals referring to it or by challenging a soft law instrument via Article 263 TFEU. Given the limited scope for individuals to challenge soft law instruments, this type of transformation takes place only rarely. A more promising avenue to observe the transformation from soft law to hard law is the interplay between the Commission and the EU courts. Namely, by publishing its soft law, the Commission makes it accessible and effectively disseminates it, but also runs the increased risk of it being challenged in court. Snyder (1993, p. 65) has pointed out that the CJEU might decide that ‘the putatively soft law has hard legal consequences’ which would blur the distinction between soft and hard law. Hofmann (2006, p. 165) has also argued that by recognizing legal effects through the operation of principles such as legitimate expectations or legal certainty, the EU courts furnish the competition guidelines of the Commission with certain hard features. Finally, Österdahl (2004, p. 51) has argued that the repeated references to soft law instruments by the Advocates General and by both courts of the EU might contribute to their ‘hardening’, even if they are qualified as not legally binding. For Österdahl, the transformation of soft law into hard law does not require any judicial gimmicks as such, the hardening simply takes place due to the positive reputation soft law instruments gain from being used by the courts. There is an element of repeat play to the interactions between the Commission and EU courts. While analyzing the consequences of the Deufil judgment (Case 310/85), Snyder (1994, p. 204) came to the conclusion that the activity of the Court and that of the Commission have an effect on each other ‘such that the result of each institution’s decisional processes are incorporated as an input into the decisional processes of the other’. His detailed explanation of the mutual dynamics is that a soft law instrument is issued by the Commission and subsequently taken into consideration by the EU courts in a judgment. The Commission then uses the court’s decision, while elaborating upon and generalizing the judicial language in order to justify the incorporation of the soft law instrument into hard law. Hence, Snyder (1993, p. 216) argues that ‘soft law, based partly on a court judgment, is transformed into hard law by administrative decision’. While not acknowledging it explicitly, many of the above arguments see the risk associated with the transformation of soft law into hard law in erasing the boundary between what is law and what is soft law. It should be clarified that the hardening of soft law is not seen to threaten hard law by softening or ‘delegalizing’ it (see also Terpan 2015). However, there is another risk that relates to the authority of the court itself. If the court is too relaxed with soft law, it may signal the erosion in the authority of the court to uphold the law. 4.2

The Appropriateness of Courts as a Forum to Fix the Problems with Soft Law

A second group of scholars has also called for ‘less’ courts in respect of soft law. Their views are, however, not linked to the notion of hard versus soft law but to the courts themselves as

202  Research handbook on soft law loci of control. These views all share the premise according to which the judiciary should keep its distance from soft law. Some argue that this is because the courts have a less than ideal tendency to subordinate the rationality of soft law process to their own professional and political preferences (for professional and political projects with respect to soft law, see di Robilant 2006). There are also scholars who think that courts are not the appropriate place to fix soft law’s possible problems because soft law frameworks may already entail capacities to solve problems without judicial intervention. Contextualizing her argument in the implementation of the Water Framework Directive (WFD) in particular, Korkea-aho (2014) has suggested that managing implementation through soft law-based networks reduces courts’ caseloads, as problems are solved by participants, without judicial intervention. Furthermore, soft enforcement mechanisms such as peer review processes might streamline legal actions, by offering the information and the administrative resources necessary to manage infringement proceedings effectively. Similarly, Armstrong (2011) thinks that adjudication as an activity pertains to more than simply applying the rules and often involves complex assessments of interests and values that could ultimately encourage litigants to rely on new forms of governance, soft law and cooperation as an alternative to courts. These views are premised on the notion that participants to governance networks are epistemically better qualified to solve problems than courts, and hence the former should be preferred over the latter. But the courts’ wavering epistemic authority is not the only reason to prefer soft law and cooperation. In his 1981 book Democracy and Distrust, John Hart Ely raised a fundamental question of whether the judiciary as a non-accountable institution should have authority over the democratic process – a question that became central to the ‘new governance’ research (in the context of which soft law was an important instrument to study) in the early 2000s. The common claim was that, in general, the new governance context should prompt nothing less than a radical rethinking of the role of courts. In this context, where peer-review mechanisms are as important as judicial control, where governance is dispersed through multiple layers of authorities (and might possibly also include private parties), where rules have different ‘degrees’ of bindingness, courts should not be seen as ‘mere’ enforcers of legal rules but rather ‘a source of communicating ideas and experience [...] without being specifically prescriptive in relation to any particular form’ (Scott and Sturm 2007, p. 572). Providing ‘a bottom-up analysis of judicial review’, Komesar and Wagner (2017) argue that different constellations of democratic participation ‘help determine the behaviour of the political and administrative processes’, and also ‘the behaviour of the adjudicative process, and, therefore, of judicial review’. In other words, while the courts may not be the best place to resolve problems that emerge in relation to soft law, they may nevertheless be useful, especially if willing to reinvent their role and abandon prescriptive tendencies and further support bottom-up participation.

5.

SOFT LAW AND COURTS: SAVIOURS OR SABOTEURS OF THE RULE OF (SOFT) LAW?

The above discussion has shown that very different views exist in relation to how much courts should (not) engage with soft law. Despite this plurality, two general observations can be made.

Soft law and courts  203 First, what we above identified as the rule of law values can be used to either support a stronger role for courts as well as to advocate for a view that courts should keep their distance from soft law. Legal certainty and the prevention of abuses, the first and second dimensions of the rule of law, are in particular such versatile rule of law values, being able to provide normative ammunition for both ‘saviours’ and ‘saboteurs’ scholars. On the one hand, Chinkin (1989) has argued that soft law and courts should not mix. Due to its informative and educative role (as opposed to prescriptive role), soft law is unsuitable for judicial dispute settlement, but potential disputes could be subjected to non-judicial forms of dispute settlement. Even though she acknowledges the situations where courts have paid attention to soft law instruments, Chinkin concludes that soft law is unsuitable for adjudication and that any court enforcement of soft law instruments may come at the expense of legal certainty. On the other hand, those scholars who argue that soft law should be controlled for potential abuses also invoke the value of legal certainty. Courts can review and reject the logics of soft law governance if it is considered contrary to hierarchically superior legal requirements. For scholars, it seems potentially dangerous to let for instance the Commission adopt soft law without courts controlling it against the rule of law values such as legal certainty. The third dimension of the rule of law is non-discrimination. This may be the hardest to translate into soft law context. However, this particular element could be seen to offer support to those scholars who are initially wary of courts intervening in the soft law environments. For them, courts’ engagement with soft law interferes with soft law’s experimentalist nature, and any such active role, especially if being ‘specifically prescriptive in relation to any particular form’ (Scott and Sturm 2007, p. 572), would discriminate against soft law by taming its innovative energy. However, as argued by Emerson in the US context (this volume, p. 246), soft law ‘may convey “official recognition” that a person’s identity or status is worthy or respect or that the harms they have suffered deserve redress [text and citations removed] Because the guidance issues from officials entrusted with the authority to interpret and administer law, it carries with it the badges of trust, respect, and good faith that attach to the official role’. For this reason, the principle of non-discrimination should encourage courts (and scholars) to closely assess soft law and its moral and political impact, not just its legal effects. The fourth dimension of the rule of law relates to access to justice. At a first sight, this could be seen to provide support for those who are in favour of courts, and this dimension remains a strong reminder of the importance of a guarantee that soft law not be cut off from effective judicial protection. However, recognition of the importance of access to justice in the soft law context involves reimagining what the courts can do. For instance, Scott and Sturm (2007) see the potential of the courts in ensuring the observance of rule of law values and principles within governance structures. In recent US literature on agency rule-making, scholars have set out to reimagine the role of courts. Agency rule-making is an ideal environment to raise questions about the courts’ role, because US agencies use guidance (the term used in the US to refer to soft law-type rules) to regulate. Kessler and Sabel (2021, p. 191) note that guidance ‘is the kind of law that uncertainty admits’, suggesting that the more uncertain regulators become the more attractive soft law looks to them. Asking what the courts’ proper role would be in these circumstances they continue suggesting that (p. 200): ‘our alternative program would refocus courts and agencies on what some commentators have called process review: review of the chain of reasoning that has led an agency to adopt and maintain a particular guidance regime’. Transferred into soft law-making, this would mean that courts, when dealing with soft law, should be more interested in reasons that have led to the adoption of soft law, rather than

204  Research handbook on soft law checking if procedural requirements have been met or that an outcome makes sense or is substantively correct and effective. The second broader observation relates to the nature of soft law. We suggest that those authors who are sceptical of courts’ engagement with soft law tend to think that soft law is a mature form of regulation and not only some immature form of legal rules or stepping stones towards legislation. In other words, courts are not needed to make soft law complete or suited to regulating in complex environments. Soft law is not hard law, but neither is it some inferior form of law that needs be put right by courts. This, of course, varies between policy fields, and in some areas, where soft law is the illustration of principles laid down by hard law, it is easier for courts to endorse a ‘soft law’ position. In this respect, scholars who see soft law as a mature form of regulation also tend to hold that the relationship between courts and soft law could be represented as one of mutual transformation. One avenue for such transformation opens in situations whereby the interpretation given by the courts to legal instruments becomes ‘embedded in the context and experience of “new governance”, bargaining or informal settlement processes’ (Hervey 2010, p. 144). In the introduction, we described the courts’ current approach to soft law as an effort to have ‘the best of both worlds’, while wondering if it may be realistic or desirable. Based on the engagement with different arguments above, we cautiously admit that it may be possible to have the best of both worlds, if the courts judge wisely in the area that emerges between the need to ensure effective judicial protection and protect individual rights on the one hand and the need to preserve the resilient, uncertainty-embracing nature of soft law on the other. While this is not an easy task to pull off, Member States’ courts show that it can be done.

6. CONCLUSION Soft law is nowadays part and parcel of the regulatory toolbox of international, supranational, national and regional decision-makers. Despite its ubiquitous presence, its legal effects cannot be clearly determined, yet it is an authoritative source of rules of behaviour, capable of nudging public authorities and individuals alike into following certain conduct. The ‘gray area’ in which soft law moves has created a sense of ‘unease’ in both the courts and in the scholars discussing the role that courts should have vis-à-vis soft law. Our chapter has attempted to disentangle the various arguments used to justify more or less court engagement with soft law, with a specific focus on EU soft law and the case law of the CJEU. The identification and analysis of these arguments showed that rule of law-related values (such as the principles of legal certainty, non-discrimination or effective judicial protection) can serve to justify and support both a stronger role for courts in the control of soft law (‘the saviour role’), and the argument that courts should keep their distance from soft law so that they do not occupy ‘the saboteur role’. The CJEU has so far been very cautious with engaging and controlling soft law, while national courts have been much more open and flexible in this respect. There are good arguments for the latter having done this, not least because of the integral role that soft law plays in the EU legislative framework. Will this then foster illegitimate ways of decision-making? Will it blur the distinction between hard and soft law? We posit in this chapter that a middle-way is perhaps possible if judges are sensitive to both the imperatives of preserving the principle of legality as well as those of not depriving soft law of its unique features and regulatory advantages as a resilient ‘bamboo’ type of law as suggested

Soft law and courts  205 by Snyder in this volume. Whether the courts are able to find and walk this middle way should be further researched with respect to not only the EU level courts, but also national and international courts through comparative research.

NOTES 1.

See https://​erasmus​-plus​.ec​.europa​.eu/​projects/​search/​details/​575097​-EPP​-1​-2016​-1​-FI​-EPPJMO​ -NETWORK [viewed 14 June 2023]. 2. See https://​www​.venice​.coe​.int/​WebForms/​pages/​?p​=​02​_Rule​_of​_law​&​lang​=​EN [viewed 14 June 2013]. 3. See https://​www​.venice​.coe​.int/​WebForms/​pages/​?p​=​02​_Rule​_of​_law​&​lang​=​EN [viewed 14 June 2013]. 4. See also Annex I ‘The Rule of Law as a foundational principle of the Union’ to the Communication from the Commission to the European Parliament and the Council, ‘A new framework to strengthen the Rule of Law’, COM(2014)158 final, 11 March 2014, p. 2–3).

REFERENCES Armstrong, K., (2011). The Character of EU Law and Governance: From ‘Community Method’ to New Modes of Governance. Current Legal Problems. 64(1), 179–214. Arnull, A., (1990). The Legal Status of Recommendations. European Law Review. 15(4), 318–321. Arnull, A., (2018). EU Recommendations and Judicial Review: ECJ 20 February 2018, Case C-16/16 P, Kingdom of Belgium v European Commission. European Constitutional Law Review. 14(3), 609–621. Case C-22/70, Commission v Council (ERTA), EU:C:1971:32. Joined Cases C-212 to 217/80, Amministrazione delle finanze dello Stato v Salumi, EU:C:1981:270. Case 310/85, Deufil GmbH & Co. KG v Commission, EU:C:1987:96. Joined Cases C-46/87 and 227/88, Hoechst v Commission, EU:C:1989:337. Case C-322/88, Grimaldi v Fonds des maladies professionnelles, EU:C:1989. Case C-366/88. France v Commission, EU:C:1990:348. Case C-325/91, France v Commission, EU:C:1993:245. Case C-311/94, IJssel-Vliet Combinatie BV v Minister van Economische Zaken, EU:C:1996:383. Case C-496/99 P, Commission v CAS Succhi di Frutta, EU:C:2004:236. Case C-189/02 P, Dansk Rørindustri and Others v Commission, EU:C:2005:408. Case C-550/07 P, Akzo Nobel Chemicals and Akcros Chemicals v Commission, EU:C:2010:512. Case C-360/09, Pfleiderer AG v Bundeskartellamt, EU:C:2011:389. Case C-550/09, E and F, EU:C:2010:382. Case C-226/11, Expedia Inc. v Autorité de la concurrence and Others, EU:C:2012:795 Case C-106/14, Fédération des entreprises du commerce et de la distribution (FCD) and Fédération des magasins de bricolage et de l’aménagement de la maison (FMB) v Ministre de l’écologie, du développement durable et de lʼénergie, EU:C:2015:576. Case C-526/14, Tadej Kotnik and Others v Državni zbor Republike Slovenije, EU:C:2016:767. Case C-613/14, James Elliott Construction Limited v Irish Asphalt Limited, EU:C:2016:821. Case T‑721/14, Belgium v Commission, EU:T:2015:829. Case C-28/15, Koninklijke KPN NV and Others v Autoriteit Consument en Markt (ACM), EU:C:2016:692. Case C‑16/16 P, Belgium v Commission, EU:C:2018:79. Case C-911/19 Fédération bancaire française (FBF) v Autorité de contrôle prudentiel et de résolution (ACPR), EU:C:2021:599. Chinkin, C., (1989). The Challenge of Soft Law: Development and Change in International Law. The International and comparative law quarterly. 38(4), 850–866.

206  Research handbook on soft law Cuyvers A. and Dam J. C. A. van, (2021). Where Soft Law Meets CILFIT: Is a Court Ever Obligated to Refer Questions on Soft Law?. In J.W. Verboeket, J.E. van den Brink, A. Drahmann, M. J. Jacobs and R. Ortlep, eds. Bestuursrecht in het echt. Vriendenbundel voor prof. mr. drs. Willemien den Ouden. Deventer: Wolters Kluwer, pp. 298-318. Dawson, M., (2009). Soft Law and the Rule of Law in the European Union: Revision or Redundancy?. EUI Working Papers (RSCAS) 2009/24. Available from: https://​ssrn​.com/​abstract​=​1415003 [viewed 15 June 2023]. Dam, J. C. A. van, (2017). Guidance documents of the European Commission: a typology to trace the effects in the national legal order. Review of European Administrative Law. 10(2), 75–91. Eliantonio, M., (2018). Soft Law in Environmental Matters and the Role of the European Courts: too much or too little of it? Yearbook of European Law, 37, 496–524. Eliantonio, M., (2021). Judicial Review of Soft Law before the European and the National Courts: A Wind of Change Blowing from the Member States? In: M. Eliantonio, E. Korkea-aho and O. Ştefan, eds. EU Soft Law in the Member States: Theoretical Findings and Empirical Evidence. Oxford: Hart Publishing. pp. 283–301. Eliantonio M., Korkea-aho E. and Ştefan O., eds., (2021). EU Soft Law in the Member States: Theoretical Findings and Empirical Evidence. Oxford: Hart Publishing. Ely, J. H., (1981). Democracy and Distrust. A Theory of Judicial Review. Harvard University Press. Fuller, Lon L., (1964). The Morality of Law. Yale University Press. Ganz, G., (1987). Quasi-Legislation: Recent Developments in Secondary Legislation. Sweet & Maxwell. Gentile, G., (2020). Ensuring Effective Judicial Review of EU Soft Law via the Action for Annulment before EU Courts: a plea for a liberal-constitutional approach. European Constitutional Law Review. 16(3), 466–492. Gentile, G., (2021). To be or not to be (legally binding)? Judicial review of EU soft law after BT and Fédération Bancaire Française. Revista de Derecho Comunitario Europeon, 70, 981–1005. Georgieva, Z., (2014). Soft Law in EU Competition Law and its Judicial Reception in Member States – A Theoretical Perspective (October 2014). TILEC Discussion Paper No. 2014-035. Available from: https://​ssrn​.com/​abstract​=​2506617 [viewed 15 June 2023]. Hartlapp, M. and Hofmann, A., (2021). The Use of EU Soft Law by National Courts and Bureaucrats: how relation to hard law and policy maturity matter. West European Politics. 44(1), 134–154. Hervey, T., (2010). Adjudicating in the Shadow of Informal Settlement?: The Court of Justice of the European Union, ‘New Governance’ and Social Welfare. Current Legal Problems. 63(1), 92–152. Hofmann, H. C. F., (2006). Negotiated and Non-Negotiated Administrative Rule-Making: the Example of EC Competition Policy. Common Market Law Review. 43(1), 153–178. Kessler, J. K. and Sabel, C. F., (2021). The Uncertain Future of Administrative Law. DAEDALUS. 150(3), 188–207. Available from: https://​scholarship​.law​.columbia​.edu/​faculty​_scholarship/​2737 [viewed 15 June 2023]. Klabbers, J., (1998). The Redundancy of Soft Law. Nordic Journal of International Law. 65(2), 167–182. Komesar, N. and Wagner, W., (2017). The Administrative Process from the Bottom up: reflections on the role, if any, for judicial review. Administrative Law Review. 69(4), 891–949. Korkea-aho, E., (2014). Watering Down the Court of Justice? The Dynamics Between Network Implementation and Article 258 TFEU Litigation. European Law Journal. 20(5), 649–666. Korkea-aho, E., (2015a). Adjudicating New Governance: Deliberative Democracy in the EU. Routledge. Korkea-aho, E., (2015b). Legal Interpretation of EU Framework Directives: A Soft Law Approach. European Law Review. 40(1), 70–88. Korkea-aho, E., (2018). National Courts and European Soft Law: Is Grimaldi Still Good Law? Yearbook of European Law. 37, 470–495. Marjosola, H., van Rijsbergen, M. and Scholten, M., (2022). How to Exhort and to Persuade With(out legal) Force: Challenging soft law after FBF. Common Market Law Review. 59(5), 1523–1542. Österdahl, I., (2004). The ECJ and Soft Law: Who’s Afraid of the EU Fundamental Rights Charter? In: U. Mörth ed. Soft Law and Governance and Regulation: An Interdisciplinary Analysis. Edward Elgar. di Robilant, A., (2006). The Genealogies of Soft Law. The American Journal of Comparative Law. 54(3), 499–554. Rubio, N. and Ştefan, O., (2021). No Longer Small Enough to Fall through the Cracks: A French Story of Adaptation to the Petites Sources of EU Law. In: M. Eliantonio, E, Korkea-aho and O. Ştefan, eds.

Soft law and courts  207 EU Soft Law in the Member States: Theoretical Findings and Empirical Evidence. Hart Publishing. pp. 135–149. Scott, J., (2011). In Legal Limbo: Post-legislative Guidance as a Challenge for European Administrative Law. Common Market Law Review. 48(2), 359–356. Scott, J. and Sturm, S., (2007). Courts as Catalysts: Rethinking the Judicial Role in New Governance. Columbia Journal of European Law. 13(3), 565–594. Senden, L., (2004). Soft Law in European Community Law. Oxford: Hart Publishing. Senden, L., (2013). Soft Post-Legislative Rulemaking: A Time for More Stringent Control. European Law Journal. 19(1), 57–75. Smismans, S., (2010). From Harmonization to Co-Ordination? EU Law in the Lisbon Governance Architecture. Journal of European Public Policy. 18(4), 504–524. Snyder, F., (1993). The Effectiveness of European Community Law: Institutions, Processes, Tools and Techniques. Modern Law Review. 56(1), 19–54. Snyder, F., (1994). Soft Law and Institutional Practice in the European Community. In: S. Martin ed. The Construction of Europe: Essays in Honour of Emile Noël. Dordrecht/Boston: Kluwer Academic Publishers. Ştefan, O., (2012). Soft law in Court - Competition Law, State Aid, and the Court of Justice of the European Union. Alphen aan den Rijn: Kluwer Law International. Ştefan, O., (2014). Helping Loose Ends Meet? The Judicial Acknowledgement of Soft Law as a Tool of Multi-Level Governance. Maastricht Journal of European and Comparative Law. 21(2), 359–379. Ştefan, O., (2017). Soft Law and the Enforcement of EU Law. In: A. Jakab and D. Kochenov, eds. The Enforcement of EU Law and Values: Ensuring Member States’ Compliance. Oxford University Press, pp. 200-217. Ştefan, O., Avbelj, M., Eliantonio, M., Hartlapp, M., Korke-aho, E., and Rubio, N., (2019). EU Soft Law in the EU Legal Order: A Literature Review (March 4, 2019). Available from: https://​ssrn​.com/​ abstract​=​3346629 [viewed 15 June 2023]. Stone Sweet, A., (1999). Judicialization and the Construction of Governance. Comparative Political Studies. 31(2), 147–184. Terpan, F., (2015). Soft Law in the European Union - The Changing Nature of EU Law. European Law Journal. 21(1), 68–96. Türk, A. H., and Xanthoulis, N., (2019). Legal Accountability of European Central Bank in Bank Supervision: A Case Study in Conceptualizing the Legal Effects of Union Acts. Maastricht Journal of European and Comparative Law. 26(1), 151–164. Wellens, C. and Borchardt, G. M., (1989). Soft Law in European Community Law. European Law Review. 14(5), 267–321. Venice Commission., (2016). Rule of Law Checklist, adopted by the Venice Commission at its 106th Plenary Session (Venice, 11–12 March 2016). Available from: https://​www​.venice​.coe​.int/​webforms/​ documents/​default​.aspx​?pdffile​=​CDL​-AD(2016)007​-e [viewed 15 June 2023]. Xanthoulis, N., (2021). Legal Effects and Reviewability of EU Acts: Between Institutional Intentions and Perceptions of Concerned Parties. In: M. Eliantonio, E. Korkea-aho and O. Ştefan, eds. EU Soft Law in the Member States: Theoretical Findings and Empirical Evidence. Oxford: Hart Publishing. pp. 303–321.

14. EU economic governance, agencies and soft law: an accountability challenge for the courts? Jacint Jordana and Joan Solanes Mullor

1. INTRODUCTION EU governance is a complex system which tries to encompass both hierarchical and non-hierarchical modes of governance (Tömmel 2016), to interlink multiple territorial levels of government (Schakel et al. 2015), to respect diversity while promoting unity and harmonization (Bieber and Bieber 2020), and to pursue the creation and promotion of an entirely new system of law – EU law – which emphasizes the supranationalization of the national legal orders (Solanes Mullor 2019a). In short, EU governance has been a revolution in the last century from the point of view of the social sciences. This revolution has also entailed new challenges from the perspective of accountability which had to be adapted to the complexity of the new phenomenon (Bovens 2007). This chapter examines the use of soft law as one of the key developments of EU governance, aiming to discuss the political and legal accountability challenges it has created. Soft law has been relevant within EU governance since its beginning (Peters 2011), but it has gained momentum with both the EU agencification process – largely supporting the consolidation of the single market – and the unique characteristics of some sectors, such as the so-called ‘EU economic governance’. This key role of soft law also entails a debate about its accountability (Dawson 2015). Here we consider a problem emerging from a particular type of accountability and its difficulties in overseeing soft law – legal accountability – that has been looking for its proper role together with political and social accountability (Markakis 2020). The chapter is divided into two parts. The first part presents the singularities and the position of soft law within EU economic governance. It explores the concept of soft law vis-à-vis the ordinary EU legislative procedure and highlights its importance for the EU agencification process and other institutions involved in EU economic governance. In particular, this first part briefly explores the relevance of soft law for the activity of EU agencies and economic institutions like the ECB and its place in the overall structure of EU economic governance. The second part of the chapter presents two cases – the Economic and Monetary Union (EMU) and the European System of Financial Supervision (ESFS) – in which the use of soft law presents different patterns: while the former takes form in a context of a highly political institutional framework with less room for agencies, the latter finds accommodation in a context with a deep presence of the EU agencification process and in a more advanced European integration stage. The role of the European Central Bank (ECB) also differs in the two cases. We will see that its less defined powers within the EMU, in which soft law and intergovernmentalism are at the peak, have provoked conflicts with national judiciaries, while to the contrary, the more defined position of the ESFS has facilitated its control by the Court of Justice of the European Union (CJEU). Thus, the implications for legal accountability are different in the two cases: the different intensification of the use of soft law and agencies lead 208

EU economic governance, agencies and soft law  209 to greater difficulties in combining legal accountability with other forms of accountability in the case of EMU, while it emerges as more articulated in the case of ESFS.

2.

SOFT LAW IN THE EU

2.1

Defining Soft Law as a Part of EU Governance

Abbott and Snidal (2000) identify three views on soft law. First, soft law can be seen as a second best option that can destabilize the existing legal framework by creating a decoupling of norms. Second, it can be seen as an intermediate stage towards the formalization of rules that can be facilitated by the development of soft law. Third, soft law can be seen as an alternative, with its own benefits, that can be better adapted to some cases. They also suggest, regarding this third perspective, that in many cases ‘softer forms of legalization will be more attractive to states as contracting costs increase’ (Abbott and Snidal 2000, p. 436). The question of under what conditions soft law transforms into hard law is a long-standing discussion in the literature, which expects processes of legalization when adequate conditions make them possible (Saurugger and Terpan, 2021; Terpan 2015). Other perspectives do not problematize the existence of soft law but perceive it as an interesting option for many situations, in particular in international policy arenas. As Abbott and Snidal (2000, p. 422) indicate, ‘the choice between hard law and soft law is not a binary one’, meaning that many configurations are possible. Relevant actors often decide to promote soft law initiatives as a way of coping with complex policy problems, because they allow political arrangements and actors’ involvement which would not be achievable under other procedures. Many forms of softer legalization, including some mechanisms of enforcement, are easier for the actors involved to accept – compared with agreements on formal laws – which may be more difficult, given the formal delegation of sovereignty they involve. To some extent, regarding the role of soft law, we can find these perceptions in the case of the EU. Not only have views about the dangers and limitations of using soft law instruments emerged, but the proliferation of soft norms has also been understood as a way for the EU to be involved in policy areas in which it has no formal powers, thus benefiting from some critical situations, such as the 2008 Euro crisis (Graziano and Halpern 2016). Finally, others suggest that the combination of soft and hard rules also provides extended capacities for the governance of the EU, both from the point of view of EU institutions and Member States (Ştefan et al. 2019; Saurugger and Terpan 2021). As a point of departure, we define soft law in EU governance as all rules that are not framed in the Community method or are not the result of a legislative procedure established by EU treaties (Terpan 2015). There are many policy areas, particularly since the Maastricht Treaty, in which EU institutions are very active and promote initiatives and programs, and even adopt legal rules, outside the legislative process and without judicial control by the CJEU. Forms of intervention are very diverse, including mechanisms of coordination, benchmarking, agreements on targets and thresholds, developing isomorphic institutions, etc. It is also important to highlight that there is no single type of soft law in the EU, but in all cases, we must exclude from its definitions primary and secondary sources of EU law, including regulations and directives. Indeed, soft law in the EU manifests itself in many different forms of policy instruments and political activity. Soft law is usually defined broadly to cover

210  Research handbook on soft law a large variety of regulatory instruments that have no capacity for legal enforcement. Terpan (2015) identifies two criteria, obligation and enforcement – adapted from Abbott and Snidal (2000) – to assess the soft–hard continuum in EU law. This suggests that many combinations are possible. He argues that when there is no obligation and soft enforcement, a measure can be considered soft law, whereas when there is a hard obligation – with or without enforcement – this is a case of hard law (see also Saurugger and Terpan 2021). Soft regulatory instruments include recommendations, incentives, guidelines, objectives, and targets, as well as imprecise norms included in legal acts, or even legal obligations not controlled by the CJEU (Ştefan et al. 2019, p. 13). What is different about EU soft law when it is compared to soft norms established by other international organizations? In fact, there are no major differences, in the sense that in both cases we find international organizations that operate under international treaties, are involved in intergovernmental relations, and actively develop new forms of policy intervention within their jurisdiction. However, what is indeed distinctive about EU soft law, compared to that produced by other international organizations, is its expansion into multiple policy areas in which the EU has no supranational powers, including the case of economic governance, where the use of soft law has been very intense since the 2008 Euro crisis. As discussed above, the production of EU soft law can be interpreted as an intermediate step towards the increasing legalization of EU governance in multiple areas where the EU has no competence to produce hard law. However, it can also be interpreted as a policymaking development that finds its place in the complex institutional arrangements between Member States and EU institutions. 2.2

A Decisive Step in EU Governance and Soft Law: OMC

To understand why soft law has developed so intensively in EU economic governance, it is necessary to refer to the Open Method of Coordination (OMC), which emerged as a soft law framework to allow EU-wide action in many different policy areas (Borrás and Jacobsson 2004; see also van Gerven and Stiller in this volume). This method has been considered one of the most common procedures to move policies forward when EU treaties do not provide well-defined legal mandates, and we suggest that it is very relevant for further developments that have occurred in EU economic governance. This method avoided defining complex institutional procedures for the approval of joint decisions, and instead, it aimed at providing flexibility for each Member State to make its own decisions, constrained by a EU-wide agenda of policy frameworks and common policy goals. As Borrás and Jacobsson (2004) detail, the OMC brought some significant innovations to the tradition of soft law in the EU. One of the major differences was the political salience of the entire process, compared with the administrative nature of EU soft law which was prominent until that point. Also, the intergovernmental and participatory framework in which the OMC operated displayed a relevant difference compared with the earlier tradition of supranational bureaucratic governance. As the European Council was the visible locus of the OMC, and the direct involvement of national politics in the related discussions was very visible and relevant, the nature of how soft law was used by the EU – and linked more intensively to the different levels of government in Europe – was transformed. All these mechanisms were adopted further by the EU economic governance methods that emerged after the 2008 Euro crisis. An additional element was the pre-eminence of peer pressure as a sanction mechanism. This was often reinforced through the use of negative publicity focused on those Member States that

EU economic governance, agencies and soft law  211 did not fulfil the established policy goals and was often highlighted by means of individual recommendations – a very visible aspect also in recent economic governance. Transforming the nature of soft law, as previously used by EU institutions, the OMC prepared the ground for the introduction of governance mechanisms that were to become prominent in the economic and financial areas. 2.3

A Major Role for Soft Law: EU Agencies

Beyond the consolidation of procedures and mechanisms for EU governance at the sector level during the first decades of the new century, there is an institutional dimension to the expansion of soft law in the EU, namely its adoption by EU agencies, including those related to the expansion of EU economic governance. Soft law is also often used by the ECB. This institution can be set apart from EU agencies, as it is included in the treaties, and its independence is clearly stated. The ECB, given the particular characteristics it entails, no doubt represents a relevant step in delegating authority to European-wide autonomous institutions, constituting a cornerstone of an emerging EU economic governance toolbox. EU agencies have specialized tasks that contribute to the development of EU policies and support coordination between the EU and national governments (see also Vaughan in this volume). They combine technical expertise, establish common interpretative criteria, and develop policy assessments and recommendations. The European agencification process can be interpreted as a part of the complex dynamic of establishing a European executive power, in which the Commission delegates non-core activities to semi-autonomous agencies (Busuioc 2013). EU agencies are financed to perform specific tasks by contributions from the EU general budget and by revenues from the Commission, and in some cases, they have their own sources of funding (ECA, 2020). Most EU agencies were designed and launched during the 2000s. They were related to the areas of the single market that required active regulation and intense coordination between regulatory agencies and Member State governments. They also contributed to the implementation of new EU policies in highly complex areas with important and variable risks and were needed to pool multiple technical and scientific resources, being an important source of soft law developments. In the economic governance area, there are agencies that support the internal market policies and are responsible for registrations, certifications and authorizations at EU level, such as European Chemicals Agency (ECHA), European Food Safety Agency (EFSA), European Union Intellectual Property Office (EUIPO). Other agencies are involved in rule-making and have supervisory responsibilities in finance activities. These include the European Banking Authority (EBA), the European Securities and Market Authority (ESMA), and the European Insurance and Occupational Pensions Authority (EIOPA). In most cases, EU agencies do not possess powers to formally adopt binding measures. The reason is partly related to the Meroni doctrine, although other factors most probably have a role to play in this particular configuration. This doctrine, established by the CJEU (Case 9-56 Meroni v High Authority), introduced certain limits to delegating tasks to agencies and private entities. The ruling dates back to 1958, and it prevented EU institutions from conferring regulatory responsibilities on bodies that were not defined in EU treaties. The Meroni doctrine was further elaborated by the CJEU in the Romano case (C-98/80 Romano), which excluded any administrative body from adopting formally binding decisions if this power was not defined in EU treaties. It was only possible for them to adopt non-binding decisions,

212  Research handbook on soft law such as recommendations and technical reports. In this sense, such constraints also push for stronger use of soft law mechanisms as a way to overcome the Meroni doctrine. More recently, the CJEU reaffirmed the Meroni doctrine in the ESMA case (C-270/12 United Kingdom v Parliament and Council). Based on this ruling, and building on the Meroni and Romano rulings, only well-defined regulatory tasks can be delegated to EU agencies. This includes, however, the power to prepare normative proposals for consideration by other EU institutions – in particular by the European Commission – and to draw up guidelines or technical documents, used to set criteria and to steer policy developments in the areas in which they are involved (see also Vaughan in this volume). 2.4

A Multiple Accountability Framework for Soft Law in EU Economic Governance

A relevant problem for the use of soft law in democratic societies, which appears to be aggravated in the case of the EU, is that the most active promoters of soft law are not politicians in parliaments and executive offices, but bureaucrats, officials, and experts in a variety of administrative spaces. This is particularly the case for specialized agencies that are formally separate from political institutions, such as EU agencies. Here, the perspective of liberal democracy would demand the introduction of transparency and accountability systems to these entities, to make their soft law activity more visible to citizens and subject to public scrutiny (Mörth 2004, see also Mörth in this volume). Also, from the agencies’ point of view, the need to enhance the legitimization of their policy activities and the implementation of multiple accountability mechanisms, even beyond legal mandates, emerges as a very relevant reason for the purpose of increasing transparency and accountability. Agencies’ strategic accountability practices can be considered under different circumstances and asymmetrical power relations (Apaydin and Jordana 2020). In the case of EU agencies making intensive use of soft law instruments, combined with an increase in their numbers and the resources available, it is critical for them to have the capacity to voluntarily establish additional accountability mechanisms (Koop 2014). Active accountability practices allow agencies to be more answerable to multiple stakeholders and to the constituencies that may interact with them (Busuioc 2013). EU agencies show different dimensions of accountability. On the one hand, there are mechanisms of political accountability to different EU institutions and Member States. On the other hand, they are socially accountable to those stakeholders and citizens who interact with them. Finally, EU agencies are legally accountable, in particular to the CJEU. Considering the political accountability mechanisms in relation to EU agencies, during the 2000s, both the Commission and Parliament emphasized the importance of increasing the accountability of EU agencies to keep them under democratic control (Busuioc 2013). Several EU institutions are involved in the supervision of the EU agencies, including the European Parliament (EP), the Council, the Commission, the Court of Auditors (ECA), and the European Ombudsman, and various mechanisms were established to keep EU agencies politically accountable. The role of the EP is probably the most relevant from the point of view of democratic accountability, and it has been reinforced thanks to legislative amendments that enhance the parliamentary oversight of EU agencies. Parliamentary committees are the main mechanisms for supervising EU agencies, but their attention and interest are very varied (Busuioc 2013). To keep EU agencies accountable, the role of written parliamentary questions is also relevant.

EU economic governance, agencies and soft law  213 However, empirical analysis of the level of scrutiny by parliamentarians show that the questions are very selective and are concentrated on only a small number of EU agencies (Font and Perez 2016). The Commission and Member States keep EU agencies politically accountable, insofar as they are each represented by their management boards and stay in direct contact with EU agencies to monitor their tasks and activities. However, this means that the Commission and the Member States’ representatives are directly embedded in the EU agencies, and for this reason, they participate in the decision-making, not being able to exert a most distant, interest-free, control. When focusing on social accountability, we should examine the mechanisms established by EU agencies to be accountable to their stakeholders, while also being directly accountable to citizens in general. Here, we find that formal mechanisms, in particular those related to increasing transparency – i.e., making available information about internal decision-making processes – are relatively well developed. However, this is only one component of the social accountability processes. Actually, different mechanisms to allow the participation of non-state stakeholders are frequent in EU agencies. These include public consultations, stakeholder bodies, and representation on management boards (Arras and Braun 2017). Recent analysis of how these mechanisms work in practice show that their design and adjustment remain largely shaped by EU agencies’ priorities and strategies, and not by external political actors. In many agencies, there is a tendency towards facilitating a wider involvement of stakeholders, despite the fact that this involves greater politicization of the agencies and a less technocratic profile (Busuioc and Jevnaker 2020). From the point of view of legal accountability, the case of EU agencies and their use of soft law raises two related difficulties. On the one hand, the control of the EU agencies’ actions has always been debated, especially because, until the Lisbon Treaty, the agencies were not formally recognized in the EU treaties (Chamon 2011). Despite this lack of recognition, the CJEU had understood, even before the Lisbon Treaty, that EU agencies’ actions were reviewable (Busuoic 2010). On the other hand, the CJEU has always recognized its jurisdiction, especially in the context of the ‘action for annulment’ (Article 263 TFEU) over legally binding acts, and therefore, soft law in principle has been excluded from the CJEU’s jurisdiction. However, as we will see in Section 3 below, thanks to the preliminary reference proceedings (Article 267 TFEU), the CJEU has been able to review soft law, especially in the context of the ESFS. Moreover, in the context of the EMU, both the CJEU and the national courts have been involved in controlling soft law, although not always without difficulties. It seems clear that the CJEU and the national courts have been creative in developing legal accountability mechanisms for soft law. Moving on to discussing the position of the ECB, it should be noted that the ECB’s political accountability is particularly constrained, as it enjoys very strong independence regarding its regulatory and policy decisions (de Haan 1997). Actually, the ECB’s national representatives are appointed by the national central banks, not by Member State governments. Here, European and national parliaments, as well as Member State governments, can only establish a distant control, and their capabilities to make the ECB fully politically accountable are not very strong. Social accountability of the ECB has also followed a weak pattern, in particular during the years after the 2008 Euro crisis, as the ECB did not establish significant accountability and transparency channels to explain and justify its activity to stakeholders and to citizens in

214  Research handbook on soft law general (Curtin 2017). Thereafter, several transparency accountability channels were open, although not very intensively. Finally, the legal accountability mechanisms of the ECB point to the CJEU as the main body in charge of legal disputes brought against the banks. The ECB, and its subjection to the CJEU, is particularly entrenched at treaty level where the EU treaties expressly mention that the ECB’s binding acts are subject to legal review by the CJEU (Article 263 TFEU). In line with the focus of this chapter on the mechanisms of legal accountability, it should already be noted that the two cases analyzed later in this chapter show some of the difficulties surrounding the legal control of soft law measures. The intense use of soft law in the context of the EMU, also by the ECB, has hindered the legal review of the CJEU and has even led to a conflict with national judiciaries. To the contrary, the ESFS and the reduced role of soft law facilitated – as also in the case of the ECB – the control of the CJEU. As described below, the CJEU has developed an intense judicial review of the ECB’s actions under the context of the ESFS, and this control has been carried out without conflict with the national judiciaries. The use of binding acts and, therefore, the reduction of soft law, bolsters legal accountability also in the case of the ECB action. Instead, in a highly political context such as the EMU, in which soft law, intergovernmentalism and the agencification process are less developed, political and social accountability seem to be more appropriate. Legal accountability, as the EMU case will show, is less suitable for this scenario.

3.

SOFT LAW IN THE EMU AND THE ESFS

3.1

EMU: A Case for Political Oversight

As a result of a combination of the Community method and intergovernmentalism, the structure of the EMU is unique in the history of monetary unions (Issing 2002, p. 345). Indeed, the fundamentals of the EMU are a combination of primary and secondary EU law on the one hand (see Protocol 12 to the TFEU on Excessive Deficit, and particularly the package of directives and regulations approved under the so-called ‘six-pack’ in 2011 and the ‘two-pack’ in 2013) and classic international treaties on the other hand (Treaty on Stability, Coordination and Governance in the Economic and Monetary Union of 2012 (TSCG) and the Treaty Establishing the European Stability Mechanism of 2012 (ESM Treaty)). Against this backdrop, the resulting framework – EU economic governance – displays three characteristics: the intense use of soft law, intergovernmentalism, and the lack of the EU agencification process. These characteristics make legal oversight difficult but bolster political accountability. Member States, whether on their own or in the Council, and the Commission are the protagonists of the EU economic governance framework, but the results of the designated procedures are ambiguous from the perspective of legal nature and the capability of the CJEU for overseeing them. Soft law is a common feature in all the supervisory procedures of the EMU (Hodson and Maher 2004, pp. 800–802). The EMU relies on control mechanisms of the deficit and public debt of Member States which, in accordance with EU sources of law, are not legally binding per se. When an Excessive Deficit Procedure (EDP) is launched against a Member State, the Commission and the Council issue ‘recommendations’, and only final sanctions are adopted in the form of a ‘decision’ (Article 126 TFEU). Under the European Semester, the Commission

EU economic governance, agencies and soft law  215 and the Council also issue ‘specific recommendations’ following up with National Reform Programmes (NRP) and Stability or Convergence Programmes sent by the Member States (see Regulation (EU) 1175/2011). When the Macroeconomic Imbalance Procedure (MIP) and the Excessive Imbalance Procedure (EIP) are launched against a Member State, the Commission and the Council issue ‘alert mechanism reports’ and ‘recommendations’ (see Regulation (EU) 1176/2011). Even the Enhanced Surveillance mechanism relies on ‘macroeconomic adjustments programs’ and ‘recommendations’ (see Regulation (EU) 472/2013). This language of plans and recommendations refers to soft law, in the sense that EU institutions do not adopt decisions through the regular EU legislative process, and these decisions do not have legally binding force. These soft law measures adopted within the EMU framework cannot be directly controlled by the CJEU through the action for annulment (Article 263 TFEU), as the non-binding nature of most of the decisions taken under the EMU framework impinges on the direct intervention of the CJEU. At best, the CJEU is able to analyze the legally binding nature of a decision on a case-by-case basis to declare itself competent under the framework of the action for annulment. That was the case in the debate regarding the conclusions of the Council adopted on 25 November 2003 about holding the EDP in abeyance for France and Germany (C-27/04 Commission v Council). The CJEU declared that the conclusion of the Council did have legal effects because in practice it held the ongoing EDP in abeyance, and, therefore, it modified the recommendations previously adopted by the Council (paras 44–51). This goes to show that the ambiguity of EMU’s soft law, which is present in all of the supervisory mechanisms, brings into question the competence of the CJEU and the availability of the action for annulment. The difficulties of the legal oversight of EMU’s soft law are also present in the framework of the preliminary reference proceedings (Article 267 TFEU). The cooperation between the national courts and the CJEU could be an alternative to the limitations of the action for annulment for controlling EMU’s soft law. Despite the potential of this multilevel mechanism, it has been scarcely used in the EMU’s framework, and because of the ambiguity of soft law in that context, judicial dialogue has been difficult and has even led to a crisis between the EU and some national judicial levels, as was the case in Germany. The crisis between the CJEU and the Federal Constitutional Court of Germany (BVerfG) in Gauweiler (C-62/14 Gauweiler and Others) started with a review of a press release of the ECB, which announced its intent to purchase government bonds on secondary markets as a measure to mitigate the Euro crisis (the so-called OMT program) (paras 18–31). In Gauweiler, the core of the discussion between the two courts was the scope and boundaries of the competence of the ECB and the EU in the field of monetary policy, that is, whether the OMT program was a momentary decision or part of an economic policy beyond the ECB’s treaty mandate (paras 46–65). The crisis deepened in Weiss (C-493/17 Weiss and Others), but in this case the CJEU was called on to review a ‘decision’ of the ECB which launched the public sector purchase program (the so-called PSPP program) – a decision which was a regular EU source of law – without any doubt regarding its legally binding nature. The answers of the CJEU both in Gauweiler and Weiss, arguing that all the decisions of the ECB were in accordance with EU law, were not convincing for the BVerfG which finally declared the decisions of the ECB under the PSPP program unconstitutional and inapplicable in Germany because of the ECB acting ultra vires (BVerfG, 2 BvR 859/15, Judgment of the Second Senate of 5 May 2020). This led to an unprecedented crisis in which, for the first time, the BVerfG decided not to

216  Research handbook on soft law comply with a judgment of the CJEU and questioned the principle of primacy of EU Law (Bobic and Dawson 2020, pp. 1982–89). The reactions of the CJEU (2020) and the President of the European Commission (2020) emphasized the need to preserve the primacy of EU law which shows the seriousness of the crisis. At the end, it was a press release in Gauweiler that triggered a contest between the two levels in which the competence of the ECB and the proportionality of its measures during the Euro crisis were tested (Solanes Mullor 2019b, pp. 155–171). The unprecedented legal review of measures not yet implemented, and only announced through a press release, triggered a case in which the powers of the ECB were subject to control. The clash between the BVerfG and the CJEU emerged in a context in which the measures were taken through an ambiguous and unclear form of soft law, making judicial dialogue more difficult because of the uncertainty over the nature of the soft law at stake. In that regard, the CJEU was forced to discuss the admissibility of the preliminary reference because of the alleged artificiality of the case, that is, whether the CJEU has a sort of preventive jurisdiction over measures only announced in a press release and not yet implemented (C-62/14 Gauweiler, paras 11–31). Beyond the use of soft law, intergovernmentalism also further exacerbates challenges of legal accountability in the context of the EMU, whose structure partially relies on classic international treaties (Puetter 2012, pp. 166–168). Despite the intention of the Commission to fully unionize the EMU’s framework, its attempts until now have not been successful (EU Commission 2017). The fact that the EMU’s normative framework is partially outside EU law hinders the oversight of the CJEU. The CJEU has recognized that EU institutions are relevant actors in framing and applying the Memorandums of Understanding (MoU), a key tool for addressing the Euro crisis, but because of the intergovernmental nature of MoUs, the CJEU is only prepared to declare non-contractual responsibility for the potential damages caused by actions of the EU institutions (Article 268 TFEU). In Ledra Advertising, the CJEU reviewed the MoU between Cyprus and the ESM and analyzed whether the Commission and the ECB, when subscribing to it, had engaged in an analysis of the compatibility of the MoU with EU law. In this case, the examination by the Commission and the ECB in signing the MoU should have included the issue of whether taking over the insured deposits of its clients by the Bank of Cyprus was compatible with the right to property recognized in Article 17(1) of the Charter of Fundamental Rights of the EU (C-8/15 P, Ledra Advertising, paras 62–76). The CJEU recognized its inability to directly control the compatibility of MoUs with EU law. It was only prepared to declare compensation for damages in cases where EU institutions were not engaged in that analysis. Later, in Chrysostomides (C-597/18 P Council v K. Chrysostomides & Co and Others), the CJEU discussed the non-contractual responsibility of the Eurogroup. In the context of its role in supporting the MoUs with Cyprus, and its intergovernmental character outside EU law, the CJEU disqualified the Eurogroup as a body subject to non-contractual responsibility. At the national level, a limited number of courts have been actively reviewing the intergovernmental measures taken within the EMU, especially MoUs with different outcomes (González Pascual 2017, pp. 121–125). On the one hand, the Portuguese Constitutional Court has considered itself as a mechanism of control and has decided to review the Portuguese MoU against national constitutional law. It has declared unconstitutional some measures taken by the Portuguese government following the MoU (Portuguese Constitutional Judgment no. 396/2011, para. 9; Judgement no. 353/2012, para. 6; judgement no. 575/2014, para. 25). On the other hand, the Spanish Constitutional Court has not questioned the MoU subscribed by

EU economic governance, agencies and soft law  217 Spain and has even used it as a parameter in favour of the austerity measures taken by the Spanish Government (Spanish Constitutional Orders no. 69/2014 and no. 115/2014; see also the Spanish Constitutional Judgement no. 93/2015). In both cases, national courts, because of the intergovernmental character of MoUs, and MoUs not being part of EU law, have not engaged in judicial dialogue with the CJEU by sending a preliminary question. The diminished role of courts in the EMU emphasizes political accountability in the EMU’s framework (Dawson and Maricut-Akbik 2021, pp. 1716–1719). This strengthened political nature of the EMU’s framework also relegates agencies to the background. Indeed, only the ECB has a role in the overall framework of the EMU. Clearly, the EU agencification process, which is present in many fields in the EU (Solanes Mullor 2018, pp. 86–88), has not been seen as the proper tool in the context of the EMU. In this way, the EMU is seen as a political dialogue between the Commission, the Council, and the Member States, in which both the CJEU and the EU agencies have secondary roles. 3.2

The ESFS: A Case for Legal Accountability

The ESFS has a strong presence in the EU agencification process (Schammo 2012, pp.  771–774; see also Türk in this volume). In the context of the ESFS, especially following the Euro crisis of 2008, Member States and EU institutions have intensified the Europeanization of banking supervision and have decided to create a supervisory framework which relies on EU agencies (Schemmel 2016, pp. 455–459). On the one hand, the ECB was reinforced, thanks to its new supervisory powers under the umbrella of both the Single Supervisory Mechanism (SSM, see Council Regulation (EU) 1024/2013) and the Single Resolution Mechanism (SRM, see Regulation (EU) 806/2014), which introduced a new agency: the Single Resolution Board (SRB). On the other hand, three EU agencies were created, commonly called the European Supervisory Authorities (ESA): the European Banking Agency (EBA, see Regulation (EU) 1093/2010), the European Insurance and Occupational Pension Authority (EIOPA, see Regulation (EU) 1094/2010), and the European Securities and Market Authority (ESMA, see Regulation (EU) 1095/2010). The ESFS legal framework gives room to the adoption of hard law rather than soft law measures. In particular, the role of the ECB in the ESFS minimizes soft law. The legally binding nature of most of the ‘decisions’ of the ECB, which are part of the EU’s regular sources of law, facilitates the legal oversight of the CJEU, which is able to control the ECB through the action for annulment and preliminary reference proceedings. In fact, the CJEU has been active in this field and has exercised control over the ECB’s decisions. For instance, in the Landerskreditbank case (T-122/15 Landeskreditbank Baden-Württemberg – Förderbank), the CJEU scrutinized the ECB’s decision taken in the SSM framework, in which a national bank was classified, on account of its size, to be subject to the sole supervision of the ECB instead of being subject to shared supervision. The CJEU upheld the decision of the ECB, but it applied strict judicial oversight of the reasoning of the ECB’s decision (paras 113–136). Another example of this active role is the Berlusconi case (C-219/17 Berlusconi and Fininvest), which also involved the SSM framework. The CJEU claimed sole jurisdiction of the oversight of the ECB’s decisions on authorization to acquire a qualifying holding in a credit institution. The SSM mechanism provides for a mechanism of collaboration between national banking authorities, which have an advisory role, and the ECB, which has the decision-making power. The CJEU, when answering the questions raised by the Italian

218  Research handbook on soft law courts, argued that national courts have no competence to review the national phase of the proceedings which are also to be controlled by the CJEU (paras 40–59). These two cases show the dynamism of the CJEU, which perceives that it can supervise the actions of the ECB and national authorities in the field of the ESFS. Soft law is still relevant in the ESFS framework. The design of the ESAs responds to a common trend of the EU agencification process which has in most instances delivered a powerless model of agency (Solanes Mullor 2018, pp. 92–97). EU institutions and scholars have provided different criteria for assessing EU agencies, although a consensus about their nature and classification has not been reached (Solanes Mullor 2016, pp. 152–160). However, ESAs seem to respond to the ‘quasi-regulatory’ type of agencies (a category introduced in Busuioc 2013, pp. 37–42). This is indeed the case, because ESAs may develop draft regulatory technical standards which will ultimately be adopted by the Commission when it has been granted delegated powers pursuant to Article 290 TFEU. Furthermore, the ESAs can approve guidelines and recommendations (Articles 10 and 16 of EBA, EIOPA and ESMA regulations). Thus, hard normative power is in the hands of the Commission, whereas the agencies are only capable of adopting soft law in the form of guidelines and recommendations (on ESAs’ soft lawmaking powers, see also Türk in this volume). This institutional framework of the ESFS, which relies on a long-standing development of the EU agencification process, facilitates the legal oversight of the CJEU. The CJEU is used to dealing with the quasi-regulatory powers of EU agencies, particularly with the power of drafting opinions on which the Commission bases its decisions. From the very beginning, the CJEU has felt prepared to control these preliminary acts indirectly by reviewing the final legally binding act of the Commission (Busuioc 2010, pp. 173–174). In the end, the normative proposals and drafts of EU agencies are part, internally, of the formation of hard law, which is clearly under the jurisdiction of the CJEU. Despite the reduced role of soft law, there still remains the issue of its legal accountability in the context of ESFS. In the EBA’s Guidelines case (C-911/19 FBF), the CJEU was confronted with the question of the legal nature of the guidelines issued by the EBA. The answer was clear. On the one hand, EBA guidelines are soft law, in the sense that they do not have binding legal effects, and therefore they are not reviewable through the action for annulment of Article 263 TFEU (paras 35–51). On the other hand, the CJEU expressly declared that it is prepared to review the validity of the guidelines through the preliminary reference procedure of Article 267 TFEU (paras 52–65). In this case, the CJEU reviewed the criteria set out in the EBA’s regulation for issuing guidelines and concluded that they were approved in accordance with those criteria (paras 66–132). The CJEU furthermore specifically stated that the fact that the guidelines do not produce binding legal effects does not affect the scope of the review (para. 55). This attitude of the CJEU – and its willingness to control the validity of soft law through the preliminary reference procedure – has emerged more intensively in a context in the ESFS in which supranationalism is more intense than in EU economic governance. In EU economic governance, intergovernmentalism hinders legal accountability, whereas the intense use of soft law by political actors with a high degree of discretion limits the intervention of the CJEU even with the possibility of its control through the preliminary reference procedure. In this kind of procedure, the CJEU needs the help of national courts, which have mostly been inactive in the field of EU economic governance. Even when they have been active, the unclear nature of soft law has resulted in open confrontation rather than in calm judicial dialogue

EU economic governance, agencies and soft law  219 between the EU and the national level. In contrast, supranationalism in the context of ESFS, especially the strong presence of the EU agencification process and the circumscribed use of soft law in the hands of EU agencies, has opened the door to the intervention of the CJEU, which has perceived itself as a pivotal supervisory body.

4. CONCLUSION Soft law, understood as a non-binding measure and as an instrument that is outside the procedures foreseen for the regular EU sources of law, is a widespread phenomenon in EU governance. Soft law has been used in multiple areas in which the EU has limited formal powers. As a consequence, soft law has deepened European integration in fields in which hard law has not naturally facilitated supranationalism. The intense use of soft law in EU economic governance is only one example of the positive implications that soft law has for the advancement of European integration in nationally sensitive policy fields. The EU agencification process and the presence of the ECB have a role to play in that evolution. For EU agencies and the ECB, soft law has been a useful resource in domains in which EU institutions have lacked strong powers. The Euro crisis of 2008 has generated valuable insights about soft law, EU governance, and the EU agencification process. In the EMU, soft law has been used intensely and has been accompanied by intergovernmentalism and the lack of EU agencies. This combination has hindered legal accountability, as the CJEU and the national courts have experienced difficulties in their capacity to control soft law. The EMU has relied instead on political oversight. In contrast, the ESFS case shows a different picture. The strong presence of EU agencies and the reduced role of intergovernmentalism facilitate the intervention of courts. Soft law is still relevant, but the CJEU has shown that it is more comfortable reviewing it because it is more familiar with soft law instruments used by EU agencies than with the newer and more diffuse forms of soft law within the EMU’s structure. In the EMU’s framework, thus, legal accountability is troublesome. European integration here adopts a shape with which courts are not at ease. Intergovernmentalism and its widespread presence in the EMU’s framework exacerbates the difficulties for the CJEU, as it keeps decisions outside EU law and outside the CJEU’s jurisdiction. The efforts of the CJEU to control the EMU’s decisions have been very limited: the treaty framework does not facilitate the review of EMU’s soft law measures through the action for annulment, and the use of preliminary reference has led to struggles with national judiciaries and the non-contractual responsibility for controlling MoUs have scarcely produced results. The dispute between the CJEU and the BVerfG in Gauweiler and Weiss shows a difficult judicial dialogue in which the lack of clarity of the nature of soft law at stake makes interaction between the two courts difficult. In the ESFS’s framework, to the contrary, legal accountability has been easier to implement. The combination of several factors – less intense use of soft law, less intergovernmentalism, and a much more secured presence of EU agencies and the ECB – have bolstered the action of the CJEU. The court has reviewed the binding acts of the ECB in the context of the ESFS, and it has declared itself competent to control the ESAs’ soft law through the preliminary reference proceeding. The CJEU has adopted this active role without conflicts at the national level. The

220  Research handbook on soft law case of ESFS shows that the EU agencification process is conducive to legal accountability because the CJEU is familiar with the control of EU agencies and their use of soft law. The comparison between the EMU and ESFS shows the need for further clarification of the accountability framework of soft law, in which the proper role of all relevant actors – political bodies and the courts – should be reassessed. The study of the EMU and the ESFS, and the role of soft law and the EU agencies in these contexts, can add new insights for the analysis of accountability, especially legal accountability, within EU governance. The above comparison of the EMU and ESFS cases reveals the shortcomings of addressing the accountability challenge of soft law from only one perspective. Legal accountability mechanisms appear less suitable in the context of the EMU than in the context of the ESFS. Thus, the proper tools for making soft law accountable will depend on a detailed and in-depth analysis of the institutional framework in which soft law operates and, depending on that framework, political and social accountability can replace legal accountability.

REFERENCES Abbott, K. and Snidal, D., (2000). Hard and Soft law in international governance. International Organization. 54(3), 421–456. Apaydin, F. and Jordana, J., (2020). Varying Power Configurations and the Accountability of Independent Regulatory Agencies. International Review of Public Policy. 2(3), 342–357. Arras, S. and Braun, C., (2018). Stakeholders wanted! Why and how European Union agencies involve non-state stakeholders. Journal of European Public Policy. 25(9), 1257–1275. Bieber, F. and Bieber, R., (2020). Negotiating Unity and Diversity in the European Union. Gewerbestrasse: Palgrave Macmillan. Bobic, A. and Dawson, M., (2020). Making sense of the ‘incomprehensible’: The PSPP Judgment of the German Federal Constitutional Court. Common Market Law Review. 57(6), 1953–1998. Borrás, S. and Jacobsson, K., (2004). The open method of co-cordination and new governance patterns in the EU. Journal of European Public Policy. 11(2), 185–208. Bovens, M., (2007). New Forms of Accountability and EU-Governance. Comparative European Politics. 5(1), 104–120. Busuioc, M., (2010). The Accountability of European Agencies. Legal Provisions and Ongoing Practices. Delft: Eburon. Busuioc, M., (2013). European Agencies. Law and Practices of Accountability. Oxford: Oxford University Press. Busuioc, M. and Jevnaker, T., (2020). EU agencies’ stakeholder bodies: vehicles of enhanced control, legitimacy or bias?, Journal of European Public Policy. 29(2), 155–175. Case 9-56 Meroni & Co., Industrie Metallurgiche, SpA, ECLI:EU:C:1958:7. Case 98/80 Giuseppe Romano v Institut national d’assurance maladie-invalidité, ECLI:EU:C1981:104. Case C-27/04 Commission of the European Communities v the Council of the European Union, ECLI:EU:C:2004:436. Case C-270/12 United Kingdom v European Parliament and Council, ECLI:EU:C:2014:18. Case C-62/14 Gauweiler and Others v Deutscher Bundestag, ECLI:EU:C:2015:400. Case C-8/15 P Ledra Advertising v Commission and ECB, ECLI:EU:C:2016:701. Case T-122/15 Landeskreditbank Baden-Württemberg – Förderbank v European Central Bank, ECLI: EU:T:2017:337. Case C-219/17 Silvio Berlusconi and Finanziaria d’investimento Fininvest SpA (Fininvest) v Banca d’Italia and Istituto per la Vigilanza Sulle Assicurazioni (IVASS), ECLI:EU:C:2018:1023. Case C-493/17 Weiss and Others, ECLI: EU:C:2018:1000. Case C-597/18 P Council v K. Chrysostomides & Co. and Others, ECLI:EU:C:2020:1028. Case C-911/19 Fédération bancaire française (FBF) v Autorité de contrôle prudentiel et de résolution (ACPR), ECLI:EU:C:2021:599.

EU economic governance, agencies and soft law  221 Chamon, M., (2011). EU agencies between Meroni and Romano or the Devil and the Deep Blue Sea. Common Market Law Review. 48(4), 1055-1075. Council Regulation (EU) No 1024/2013 of 15 October 2013 conferring specific tasks on the European Central Bank concerning policies relating to the prudential supervision of credit institutions (2013). Official Journal. 2013. L287, 63–89. Court of Justice of the European Union., (2020). Press release following the judgment of the German Constitutional Court of 5 May 2020. Press Release No. 58/20. Curtin, D., (2017). ‘Accountable independence’ of the European Central Bank: seeing the logics of transparency. European Law Journal. 23(1-2), 28–44. Dawson, M., (2015). The Legal and Political Accountability Structure of ‘Post-Crisis’ EU Economic Governance. Journal of Common Market Studies. 53(5), 976–993. Dawson, M. and Maricut-Akbik, A., (2021). Procedural vs substantive accountability in EMU governance: between payoffs and trade-offs. Journal of European Public Policy. 28(11), 1707–1726. De Haan, J., (1997). The European Central Bank: independence, accountability and strategy: a review. Public Choice. 93(3), 395–426. European Court of Auditors., (2020). Future of EU agencies – Potential for more flexibility and cooperation, Special report 22, Brussels. European Commission., (2017). Proposal for a Council Directive laying down provisions for strengthening fiscal responsibility and the medium-term budgetary orientation in the Member States. COM(2017) 824 final. European Commission., (2020). Statement by President Von Der Leyen. STATEMENT/20/846. Available from: https://​ec​.europa​.eu/​commission/​presscorner/​detail/​en/​statement​_20​_846 [viewed 16 June 2023]. Font, N. and Ixchel Perez Duran, I., (2016). The European Parliament oversight of EU agencies through written questions. Journal of European Public Policy. 23(9), 1349–1366. German Federal Constitutional Court (BVerfG), 2 BvR 859/15, Judgment of the Second Senate of 5 May 2020. González Pascual, M., (2017). Constitutional Courts before the Euro-crisis law in Portugal and Spain; A comparative perspective. e-Pública. Revista Electrónica de Direito Público. 4(1), 110–127. Graziano, P. R., and Halpern, C., (2016). EU Governance in Times of Crisis: Inclusiveness and Effectiveness beyond the ‘Hard’ and ‘Soft’ Law Divide. Comparative European Politics. 14(1), 1–19. Hodson, D. and Maher, I., (2004). Soft law and sanctions: economic policy co-ordination and reform of the Stability and Growth Pact. Journal of European Public Policy. 11(5), 798–813. Issing, O., (2002). On Macroeconomic Policy Co-ordination in EMU. Journal of Common Market Studies. 40(2), 345–358. Koop, C., (2014). Theorizing and explaining voluntary accountability. Public Administration. 92(3), 565–581. Markakis, M., (2020). Accountability in the Economic and Monetary Union. Foundations, Policy, and Governance. Oxford: Oxford University Press. Mörth, U., (2004). Soft Law in Governance and Regulation: An Interdisciplinary Analysis. Cheltenham: Edward Elgar Publishing. Peters, A., (2011). Soft law as a New Mode of Governance. In: U. Diedrichs, W. Reiners and W. Wessels eds. The Dynamics of Change in EU Governance. Edward Elgar Publishing. Portuguese Constitutional Court, Ruling No. 396/2011, 21 September 2011. Portuguese Constitutional Court, Ruling No. 353/12, 5 July 2012. Portuguese Constitutional Court, Ruling No. 575/2014. 14 August 2014. Puetter, U., (2012). Europe’s deliberative intergovernmentalism: the role of the Council and European Council in EU economic governance. Journal of European Public Policy. 19(2), 161–178. Regulation (EU) 1093/2010 of the European Parliament and of the Council of 24 November 2010 establishing a European Supervisory Authority (European Banking Authority), amending Decision No 716/2009/EC and repealing Commission Decision 2009/78/EC (2010). Official Journal. 2010. L331, 80–115. Regulation (EU) 1094/2010 of the European Parliament and of the Council of 24 November 2010 establishing a European Supervisory Authority (European Insurance and Occupational Pensions

222  Research handbook on soft law Authority), amending Decision No 716/2009/EC and repealing Commission Decision 2009/79/EC (2010). Official Journal. 2010. L331, 48–83. Regulation (EU) 1095/2010 of the European Parliament and of the Council of 24 November 2010 establishing a European Supervisory Authority (European Securities and Markets Authority), amending Decision No 716/2009/EC and repealing Commission Decision 2009/77/EC (2010). Official Journal. 2010. L331, 84–119. Regulation (EU) 1175/2011 of the European Parliament and of the Council of 16 November 2011 amending Council Regulation (EC) No 1466/97 on the strengthening of the surveillance of budgetary positions and the surveillance and coordination of economic policies (2011). Official Journal. 2011. L306, 221–233. Regulation (EU) 1176/2011 of the European Parliament and of the Council of 16 November 2011 on the prevention and correction of macroeconomic imbalances (2011). Official Journal. 2011. L306, 25–32. Regulation (EU) 472/2013 of the European Parliament and of the Council of 21 May 2013 on the strengthening of economic and budgetary surveillance of Member States in the euro area experiencing or threatened with serious difficulties with respect to their financial stability (2013). Official Journal. 2013. L140, 1–10. Regulation (EU) No 806/2014 of the European Parliament and of the Council of 15 July 2014 establishing uniform rules and a uniform procedure for the resolution of credit institutions and certain investment firms in the framework of a Single Resolution Mechanism and a Single Resolution Fund and amending Regulation (EU) No 1093/2010 (2014). Official Journal. 2014. L225, 1–90. Saurugger, S. and Terpan, F., (2021). Normative Transformations in the European Union: On Hardening and Softening Law. West European Politics. 44(1), 1–20. Schakel, A., Hooghe, L., and Marks, G., (2015). Multilevel Governance and the State. In S. Leibfried, E. Huber, M. Lange and J. Levy eds. The Oxford Handbook of Transformations of the State. Oxford: Oxford University Press. Schammo, P., (2012). EU Day-to-Day Supervision or Intervention-based Supervision: Which Way Forward for the European System of Financial Supervision? Oxford Journal of Legal Studies. 32(4), 771–797. Schemmel, J., (2016). The ESA Guidelines: Soft Law and Subjectivity in the European Financial Marker-Capturing the Administrative Influence. Indiana Journal of Global Studies. 23(2), 455–503. Solanes Mullor, J., (2016). Administraciones independientes y Estado regulador. El impacto de la Unión Europea en el Derecho Público español. Madrid: Congreso de los Diputados. Solanes Mullor, J., (2018). Institutional Balance, EU and National Agencification Processes: the need for dialogue. Revista Española de Derecho Europeo. 68, 85–111. Solanes Mullor, J., (2019a). El derecho de la Unión Europea: una revolución jurídica en permanente construcción. In: F. Guirao and J. Pich eds. ¿Una Unión Europea en crisis? Madrid: Catarata. pp. 49–71. Solanes Mullor, J., (2019b). What Goes Up Must Go Up: Raising Judicial Scrutiny over the European Central Bank through Judicial Dialogue. In: M. Belov ed. Judicial Dialogue. The Hague: Eleven International Publishing. pp. 155–171. Spanish Constitutional Court, Order 69/2014, 10 March 2014. Spanish Constitutional Court, Order 115/2014, 8 April 2014. Spanish Constitutional Court, Judgement 93/2015, 14 May 2015. Ştefan, O. A., Avbelj, M., Eliantonio, M., Hartlapp, M., Korkea-aho, E. and Rubio, N., (2019). EU Soft Law in the EU Legal Order: A Literature Review. Available from: dx​.doi​.org/​10​.2139/​ssrn​.3346629. Terpan, F., (2015). Soft Law in the European Union - The Changing Nature of EU Law. European Law Journal. 21(1), 68–96. Treaty Establishing the European Stability Mechanism., (2012). Brussels, signed 2 February 2012, in force 27 September 2012. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​?uri​=​ CELEX​%3A42012A0202​%2801​%29 [viewed 16 June 2023]. Treaty on Stability, Coordination and Governance in the Economic and Monetary Union., (2012). Brussels, signed 2 May 2012, in force 1 January 2013. Available from: https://​eur​-lex​.europa​.eu/​legal​ -content/​EN/​TXT/​?uri​=​celex​%3A42012A0302​%2801​%29 [viewed 16 June 2023]. Tömmel, I., (2016). EU Governance of Governance: Political Steering in a Non-Hierarchical Multilevel System. Journal of Contemporary European Research. 12(1), 407–423.

15. Fictions and fuzziness: soft law rule-making among the EU’s decentralized agencies Steven Vaughan

1. INTRODUCTION This guidance is not legally binding. Nevertheless, NRAs [national regulatory authorities] are required to take the utmost account of it.1

In January 2021, the European Chemicals Agency (ECHA) re-issued a Guidance Document on the preparation of an application for authorization (a process by which harmful chemicals can stay on the EU market under certain circumstances). There was nothing particularly noteworthy about that re-issue. Save that the Guidance Document was 139 pages long and one of 40 primary Guidance Documents on REACH (the EU’s flagship chemicals regime, Regulation (EC) No 1272/2008) that ECHA has produced in relation to just one of the eight different pieces of EU law for which it has responsibility. Since ECHA was established in 2007, the agency has engaged in a wealth of soft law rule-making leading to well over one million words of ‘guidance’ on REACH; a volume that is a significant multiplier of the length of the underlying, complex legislation to which that guidance relates. ECHA’s guidance takes various forms (the 40 primary Guidance Documents, plus dozens of other standards, summaries, methods, approaches, guides etc.), does various things, and is addressed to various subjects. Guidance production, in its many differentiated forms, is ‘business as usual’ for ECHA. This is also true of many other of the EU’s decentralized agencies, the focus of this chapter. What follows unfolds in three sections. First, I offer a short account of EU agencies and their soft law rule-making powers and practices. Second, I look in depth at ECHA and REACH as a case study on EU agency soft law rule-making. This deep dive shows, at a granular level, the variation and differentiation (along multiple axes) in EU agency soft law rule-making. Third, I ask a series of questions about what strikes me as the fuzzy nature of EU agency soft law and suggest that we have a number of fictions that we seem to want to hold on to in order to preserve a sense of legal order and legitimacy that is some way from the messy reality of EU agency soft law rule-making practices.

2.

EU AGENCIES AND THEIR SOFT LAW RULE-MAKING POWERS AND PRACTICES

Physically scattered across the Member States, EU agencies (in terms of what they do, how they connect to national regulators and other stakeholders, their role as part of EU administration and so on) are an important part of the EU governance landscape (and have been since 1975; see Egeberg and Trondal 2017); an example of EU ‘functional decentralisation’ (Vos 2018, p. 5). Separate from the EU institutions (the European Parliament, the European Commission 223

224  Research handbook on soft law and so on), these agencies enjoy a ‘certain degree of autonomy from the Commission but [are] not fully insulated from the Commission’s influence’ (Chiti 2013, p. 94). Over time, EU agencies have been increasing in number (29 in 2008, 33 in 2015, over 40 today) and also significance as part of a process often labelled as ‘agencification’ (see Chamon 2010; Busuioc 2013). This is, on one take, surprising given that EU agencies do not feature in the EU treaties (those foundational instruments containing no general legal basis to create agencies) and are, instead, creatures of different pieces of secondary EU legislation, Regulations.2 EU agencies are one way for the EU to grow its regulatory capacity without the transfer of additional authority from Member States to the Commission (Keleman 2002; Howell 2019) and reflect more general trends (seen in the Member States and elsewhere) for ‘new’ approaches to multi-actor and multi-level governance (Dunleavy and Hood 1994; Levi-Faur 2011). EU agency powers are limited both by: (i) the relevant legislative text establishing each agency; and (ii) also by the jurisprudence of the EU courts which, in very broad terms, says that EU agencies have some discretion but otherwise operate within the grant of precisely delineated executive powers from the EU institutions (see the C-9/56 Meroni (1958) and the C-270/12 ESMA Short Selling (2014) cases). As in other places, the EU’s decentralized agencies cover wide swathes of policy, from EU interests in aviation safety to fisheries to labour and so on. This range of subject matter, and the differing reasons for the creation of each agency, means that, as Elizabeth Howell (2019, p. 325) counsels us, ‘caution should be exercised with respect to generalisations’. While what each EU agency actually does varies widely – Ellen Vos’s (2018) recent study showing how EU agencies can be organized and divided through the six types of work in which they engage (enforcement, the provision of information and co-operation, offering services (such as certification), etc.) – there is some sense that those agencies primarily support the smooth implementation of EU policies, ‘thus helping all the [EU] institutions, in particular the Commission, to concentrate on core policy-making tasks’ (Parliament, Council, and Commission 2012, p. 1). EU agencies have influence both within the EU and, as regards EU external relations, as global actors (Coman-Kund 2018), with Dovilė Rimkutė (2015, p. 115) labelling EU agencies ‘functional problem solvers’. A study by David Coen and Mark Thatcher (2005) suggests that the work and activities of these agencies tends to expand, especially in less visible ways, over time. We know that EU agencies vary significantly in terms of size and budget (with some wholly self-financing, and others funded either by the EU or partly by the EU and partly by other sources), and that the autonomy and the powers of EU agencies also vary (Vos 2018). There is not necessarily any such thing as ordinary practice among or in relation to the EU’s agencies. This is despite the publication, in 2012, of a ‘Common Approach’ on EU agencies by the EU Commission, Council, and Parliament. This sets out, among other matters, the role of decentralized agencies in the EU institutional landscape, the preferred structure and governance of those agencies, how EU agencies should be operated and funded, and issues relating to accountability and transparency. Vos’s (2018) recent work makes clear that the Common Approach is not always followed in practice. When it comes to soft law rule-making, the focus of this chapter, there is much less written than on EU agencies per se (Vos (2018), for example, does not even engage with the issue in her November 2018 report for the European Parliament; and others take single agencies as their primary focus). This is perhaps not so surprising, soft law often being the overlooked ‘ugly sister’ of public law scholarship. What ‘state of the nation’ studies that do exist in this

Fictions and fuzziness  225 space have considered the way in which EU agency soft law comes about (and fits in with the EU’s institutional and judicial frameworks; see Senden and van den Brink 2012), have reflected on the ‘procedural architecture’ for EU agency soft law rule-making (Chiti 2013), have looked at the actual soft law rule-making practices of EU agencies (Vaughan 2015a), and have asked important questions about the legitimacy of agency soft law rule-making (Rocca and Eliantonio 2019). Let me say something at this point about language and terminology. Other studies (for example Senden and van den Brink 2012; Chiti 2013; Senden 2013) have made distinctions between what might be thought of as typologies of EU agency instrument, contrasting ‘soft regulatory rule-making’ (involving para-law policy-steering instruments) with ‘soft administrative rule-making’ (involving post-legislative guidance instruments). I am not as convinced that the actual practices of EU agencies, which I will explore below, continue to neatly fit into these rather fixed categories. Instead, I will use the word ‘guidance’ as a blunt shorthand for the many and fuzzy forms of soft law rule-making in which EU agencies engage. This is an imperfect catch-all but, as will become apparent later on, I am also unsure about the usefulness of using ‘soft law’ as an alternative, especially in situations where the norms we are talking about seem pretty hard (and I fully appreciate the irony of making such a claim in a research handbook dedicated to soft law). I should also note here that this chapter is not concerned with extolling either the virtues of EU agency guidance (cheap, fast, flexible, sensitive to context, and so on; see Reinicke and Witte 2000; Armstrong 2011) or its demerits (concerns about legitimacy, accountability, transparency etc.; on which, see Snyder 1994; Klabbers 1998). Instead, I want to paint a picture of what EU agencies do when it comes to guidance-making, both in general and through the exploration of ECHA as a case study. On the former, we know from my own earlier work (Vaughan 2015a) that many (but not all) EU agencies produce guidance, that practices vary quite significantly among those agencies that do produce guidance (some issuing a handful of documents, others hundreds), that EU agency guidance documents range from a few pages in length to those massing over two hundred thousand words, and that EU agency guidance documents are called (as in other contexts) a vast number of different things (over two dozen different titles in total, from ‘Guidelines’ to ‘Notes’ to ‘Common Approaches’). Only a minority of EU agency guidance is produced in all of the EU’s official languages; most is only in English. My earlier work also seems to suggest that the volume of guidance produced by EU agencies is increasing. One possible reason for this is the more frequent use of framework legislation in the EU which requires – given that they are ‘laws in progress’ (Korkea-aho 2013) – post-legislative elaboration and additional rule-making; and another that EU legislation is becoming increasingly complex and so needs interpretation via guidance (Scott 2011). Senden and van den Brink (2012) have suggested that EU agencies engage in both direct and indirect administration through the use of guidance: doing the former when agency guidance influences how agencies make active decisions on specific issues, and doing the latter when agency guidance speaks to and shapes the actions of national regulators, legislatures, and courts. There is variation, nuance, and dynamism to these soft law rule-making practices. Let us look deeper at one aspect of that variation, namely how EU agencies are granted the power to issue guidance (Vaughan 2015a; Rocca and Eliantonio 2019). In the majority of cases,3 the underpinning EU legislation for the agency grants the agency a generic power to produce guidance. In a smaller group of cases,4 the legislation contains a generic power and also sets out an enumerated list of instances in which guidance on a specific topic is mandated.

226  Research handbook on soft law In a single instance, an EU agency (the EU Agency for the Cooperation of Energy Regulators (ACER)) is required to produce guidance on a specific topic but is not otherwise granted any wider guidance-making powers. Finally, there are less obvious situations in which the underpinning legislation does not explicitly grant guidance-making powers, but it would be possible to read in those powers into other language (or aims or obligations) in the legislation. We see this, for example, with the EU Maritime Safety Agency (EMSA) and the European Union Intellectual Property Office (EUIPO). Even where EU legislation is explicit (in either specific or generic terms) about the production of guidance, it is common for little to be said in that legislation about the mechanics of how EU agencies are to go about guidance production. In this context, Edoardo Chiti (2013, p. 101) observes that, ‘the overall tendency is to formalise only a very basic procedural outline’. Penelope Rocca and Mariolina Eliantonio’s study (2019) puts EU agencies into four groups depending on their level of guidance-making proceduralization (considering matters such as access to documents, transparency, rules on consultation and participation, etc.). Of the 20 EU agencies they reviewed, six had a high degree of proceduralization (and thus a high degree of legitimacy) and six had a low degree. Six others were in the ‘so-so’ category, and two in the ‘pretty poor’ category. What is particularly striking from their data is that newer agencies were not necessarily better when it came to placing in the highest proceduralization category (agency age seemingly uncorrelated to degrees of proceduralization). On yet another axis of variation, guidance produced by EU agencies (whatever its length or title) comes in a full range of drafting styles. As I have set out elsewhere (Vaughan 2015a, p. 77): It is perfectly possible to imagine, and to find examples of, heavily prescriptive, very detailed, rigid post-legislative guidance that is comparable to the most complex EU legislative text, and, equally, possible to imagine and to find examples of guidance that is wide, vague, and incapable of producing legal effects.

As Merijn Chamon (2016, p. 29) puts it, ‘some of these instruments may have more legal and practical effects than others’. This takes us into the sometimes-painful space of whether EU agency guidance is binding; painful in that scholarly debates about bindingness are often split between those who take rather fixed positions on hard versus soft law and those (myself included) who are more interested in what Joanne Scott labels (2011, p. 354), the ‘complex normativity’ of modern EU governance. This space can also be painful because the relevant debates do not often lead to very satisfying conclusions or take us all that far. On a practical level, EU agencies themselves appear unsure (at least in public) about the ‘legal’ effects of the guidance they issue. A handful routinely state that their guidance is not legally binding. Others say this sometimes, but not all the time (leading to a query as to whether this is simply the result of an ad hoc approach or something more meaningful; probably the former); and a third group say nothing about the binding nature of their guidance. Recall, from the Introduction to this chapter, how the EU Body of European Regulators for Electronic Communications frames (BEREC 2012, p. 2) the way in which its own guidance bites: ‘This guidance is not legally binding. Nevertheless, NRAs [national regulatory authorities] are required to take the utmost account of it’. The EU’s financial services agencies are also clear that their guidance is part of a ‘comply or explain’ approach; one that is also common in financial services regulation in many EU Member States and the UK (see Leino 2017). Similar is seen with other EU agencies. Both the European Medicines Agency (EMA) and the European

Fictions and fuzziness  227 Union Aviation Safety Agency (EASA), for example, set out that if those within their regulatory purview chose not to follow the guidance that the EMA or EASA has produced, then any deviations or alternative approaches will need to be fully justified, with the burden of proof on the regulatee (EMA 2008; EASA 2015). As Chamon (2016, p. 35) has argued in the context of the EMA’s guidance, it is ‘worth stressing here that while the non-binding character of EMA’s soft law is acknowledged by the agency itself, it still requires parties to justify their intended non-compliance with such non-binding guidance’. While, as a result of changes from the Lisbon Treaty, certain acts of EU agencies are in theory amenable to judicial review by the EU courts (see Article 263 TFEU), reality is more complex (Eliantonio and Ştefan 2018; Eliantonio 2021). Although the EU courts appear reluctant to use the term ‘soft law’ itself (Korkea-aho 2018, p. 477), there is a body of EU jurisprudence on these sorts of norms in general (which, by analogy, is relevant to, and interesting in the context of, EU agency guidance) and then a smaller body of case law specifically on EU agency guidance. As Mariolina Eliantonio and Oana Ştefan (2018) make clear, while the EU courts have acknowledged the binding effects of EU soft law in some limited circumstances and have accepted that soft law can guide the exercise of EU administrative discretion (Ştefan 2014), those courts have been ‘reluctant to give much weight to the effects that soft law can entail’ (Eliantonio and Ştefan 2018, p. 462). Those courts have also been engaged (and continue to engage) in gatekeeping exercises: taking views on whether it is acts that are intended to have legal effects or acts that are capable of producing legal effects that are justiciable, and on what ‘legal effects’ might mean (for a fuller discussion see Gentile 2020; Eliantonio 2021; Xanthoulis 2021). Put together, what this means then is that most EU agency guidance (and most EU soft law more generally) falls outside the ambit of both the EU courts and of the specialist Boards of Appeal that are attached to many EU agencies (Ştefan and Petri 2018). EU agency guidance is produced in large volumes but rarely litigated or subject to ex-post review by the courts.

3.

THE EUROPEAN CHEMICALS AGENCY

In this section, I want to explore the practices of the European Chemicals Agency (ECHA) as a case study on EU agency soft law rule-making. Since 2007, REACH has been the primary piece of control legislation for chemicals in the EU (Stokes and Vaughan 2013; Haigh 2015). At its most basic, REACH requires the generation of data on the intrinsic properties of certain chemicals (around 45,000 of the 105,000 substances currently on the EU market) by the private sector (the manufacturers, importers and (sometimes) downstream users of those chemicals) followed by the registration of those chemicals (accompanied by their testing data) with ECHA (for a more detailed review, see Bergkamp 2013; Drohmann and Townsend 2013; Vaughan 2015b). Over time, ECHA has also become responsible for seven other pieces of EU chemicals regulation, but my focus here is on its role in relation to REACH. Like many other of the EU’s decentralized agencies, ECHA is intended to act, in various ways, as a conduit between the private sector, Member States, the Commission, and other stakeholders (such as consumers and NGOs) relevant to the functioning of REACH. At the same time, ECHA is also mandated to take regulatory decisions (for example, on whether or not to accept a chemical registration dossier). As a piece of legislation, REACH is vast (the most recent unofficial consolidated version is 552 pages long), and the norms in the

228  Research handbook on soft law Regulation take a variety of forms: some are prescriptive and detailed, others are bare framework commands without any underlying substance in the Regulation (see Fisher 2008; Scott 2009; Kjaer 2010). ECHA’s power to produce guidance comes, in part, from Article 77(2) of REACH which details that one of the tasks of ECHA’s Secretariat is to: … (g) provide technical and scientific guidance and tools where appropriate for the operation of this Regulation … ; (h) provide technical and scientific guidance on the operation of this Regulation for Member State competent authorities and providing support to the helpdesks established by Member States … ; [and] (i) provide guidance to stakeholders including Member State competent authorities on communication to the public of information on the risks and safe use of substances …

This mandate is wide and non-specific. However, the Regulation then also details a small number of specific instances where ECHA is obliged to produce guidance: for example, cost sharing guidance for what are known as Substance Information Exchange Fora (Article 27(1) of REACH), and in relation to certain applications for a time-limited authorization for a harmful chemical to remain on the market (Annex XVI of REACH). Since 2003 (and partly via preparatory REACH Implementation Projects which began before REACH was passed), more than one million words of official guidance have been produced by ECHA on REACH. Put another way, notwithstanding the length and complexity of the underlying Regulation, the core Guidance Documents produced by ECHA are almost ten times as long as the text of REACH. It is no hyperbole to say that, without the ECHA guidance, REACH simply would not work. As was set out above, just as (hard) law is pluralized and differentiated, so too are the soft law rules produced by EU agencies (and soft law more generally). ECHA’s guidance comes in a variety of forms. In relation to REACH, ECHA produces ‘Guidance Documents’, ‘Guidance in a Nutshell’, ‘Guidance Factsheets’, ‘Practical Guides’, ‘Formats’, ‘Practical Examples’, ‘Testing Methods’, ‘Recommendations to Registrants’, and an ‘Identify Your Obligations’ tool, together with responses to over 1,300 separate FAQs. ECHA sees its guidance as living documents and comments that, ‘Guidance reflects the “state-of-the-art” in the practical implementation of the legislation’ (ECHA 2012, p. 46). As well as differing in terms of title and content, ECHA’s guidance also varies in length and is subject to revision over time. So, for example, Version 1.0 of guidance addressed to Member States on the preparation of a dossier on very harmful chemicals was 58 pages long, more than 22,000 words. Version 2.0 of that guidance was a mere nine pages long. ECHA (2012, p. 46) has previously accepted that some of its guidance is dense and complex and committed to working to make its guidance more accessible and, where possible, simpler. ECHA labels everything save for the core Guidance Documents as ‘quasi guidance’ (ECHA 2012, p. 29) – an interesting term which is also largely meaningless; ECHA’s stated intent being that these ‘quasi guidance’ documents are ‘in simple terms’ and particularly intended for Small and Medium Enterprises (ECHA 2012, p. 29). ECHA’s core Guidance Documents, meanwhile, are labelled by the agency as ‘“high level” or “formal” guidance’ (ECHA 2013, p. 2). This sort of fuzzy logic (discussed further below) was also seen in N.V. Elektriciteits v ECHA (2011), a case before the ECHA Board of Appeal. In the case, the Board commented (2011, para. 56) on the differences between the core Guidance Documents produced by ECHA and the Agency’s FAQs.

Fictions and fuzziness  229 The legal nature of the FAQs needs to be distinguished from the REACH Guidance, which are drafted and issued in close co-operation with the stakeholders. Compared to the REACH Guidance, the legal nature of the FAQs is different and less complex as the Agency alone decides on the contents of the FAQs and their purpose is to directly inform registrants of the Agency’s administrative practice.

Exactly what it means to say that the ‘legal nature’ of FAQs is ‘different’ to other forms of ECHA guidance is not elaborated on by the Board, and this is even more confusing when we reflect on the binding effects (legal or practical) of ECHA’s guidance (discussed immediately below). What we also see with ECHA is some form of regulatory evolution in guidance-making. In papers linked to a December 2019 ECHA Management Board meeting we are told that (ECHA 2019, p. 1): Guidance documents addressed to authorities were in the past used to ensure a common understanding of the topic at stake. While this was justified in the early years of the regulatory processes, these documents can now in many cases be gradually replaced by a more flexible approach (for example through ‘approach documents’) in order to ensure a more dynamic and efficient buy-in.

Again, and at the risk of belabouring the point about the lack of clarity, how an ‘approach document’ is different to a guidance document is not immediately clear. What we also learn from this meeting is that the Agency intends to make Factsheets and Nutshells ‘gradually obsoleted’ (ECHA 2019, p. 1), reinforcing the idea that agency guidance has a life and rhythms. Each of ECHA’s Guidance Documents contains a ‘Legal Notice’ in the following terms: ‘Users are reminded that the text of the REACH Regulation is the only authentic legal reference and that the information in this document does not constitute legal advice. The European Chemicals Agency does not accept any liability with regard to the contents of this document.’ Given the volume of ECHA’s guidance, and my earlier comment that REACH simply could not work without ECHA’s guidance, these disclaimers feel more performative than substantive. Certainly, there is an expectation on the part of the agency that those subject to REACH follow ECHA guidance, and this is an expectation that is more widely appreciated. A leading chemicals attorney, Lucas Bergkamp (2017, p. 190), has written about ECHA’s guidance that, ‘[t]o practising lawyers, guidance often is a helpful tool in advising clients. Guidance may provide instructions on how to comply with the law; it may fill the gaps, or address uncertainties or ambiguities’. And the agency itself speaks (ECHA 2021a, p. 23) of the force of its own guidance in the following terms: ‘A significant improvement in compliance is achieved if more companies make full use of the tools, templates and guidance that ECHA has developed in collaboration with industry associations.’ In Momentive Specialty Chemicals v ECHA (2014), the agency had rejected an appellant’s registration dossier as inadequate. This was because Momentive had failed to comply with ECHA’s guidance on read-across (where endpoint information for one chemical is used to predict the same endpoint for another chemical). In this case, the ECHA Board of Appeal commented (2014, para. 61) that: … in not following the available guidance the Appellant did not avail itself of a tool designed to help registrants to prepare and submit their read-across proposals in an effective way. The Board of Appeal observes that in so doing the Appellant may have required additional effort to justify its case compared with following the approach described in the guidance.

230  Research handbook on soft law This comment suggests that while ECHA’s guidance is not necessarily binding on third parties, in that registrants are not necessarily obliged to follow it, where third parties use standards or take approaches different to those set out in the Agency’s guidance, ‘additional effort’ will be required of them to justify taking that path. This in turn suggests that ECHA’s guidance may, in practice, only really be semi-soft. In terms of addressees, most of ECHA’s guidance is explicitly aimed at registrants. But there are also core Guidance Documents produced by ECHA aimed at Member States and at the agency itself (setting out, for example, how the agency will evaluate chemical dossiers). It is worth noting that the language that ECHA uses in its guidance differs depending on to whom the guidance is addressed. A comparison is instructive. ECHA’s Guidance on Information Requirements and Chemical Safety Assessment is the densest and most technical of all the guidance published by ECHA. It is, in fact, 28 separate guidance documents, amounting to more than 200,000 words of text across more than 2,000 pages. These 28 documents are (implicitly) addressed to chemists and regulatory scientists, detailing processes for chemicals data collection and assessment, the identification of data gaps, and the subsequent generation of additional data to fill those gaps. The more accessible Guidance on Registration, however, is written in layperson-friendly language for those ‘with or without expert knowledge in the field of chemicals’. We also see a more generous form of drafting in ECHA’s Guidance on the Communication of Information on the Risks and Safe Uses of Chemicals, which is addressed to Member States. Much of what is written in this Guidance Document is suggestive, with Member State Competent Authorities advised to ‘consider’ doing various things on 59 separate occasions. This suggests a reflexivity and awareness on the part of the Agency as to the impact of its guidance, and a degree of political manoeuvring. Let us turn now to some of the mechanics of how guidance is produced by ECHA. A Guidance Consultation Procedure was first adopted by ECHA’s Management Board in 2008, with full implementation of the procedures and workflows for developing and updating guidance occurring in 2009. ECHA’s aim with the development and review of its guidance is to build consensus between various actors. This, however, is not always possible. The 2008 Guidance Consultation Procedure was said to have led to, ‘protracted discussions on scientific, technical or policy issues which caused delays’ (ECHA 2012, p. 47). As a result, ECHA implemented a new Consultation Procedure on Guidance in 2011 which allowed the Agency to, ‘finalise guidance on the basis of majority views if full consensus cannot be achieved’.5 What this means then is that, even if on the face of the guidance document it seems accepted, there could have been strong, minority views against its norm shaping.6 While there are some minutes published on the Agency’s website which help with aspects of guidance histories, it is very difficult to have an in-depth sense of contestation, participation and deliberation in guidance-making for any of ECHA’s guidance documents produced (or updated) before 2020 (when a (re)revised Consultation Procedure came into effect). ECHA’s Secretariat collects data, from various sources, on difficulties with the guidance that the Agency has produced. Once issues with guidance are identified, there are four possible ways forward: a simple editorial change or correction, an update or a revision (which effects limited or more general changes to the substance of the guidance), or the issuing of a wholly new Guidance Document (ECHA 2013). Who is consulted, and how, on changes to ECHA guidance, or on the writing of new guidance, has changed over time; and has (disappointingly) become less expansive over time. As from April 2020, while the EU Commission and the Member State Competent Authorities will always be consulted, an ECHA Head of Unit is

Fictions and fuzziness  231 allowed to decide: (a) whether to skip consulting other bodies ‘for justified reasons’, and/or (b) whether to engage in consultation with consulted bodies in parallel or consecutively (ECHA 2021b). So, for example, the Head of Unit can decide to consult or not to consult with a Partner Expert Group (a group comprising Accredited Stakeholder Organisations (ASOs) and other nominated experts). Even where there is a Partner Expert Group (PEG) formed to engage in a review of ECHA guidance, this is a form of closed consultation (in that PEG members are invited to take part by ECHA project managers). ECHA’s ASOs are meant to represent differing fields of competence at the EU level. However, the vast majority of ASOs are industry representative bodies (on which, see Biedenkopf 2015). Others have suggested that the success of new governance approaches is, ‘significantly affected by the extent to which certain stakeholder groups are marginalised or absent’ (de Búrca 2010, p. 235). It is clear that the discretion granted to ECHA as regards consultation, both in terms of when to consult and who to consult about its guidance, creates the potential for a new governance failure. Recent work by Joosen (2020) has shown how certain interest groups shape EU agency rule making but generally ‘remain under the radar’ when it comes to formal, public matters like consultations. Historically, ECHA did not publish who took part in PEGs. And so, when one looks at the detail of Emilia Korkea-aho’s case study (2013) on ECHA’s Guidance on Requirements for Substances in Articles, what is interesting is that her insights into who actually participated in the generation of that guidance document come not from public sources, but from personal communications with ECHA. As Korkea-aho (2013, p. 372) notes, this lack of public information on contributions during the writing of ECHA guidance stands in stark contrast to what happens with other EU environmental regimes, for example under the Water Framework Directive. This practice has changed over time and ECHA now publishes, alongside guidance consultation documents, the composition of each PEG. The Agency comments that stakeholder engagement in the guidance updating process is important: (a) to reduce possibility of errors, and (b) to get, as ECHA terms it, ‘buy in’ from the various actors. ECHA’s 2011 review of REACH (ECHA 2012) details that, between 2006 and 2011, the Agency published 71 guidance documents, undertook 30 consultations on guidance with Member State competent authorities and consulted 254 experts as part of PEGs.7 In earlier work on ECHA’s guidance on REACH (Vaughan 2015b), I had suggested that soft law rule-making can serve four functions. I claimed that amplification occurs where guidance produced by ECHA goes beyond, but is not in direct contradiction with, the text of the Regulation. This sort of guidance simply says more of what is in the Regulation. Standardization, the second function, is a subset of the amplification function. With standardization, the goal of ECHA is to channel registrants (and others) down given avenues of action (not set out specifically in the text of REACH) to make the tasks for which ECHA is responsible more manageable. So, for example, the Regulation says X is an outcome and ECHA then publishes guidance on the processes and steps needed to get to X. The third function, translation, occurs when, despite the text of REACH being clear, the Agency, in its guidance, implicitly contests the drafting of the Regulation and so translates the relevant provisions into something else. This happens on only a handful of occasions across ECHA’s more than one million words of guidance (see Vaughan 2015b for examples). Finally, I said that extrapolation occurs where REACH is silent on a particular matter that ECHA felt was important for the operation of REACH and so guidance is issued to fill in the gap. As is probably obvious, we might think of amplification and standardization as legitimate aims of guidance produced by an EU agency, and we might also agree that extrapolation is nec-

232  Research handbook on soft law essary for the efficient day-to-day working of REACH (in that amendments to the legislation to fill obvious gaps are unlikely to happen). Translation, however, is more troubling (raising questions of legitimacy and agency over-reach), but also much less frequent. Looking back, I am not now as sure as to whether such bright-line delineations between EU agency guidance functions are especially helpful. There is some utility, I think, in grouping EU agency soft norms into those we might find more or less problematic, but the general direction of travel of my work on ECHA (here, and previously) is that we have a complex governance situation characterized by various forms of fuzziness. What we see with this ECHA case study is both a high degree of variation within a single EU agency when it comes to guidance production (in terms of approach, format, title, language, addressees, authorship, length, revisability, and so on) and a level of fuzzy logic when it comes to the function, impacts, and effects of that guidance. Let us now do some standing back and reflect further on those forms of fuzziness, seen with more generally with EU agency guidance.

4.

SOME CONCLUDING THOUGHTS ON FICTIONS AND FUZZINESS

In C-16/16 P Belgium v Commission (2017), Advocate General Bobek suggested that soft law instruments contain ‘a type of imperfect norm’; a mix of a ‘normative ambition’ to secure compliance with a lack of ‘instruments of direct coercion’. This strikes me as a useful framing catch-all. Let me try to take that thinking a little further by suggesting that the imperfection in EU agency soft law norms comes from a series of fictions and fuzziness that are told and/ or believed (by certain legal scholars, judges, and policy makers) about those norms. The first is that EU agencies have no general regulatory powers, Majone (2005, p. 93) wryly noting that ‘the only formula that seems to be acceptable to the Brussels authorities is an oxymoron: a regulatory agency without regulatory powers’. As Oana Ştefan and Marina Petri (2018, pp. 525–526) have argued (in the context of ACER), ‘under the guise of their “technical” nature, some of these [guidance] instruments can contain normative rules’, which belies the ‘strict application’ of the Meroni doctrine. The second is the fiction that any discretion granted to EU agencies has a tight backstop of clearly delineated and enumerated executive powers granted from an EU institution (see the framing in the C-270/12 ESMA Short Selling case for example). This is prima facie not true in every instance. The third is that guidance produced by EU agencies does not have legal effects, Vianello (2018, p. 555) rightly instead suggesting that ‘the legal freedom not to follow a merely conditioning act is often a mere fiction’. The fourth is that it could be harmful on legitimacy grounds were EU agencies to be engaged in norm-making that had legal effects (see, for example, the discussions in Scott 2011; Rocca and Eliantonio 2019; van Rijsbergen 2021). The problem here is that EU agencies do, regularly and routinely, engage in soft law rule-making practices which have legal effects. As Senden and van den Brink (2012, p. 16) remind us: Although soft law instruments by definition lack legally binding force as an inherent feature, they may indeed still generate various legal effects inter alia on the basis of the principles of legitimate expectations and equal treatment and by functioning as standards of administrative and judicial interpretation and review.

Fictions and fuzziness  233 The connected challenge, the idea that EU agency guidance is not binding, is that regulatees of EU agencies will/may need to explain why they have not complied with that guidance and/ or demonstrate how their alternative approach is in line with what that guidance says. This seems, at best, semi-soft rule-making (and, I suspect, almost certainly feels ‘hard’ to those subject to the guidance). We also need to recall that, while not binding on third parties, EU agency guidance can have a self-binding effect on those who author it and may also have some (query what exactly?) ‘legal significance’ (see C-233/02 France v Commission, para. 40). The fifth, related, fuzziness is the claim that, at best, all that EU agencies do is engage in quasi-regulatory, quasi rule-making (where what ‘quasi’ means is not well set out but is doing some very heavy lifting). The sixth fuzziness is that while the EU courts are not bound by EU soft law, national courts either must or perhaps may (depending on the line of EU case law one follows) nevertheless take such soft law into consideration (Eliantonio 2018; Korkea-aho 2018). Finally, there is also some fuzziness over the question of whether it is the substance and not the form of an EU agency (soft law) instrument which is of primary importance when thinking about impacts, effects, and so on (see, for example, Ştefan and Petri’s (2018) discussion about the ACER Energie-Control order). The ‘complex normativity’ (Scott 2011, p. 354) of EU agency soft law rule-making is interesting, worthy of study, messy and, I would suggest, not especially amenable to any set of bright-line conclusions. What we see with the soft law rule-making practices of the EU’s decentralized agencies is the ‘steady development of non-binding powers aimed at generating regulatory effects’ (Simoncini 2018, p. 2) and degrees of variation along multiple axes which highlight both how soft law can be just as detailed and as thick as hard law, and that hierarchy and flavours of normative differentiation can be seen in soft norms just as they can in hard.

NOTES 1. BEREC 2012, p. 2. 2. Agencies are referenced in the Lisbon Treaty, though a number of these references are an ex post facto acknowledgement of where the EU institutional landscape has come to. So, for example, Article 2(2)(f) of the Treaty of Lisbon amends the words ‘institutions or bodies’ and ‘institutions and bodies’ in the (renamed) Treaty on the Functioning of the European Union to instead refer to ‘institutions, bodies, offices or agencies’. Article 214 of the Lisbon Treaty also extends the purview of the CJEU to cover certain agency acts. 3. Seen with the EU Body of European Regulators for Electronic Communications (BEREC), the European Union Aviation Safety Agency (EASA), the European Banking Authority (EBA), the European Food Safety Authority (EFSA), the European Insurance and Occupational Pensions Authority (EIOPA), the European Union Agency for Railways (ERA), and the European Stability Mechanism (ESM). 4. Namely, with the Community Plant Variety Office (CVPO), ECHA, and the Single Resolution Board (SRB). 5. This language, however, no longer appears in the December 2013 ‘second revision’ of the consultation procedure guidance. A re(revised) Consultation Procedure was published in April 2020. 6. As from April 2020, ECHA’s latest Consultation Procedure on Guidance sets out that where consensus was not achieved, the first page of the guidance document should include a note ‘to make the reader of the guidance specifically aware of a lack of consensus and provide a cross-reference to the meeting minutes’ (ECHA 2020, p. 6). 7. ECHA’s Report on the operation of REACH and CLP (Classification, Labelling and Packaging) 2021 does not contain any similar (more current) data.

234  Research handbook on soft law

REFERENCES Armstrong, K. A., (2011). The character of EU law and governance: from ‘Community method’ to new modes of governance. Current Legal Problems. 64(1), 179–214. BEREC., (2012). BEREC guidelines on the application of Article 3 of the Roaming Regulation - wholesale roaming access [online]. Riga: BEREC. [Viewed 26 January 2022]. Available from: https://​www​ .arcep​.fr/​fileadmin/​reprise/​dossiers/​roaming/​03​-berec​-guidelines​-on​-Access​_obligation​-Art3​.pdf. Bergkamp, L., (2013). The European Union REACH regulation for chemicals: law and practice. Oxford: Oxford University Press. Bergkamp, L., (2017). Book review: EU chemicals regulation: new governance, hybridity, and REACH by Steven Vaughan. Cheltenham: Edward Elgar Publishing, 2015, 288 pp. Journal of Environmental Law. 29(1), 189–193. Biedenkopf, K., (2015). EU chemicals regulation: extending its experimentalist REACH. In: J. Zeitlin, ed. Extending experimentalist governance? The European Union and transnational regulation. Oxford: Oxford University Press. pp. 107–136. de Búrca, G., (2010). New governance and experimentalism: an introduction. Wisconsin Law Review. 2, 227–238. Busuioc, M., (2013). European agencies: law and practices of accountability. Oxford: Oxford University Press. Case A-001-2010 N.V. Elektriciteits – Produktiemaatschappij Zuid-Nederland EPZ v ECHA, Decision of 10 October 2011. Case A-006-2012 Momentive Specialty Chemicals v ECHA, Decision of 13 February 2014. Case C-9/56 Meroni & Co., Industrie Metallurgiche, SpA v High Authority of the European Coal and Steel Community ECLI:EU:C:1958:7. Case C-233/02 France v Commission ECLI:EU:C:2004:173. Case C-270/12 United Kingdom of Great Britain and Northern Ireland v European Parliament and Council of the European Union ECLI:EU:C:2014:18. Case T-721/14 Kingdom of Belgium v European Commission ECLI:EU:T:2015:829. Case C-16/16 P Opinion of Advocate General Bobek delivered on 12 December 2017, Kingdom of Belgium v European Commission ECLI:EU:C:2017:959. Chamon, M., (2010). EU agencies: does the Meroni Doctrine make sense? Maastricht Journal of European and Comparative Law. 17(3), 281–305. Chamon, M., (2016). EU agencies: legal and political limits to the transformation of the EU administration. Oxford: Oxford University Press. Chiti, E., (2013). European agencies’ rulemaking: powers, procedures and assessment. European Law Journal. 19(1), 93–110. Coen, D. and Thatcher, M., (2005). The new governance of markets and non-majoritarian regulators. Governance. 18(3), 329–346. Coman-Kund, F., (2018). European Union agencies as global actors: a legal study of the European Aviation Safety Agency, Frontex and Europol. London: Routledge. Drohmann, D. and Townsend, M., eds., (2013). REACH: best practice guide to Regulation (EC) No 1907/2006. Oxford: Hart Publishing. Dunleavy, P. and Hood, C., (1994). From old public administration to new public management. Public Money & Management. 14(3), 9–16. EASA., (2015). Response to FAQ n.19207. why should I follow an AMC? [online]. EASA. [Viewed 24 January 2022]. Available from: https://​www​.easa​.europa​.eu/​faq/​19207. ECHA., (2012). General report 2011 (ECHA-12-A-02-EN) [online]. Helsinki: European Chemicals Agency. [Viewed 24 January 2022]. Available from: https://​ echa​ .europa​ .eu/​ documents/​ 10162/​ 17210/​mb​_06​_2012​_general​_report​_2011​_final​_en​.pdf/​cffe481c​-88f6​-4ce4​-ad50​-df16c0c92613​?t​=​ 1334247381290. ECHA., (2013). Second revision to the Consultation Procedure on Guidance (MB/63/2013 final) [online]. Helsinki: European Chemicals Agency. [Viewed 24 January 2022]. Available from: https://​ echa​.europa​.eu/​documents/​10162/​17208/​mb​_63​_2013​_revision​_consultation​_procedure​_guidance​ _en​.pdf/​f32a9a0f​-fc05​-4169​-921e​-f0c16dc51ae4.

Fictions and fuzziness  235 ECHA., (2019). Updated consultation procedure for guidance. 56th meeting of the Management Board 12-13 December 2019 (MB/62/2019 final) [online]. Helsinki: European Chemicals Agency. [Viewed 24 January 2022]. Available from: https://​echa​.europa​.eu/​documents/​10162/​2166617/​ FINAL​_MB​_62​_2019​_Updated​_consultation​_procedure​_guidance​_MB56​.pdf/​f2f2f508​-beeb​-15da​ -d370​-504a149fd05b. ECHA., (2021a). ECHA Programming Document 2021-2024 (ECHA-21-R-01-EN) [online]. Helsinki: echa​ .europa​ European Chemicals Agency. [Viewed 24 January 2022]. Available from: https://​ .eu/​documents/​10162/​13609/​programming​_document​_2021​-2024​_en​.pdf/​fdc0a236​-696b​-b2a8​-a6fb​ -523066eadf30. ECHA., (2021b). Consultation procedure for guidance (PRO-0011.11) [online]. European Chemicals Agency. [Viewed 24 January 2022]. Available from: https://​echa​.europa​.eu/​documents/​10162/​17207/​ pro​-0011​_consultation​_procedure​_for​_guidance​_en​.pdf/​21fa2b20​-60cc​-481e​-833b​-9afbee9ac966. 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), 675–690. Eliantonio, M., (2018). Soft law in environmental matters and the role of the European courts: too much or too little of it?. Yearbook of European Law. 37, 496–524. Eliantonio, M., (2021). Judicial review of soft law before the European and the national courts: a wind of change blowing from the member states?. In: M. Eliantonio, E. Korkea-aho and O. Ştefan, eds. EU soft law in the member states: theoretical findings and empirical evidence. Oxford, UK: Hart Publishing. pp. 283–302. Eliantonio, M. and Ştefan, O., (2018). Soft law before the European courts: discovering a ‘common pattern’?. Yearbook of European Law. 37, 457–469. EMA., (2008). Status of EMEA scientific guidelines and European Pharmacopoeia monographs and chapters in the regulatory framework applicable to medicinal products (EMEA/42371/2008) [online]. European Medicines Agency. [Viewed 24 January 2022]. Available from: https://​www​.ema​.europa​ .eu/​en/​documents/​scientific​-guideline/​status​-emea​-scientific​-guidelines​-european​-pharmacopoeia​ -monographs​-chapters​-regulatory​-framework​_en​.pdf. Fisher, E., (2008). The ‘perfect storm’ of REACH: charting regulatory controversy in the age of information, sustainable development, and globalisation. Journal of Risk Research. 11(4), 541–563. Gentile, G., (2020). Ensuring effective judicial review of EU soft law via the action for annulment before the EU courts: a plea for a liberal-constitutional approach. European Constitutional Law Review. 16(3), 466–492. Haigh, N., (2015). EU environmental policy: its journey to centre stage. London: Routledge. Howell, E., (2019). EU agencification and the rise of ESMA: are its governance arrangements fit for purpose?. The Cambridge Law Journal. 78(2), 324–354. Joosen, R., (2020). The tip of the iceberg – interest group behaviour in rule drafting and consultations during EU agency rulemaking. Journal of European Public Policy. 27(11), 1677–1697. Keleman, D. R., (2002). The politics of ‘Eurocratic’ structure and the new European agencies. West European Politics. 25(4), 93–118. Kjaer, P. F., (2010). A hybrid within a hybrid: contextualising REACH in the process of European integration and constitutionalisation. European Journal of Risk Regulation. 1(4), 383–396. Klabbers, J., (1998). The undesirability of soft law. Nordic Journal of International Law. 67(4), 381–391. Korkea-aho, E., (2013). Laws in progress? Reconceptualizing accountability strategies in the era of framework norms. Transnational Environmental Law. 2(2), 363–385. Korkea-aho, E., (2018). National courts and European soft law: is Grimaldi still good law?. Yearbook of European Law. 37, 470–495. Leino, P., (2017). Accountability dilemmas of regulating financial markets through the European Supervisory Agencies. In: C. Harlow, P. Leino and G. della Cananea, eds. Research handbook on EU administrative law. Cheltenham, UK: Edward Elgar Publishing. pp. 209–238. Levi-Faur, D., (2011). Regulatory networks and regulatory agencification: towards a Single European Regulatory Space. Journal of European Public Policy. 18(6), 810–829. Majone, G., (2005). Dilemmas of European integration. the ambiguities and pitfalls of integration by stealth. Oxford: Oxford University Press. Parliament, Council, and Commission., (2012). Joint statement and common approach of the European Parliament, the Council of the EU and the European Commission on decentralised agencies [online].

236  Research handbook on soft law Brussels: Council of the European Union. [Viewed 26 January 2022]. Available from: https://​data​ .consilium​.europa​.eu/​doc/​document/​ST​-11450​-2012​-INIT/​en/​pdf. Reinicke, W. H. and Witte, J. M., (2000). Interdependence, globalization, and sovereignty: the role of non-binding legal accords. In: D. Shelton, ed. Commitment and compliance: the role of non-binding norms in the international legal system. Oxford: Oxford University Press. pp. 75–100. Rijsbergen, M. van, (2021). Legitimacy and effectiveness of ESMA’s soft law. Cheltenham, UK: Edward Elgar Publishing. Rimkutė, D., (2015). Explaining differences in scientific expertise use: the politics of pesticides. Politics and Governance. 3(1), 114–127. Rocca, P. and Eliantonio, M., (2019). European Union soft law by agencies: an analysis of the legitimacy of their procedural frameworks. In: M. Conticelli, M. De Bellis and G. della Cananea, eds. EU executive governance: agencies and procedures. Torino: Giappichelli. pp. 177–203. Scott, J., (2009). REACH: combining harmonization with dynamism in the regulation of chemicals. In: J. Scott, ed. Environmental protection: European law and governance. Oxford: Oxford University Press. pp. 56–91. Scott, J., (2011). In legal limbo: post-legislative guidance as a challenge for European administrative law. Common Market Law Review. 48(2), 329–355. Senden, L., (2013). Soft post-legislative rulemaking: a time for more stringent control. European Law Journal. 19(1), 57–75. Senden, L. and Brink, T. van den, (2012). Checks and balances of soft EU rule-making [online]. Brussels: European Parliament. [Viewed 25 January 2022]. Available from: https://​www​.europarl​ .europa​.eu/​thinktank/​en/​document/​IPOL​-JURI​_ET(2012)462433. Simoncini, M., (2018). Administrative regulation beyond the non-delegation doctrine: a study on EU agencies. Oxford: Hart Publishing. Snyder, F., (1994). Soft law and institutional practice in the European Community. In: S. Martin, ed. The Construction of Europe: essays in honour of Emile Noël. Dordrecht: Kluwer Academic Publishers. pp. 197–225. Ştefan, O., (2014). Helping loose ends meet? The judicial acknowledgement of soft law as a tool of multi-level governance. Maastricht Journal of European and Comparative Law. 21(2), 359–379. Ştefan, O. and Petri, M., (2018). Too weak to be controlled: judicial review of ACER soft law. Yearbook of European Law. 37, 525–550. Stokes, E. and Vaughan, S., (2013). Great expectations: reviewing 50 years of chemicals legislation in the EU. Journal of Environmental Law. 25(3), 411–435. Vaughan, S., (2015a). Differentiation and dysfunction: an exploration of post-legislative guidance practices in 14 EU agencies. Cambridge Yearbook of European Legal Studies. 17, 66–91. Vaughan, S., (2015b). EU chemicals regulation: new governance, hybridity and REACH. Cheltenham: Edward Elgar Publishing. Vianello, I., (2018). Upholding a principled approach to the use of EU administrative activities externally: what role for the Court of Justice of the European Union?. Yearbook of European Law. 37, 551–568. Vos, E., (2018). EU agencies, common approach and parliamentary scrutiny: study [online]. R. Korver, ed. Brussels: European Parliament. [Viewed 26 January 2022]. Available from: https://​www​.europarl​ .europa​.eu/​RegData/​etudes/​STUD/​2018/​627131/​EPRS​_STU(2018)627131​_EN​.pdf. Xanthoulis, N., (2021). Legal effects and reviewability of EU acts: between institutional intentions and perceptions of concerned parties. In: M. Eliantonio, E. Korkea-aho and O. Ştefan, eds. EU soft law in the member states: theoretical findings and empirical evidence. Oxford, UK: Hart Publishing. pp. 303–322.

16. Administrative guidance in the United States: the moral and political stakes of non-binding law Blake Emerson

1. INTRODUCTION In December 2021, the Omicron variant of COVID-19 began to spread through the United States. To the surprise of many, the Centers for Disease Control and Prevention (CDC) responded by substantially altering its guidance for isolation and quarantine, shifting the recommended isolation period from ten days to five. Dr. Anthony Fauci, Chief Medical Advisor to President Biden, explained the controversial change: ‘The CDC made a decision to balance what is good for the public health at the same time as keeping society running’ (Beals 2021). Given the sheer number of people who would likely become infected by the new variant, a ten-day isolation window might seriously impair the economy. A shift in guidance was necessary to ‘get society to function properly’. This was a particularly stark example of how administrative guidance can have significant normative and social impacts, despite the fact that it lacks direct, legally binding effect. Not only in the Pandemic, but across other important policy domains, such as immigration and racial and gender civil rights, government documents that lack the force of law can alter the moral self-understandings of individuals and communities and enter into high-profile political disputes. This chapter will examine administrative guidance in the US as a particular example of the broader phenomena of ‘soft law’. While the term is used in different senses, soft law roughly refers to nonbinding or hortatory law in the international as well as transnational and domestic contexts (Weil 1983; Klabbers 1996; Gersen and Posner 2008; Galbraith and Zaring 2014; Korkea-aho et al. 2021). Guidance is the term US legal scholars use to refer to such nonbinding documents in the context of federal administrative law (Anthony 1992; Mendelson 2007; Levin 2018; Parrillo 2019). These documents have been the subject of intense scholarly debate and political controversy in recent years (Emerson 2019a). Prized for its ease of issuance and flexibility in application, guidance is arguably the lifeblood of administrative process. It enables administrators to implement binding legal requirements and purposes efficiently, without prolonged procedures and without imposing strict rules on regulated parties. It creates a means for the leadership of government agencies to control the discretion of enforcement staff and other officials, so that the implementation of law is consistent, predictable, and democratically responsive. At the same time, these documents can have significant coercive and normative effects. Precisely because guidance can be issued and changed swiftly, and because it is often not subject to judicial review, there is a risk that the terms of guidance may fall outside the issuing agencies’ statutory authority or else be poorly justified. Guidance may also coerce parties to comply 237

238  Research handbook on soft law without affording them procedural protections that would apply if the agency took adjudicatory action against them or issued a rule with the force of law. The benefits and risks described above are framed by the well-worn contours of the contest between governmental power and private liberty. Guidance helps the government carry out law but by the same token risks arbitrary infringements on the private sphere. This chapter goes beyond that familiar tension to discuss how nonbinding law enters into political and social conflict and informs moral and ethical judgment. By staking out an official position on uncertain and sometimes controversial matters of policy, guidance mediates political conflict. It affirms some practices, identities, statuses, and forms of life even as it marginalizes others. By telling people how they can comply with otherwise valid obligations, guidance clarifies the content of those obligations. Unlike binding law, guidance by its terms anticipates revision, departure, and amendment. It is suited to times of political friction and uncertainty. However, the use of guidance in such moments may heighten ongoing crises of administrative legitimacy (Habermas 1975; Freedman 1978). Section 2 of this chapter lays out the legal framework in which federal administrative guidance operates. Section 3 then describes normative tensions concerning liberty and democracy that are at issue in guidance and explains some recent efforts at reform. Section 4 explains how guidance interacts with moral judgment and political conflict.

2.

THE LEGAL FRAMEWORK

Guidance is the generic term legal scholars use for federal administrative documents that lack ‘the force of law’, such as policy statements, enforcement memoranda, staff manuals, and frequently asked questions (Anthony 1992). The general legal framework that governs these documents is the Administrative Procedure Act of 1946 (APA). This part describes how the APA’s rulemaking process works and how guidance documents are exempted from that process. 2.1

Rulemaking in the Administrative Procedure Act

Enacted in the wake of the New Deal, the APA is a quasi-constitutional statute. It sets out the basic procedural requirements for administrative action, such as the issuance of general regulations and the adjudication of individual controversies, as well as provisions governing judicial review of such actions. The APA has constitutional heft not only because it fundamentally structures how the executive branch and modern government function. In addition, the Act represents a political settlement between the ideological proponents and opponents of the welfare and regulatory state that had consolidated by the mid-twentieth century (Shepherd 1996; Grisinger 2012). The APA governs any federal ‘agency’, a term which includes any ‘authority of the Government of the United States’ other than Congress, the President, ordinary courts, territorial governments, and certain military tribunals and authorities (5 U.S.C. § 551(1); Franklin v. Massachusetts 505 U.S. 788 (1992)). It thus covers not merely the so-called ‘independent’ agencies, such as the Securities and Exchange Commission, but also executive branch departments, such as Treasury or State, and the components of such departments. One of the most noteworthy and indeed famed provisions of the APA are its procedures for public participation in administrative rulemaking. Agencies’ ‘legislative’ or ‘substantive’

Administrative guidance in the United States  239 rules have equal legal validity and binding effect to a law enacted by Congress. When agencies make such rules, they must generally publish a notice of their proposal and invite comments from interested parties. Final rules must then be accompanied by a ‘concise general statement of their basis and purpose’ (5 U.S.C. § 553(c)). The courts, Congress, and presidents have embellished these formal requirements into a detailed process of administrative deliberation, consultation, and review (Mashaw 2018). An agency’s statement of basis and purpose today is rarely ‘concise’. Rather, agencies respond in detail to the significant comments they receive. They assess the costs and benefits of the regulation against other alternatives, perform environmental impact statements, and much more. As a consequence of all these analytic requirements, regulatory preambles go on for scores or even hundreds of pages, sometimes accompanied by a docket with all comments received and other supporting materials. That formal record is only a snapshot of a broader process of consultation before the rule is issued, which often involves outreach to regulated parties, and in some cases, proactive engagement with regulatory beneficiaries (Elliott 1992; Sant’Ambrogio and Staszewski 2021). When the agency finalizes its rule, aggrieved parties may challenge the rule in court. A reviewing court will invalidate rules that it finds to be ‘arbitrary’ or ‘capricious’, or which fall outside the bounds or statutory authority, or which are procedurally defective (5 U.S.C. § 706(2)). Review for arbitrariness focuses on whether the agency has offered a reasoned explanation for its action. Courts undertake that inquiry, as well as review of the agency’s legal interpretations, with some degree of deference to Congress’ delegation of authority to the agency as well as the agency’s expertise and political accountability. As a practical matter, the courts have great leeway in whether they give administrative actions a ‘hard look’ that examines the agency’s explanation in granular detail or rather a passing glance to ensure no glaring errors have been made (Sunstein 1983; Gersen and Vermeule 2016). The notice-and-comment process distinguishes US administrative law from that of other countries, such as the United Kingdom, France, and Germany, which do not have the same general, formal, judicially enforceable requirements for public participation in agency policymaking (Rose-Ackerman 2021). Supporters of the US-American process see it as an important if highly imperfect institutionalization of participatory democracy (Mashaw 2018; Emerson 2019b). Critics worry that it impedes efficient administrative action, rigidifies agency policies, and even facilitates capture by regulated parties (Wagner 2010; Bagley 2019). Administrative regulations are ‘hard law’, and the rulemaking process is accordingly quite hard to pursue. It creates multiple opportunity for input, contestation, and reversal before legally binding action is taken. Agencies must pay a steep price in time, resources, and consideration to issue a rule that changes private parties’ legal rights and obligations, and which may establish a sufficient basis for imposing penalties such as fines or even imprisonment. 2.2

The Guidance Exemption from Notice-and-Comment

The APA nonetheless gives agencies several paths to avoid these requirements (Levin 2018). Most importantly for our purposes, it exempts from notice-and-comment requirements ‘interpretative rules, general statements of policy, or rules of agency organization, procedure, or practice’ (5 U.S.C. § 553(b)(A)). These documents, which are different forms of guidance, are ‘softer’ than regulations. General statements of policy, in particular, do not purport to impose mandatory norms, but rather merely ‘advise the public prospectively of the manner in which the agency proposes to exercise a discretionary power’ (United States Department of Justice

240  Research handbook on soft law 1947, p. 30, n. 3). Such policy statements are distinct in that they do not impose ‘legally binding requirements’ (Case 12-5310 National Mining Association v. McCarthy, p. 252 ). For instance, an Environmental Protection Agency document which merely ‘recommends’ that states impose certain permitting requirements on coal mining operations, but which disclaims any ‘legal impact’ qualifies as a policy statement (National Mining Association, p. 253). While these policy statements may create strong incentives to comply, they are insufficient on their own to sustain an adverse legal action against a member of the public. The other documents exempted from rulemaking requirements—interpretive and procedural rules—are soft in some respects, hard in others. An interpretative rule ‘advises the public of the agency’s construction of the statutes and rules which it administers’ (United States Department of Justice 1947, p. 30, n. 3). Such interpretive rules are not supposed to impose ‘new’ obligations that are independent of those already imposed by other sources of binding law. But many courts will consider these interpretations, unlike policy statements, to have the force of law. For example, an interpretive rule may specify a pre-existing legal duty to provide wheelchair-bound sporting spectators with a ‘line of sight’ to the event; it can clarify that such spectators must be able see over standing as well as over seated spectators (Case 97-7005 Paralyzed Veterans of America v. D.C. Arena L.P., p. 587). That administrative interpretation would then be sufficient to uphold an enforcement action taken against a non-compliant sporting arena. By contrast, an agency cannot use a merely interpretive rule to impose a specific height requirement that the agency deems ‘appropriate’ for the structural containment of ‘big cats’ (Case 95-2571 Hoctor v. U.S. Department of Agriculture). A key question in drawing the line between impermissible and permissible use of interpretive rules is ‘how tightly the agency’s interpretation is drawn linguistically from the actual language of a statute or rule’ (Paralyzed Veterans, p. 587). While some scholars use the term ‘guidance’ to refer only to interpretive rules and policy statements (Levin 2018), a procedural rule is also soft in the narrow sense that it must be restricted to internal agency practices and cannot alter the legal rights and obligations of private parties (cf. Metzger and Stack 2017). For example, a decision not to entertain face-to-face applications for label approval would qualify as a procedural rule (Case 99-5305 James V. Hurston Associates, Inc. v. Glickman, p. 280 ). However, such rules have the force of law, in the sense that agency officials must obey them, and courts will enforce them. Indeed, such rules played a crucial role in the investigation and resignation of President Richard Nixon following the Watergate scandal (Case 73-1766 United States v. Nixon). The lines between these categories of guidance, as well as between guidance and regulations, are famously murky, even metaphysical. Given the indeterminacy of the distinctions and the often high economic and social stakes, the appropriate scope of the guidance exemption has become one of the ‘most litigated’ administrative-law issue in the federal courts (Levin 2018, p. 265). Agencies have strong incentives to resort to guidance because it is cheap to issue and rescind as compared to the onerous process of rulemaking, and because it enables the agency’s leadership to exercise control over its staff, which may number in the thousands, tens of thousands, or even hundreds of thousands. Private parties, too, have reasons to seek out notice of the agencies’ position on how to enforce vague statutory and regulatory requirements. But they may also chafe at an agency’s supposedly non-binding statement which seems in fact to require rote compliance. The next section examines the normative issues that underlie these incentives and tendencies.

Administrative guidance in the United States  241

3.

THE GUIDANCE CONTROVERSY IN A COMPLEX NORMATIVE TERRAIN

Legal and political controversy over guidance raises normative problems concerning the administrative state. The debate centres on the conflict between private liberty and governmental discretion. Critics of the proliferation of guidance argue that agencies use such documents to avoid the procedural requirements and judicial review provisions the APA imposes for rulemaking (Anthony 1992). This poses a problem not only for regulated parties, but perhaps even more so, for regulatory beneficiaries. Whereas regulated parties will have the opportunities to contest and seek pre-enforcement judicial review of industry-friendly regulations issued through the notice-and-comment process, they will not have the same opportunities with regard to informal guidance (Mendelson 2007). Proponents of the liberal use of guidance, however, highlight the crucial role that guidance plays in providing notice to the public of the agencies’ future enforcement posture, as well as ensuring the consistent and equal treatment of regulated parties (Strauss 2001; Levin 2018). Guidance also serves as an ‘internal administrative law’ that implements hierarchical accountability, filling in the interstices left by binding statutory and regulatory law with the policy judgments of politically responsible agency leadership (Metzger and Stack 2017). Depending on the jurist’s sense of the relative weight of these risks and benefits, they are apt to choose more or less stringent restrictions on the use of guidance. For instance, on the more stringent end, guidance is impermissibly binding if ‘affected parties are reasonably led to believe that failure to conform will bring adverse consequences’ (Anthony 1992, pp. 1328–1329). On the more permissive end, guidance is impermissibly binding just in case affected parties have no reasonable opportunity to contest the terms of the guidance prior to its enforcement (Elliott 1992; Levin 2018; Emerson 2019a). While such debates are framed by the contrasting values of governmental discretion versus private liberty, the relation between the two is more complex than a simple trade-off. Below I first describe the fraught relationship between liberty and guidance, and then the fraught relationship between democracy and guidance. 3.1

Liberty and Guidance

At first blush, any expansion in agencies’ authority to issue and change guidance, to strictly adhere to guidance, or to rely on guidance as a ground for an enforcement action, would seem to come purely at the expense of affected parties’ liberty. To modify an example from a previous section, a guidance document that recommends big cat fencing of eight feet—if rigidly enforced by the agency—would practically deprive the owner or dealer of lions or tigers of the option of building a fence that is only seven feet tall. If she does build a shorter fence than ‘recommended’, she risks a costly enforcement action. Nevertheless, there are several ways that guidance in fact promotes the liberty of private parties. First, if the issuing agency has significant discretionary power—such as the authority to investigate, adjudicate cases, or issue fines—then the regulated party benefits from the information the guidance conveys about how the agency intends to use that power. Where the underlying, binding rules are vague, such notice may in fact increase regulated parties’ practical ability to act without fear of government sanction. The big cat dealer who intends to build an enclosure will know, in advance, that if she builds the fence eight feet high, she is in

242  Research handbook on soft law the clear. In the absence of the guidance, she will be left to guess how much fence is enough to avoid a penalty. It is useful, in other words, to get some concrete details about what the government takes binding but broad terms like ‘appropriate’ to mean, or how it will exercise the significant coercive powers at hand. Nicholas Parrillo accordingly notes in his landmark study of guidance that regulated parties often crave guidance like ‘water in the desert’ (Parrillo 2019, p. 185). They do so when the agency has vast coercive power. The clearest example of this is the Food and Drug Administration’s (FDA) binding authority to approve drugs and devices before they reach the marketplace. With sometimes hundreds of millions of dollars on the line, regulated parties want to know how to avoid hiccups with the approval process. FDA guidance tells them. A related benefit of guidance is preservation of formal equal treatment among regulated parties, which is necessary to ensure the freedom of each consistent with every other according to general rules (Kant [1797] 1999, p. 30). If one field office of the US Department of Agriculture (USDA) thinks the right fence height for big cats is seven feet, whereas another field office believes it is nine, then a puma owner with an eight-foot-high fence will be found compliant in one jurisdiction but non-compliant in another. Guidance can thus ensure that like parties are treated alike according to general norms rather than subject to arbitrary enforcement that enlarges the discretion of some parties relative to others. A distinct consideration is the mediating role guidance plays in coercive relations among private parties. Suppose that the USDA issues no guidance whatsoever as to the appropriate fence heights for big cats, with the result that an ignorant, bumbling, or miserly cat owner builds a fence penning their jaguar only four feet high. When the jaguar jumps over the fence and mauls a passer-by, that party’s liberty will be grievously and perhaps permanently impaired. Thus, the mere fact that guidance coerces some private parties, sometimes severely, does not mean that it is merely coercive. It may also, and at the same time, empower and enable parties who are otherwise at the mercy of others. For instance, guidance from the Equal Employment Opportunity Commission (EEOC) on the use of arrest and convictions in employments decisions coerces employers to some degree not to ask questions about prior arrests unless they are genuinely related to the job in question (EEOC 2012). Such guidance by the same token increases employment opportunities for persons convicted of offences, enabling them to function as competent and independent members of society. This is not to say that the gains guidance provides to one party’s liberty always compensate for the losses to another’s. When agencies use guidance to give regulated parties, such as polluting industries, a safe harbour against otherwise binding obligations, such guidance may harm the scope of liberty overall. If improperly calibrated, guidance may give regulated parties such great reassurance of non-enforcement that others’ freedom from unsafe air and water would have been better promoted by leaving the regulated parties in the dark as to the likelihood of enforcement. As Nina Mendelson (2007) observes, regulatory beneficiaries may lack the opportunity to challenge the guidance because of the rules governing internal agency adjudication as well as judicial review of agency action. One prominent means of protecting private interests while retaining guidance’s consistency benefits is to provide an internal appeal mechanism accompanied by reason-giving requirements. The FDA and the White House’s Office of Management and Budget (OMB) have thus provided that executive agency officials may only depart from guidance ‘with appropriate justification and supervisory concurrence’ (65 FR 56468; 72 FR 3432). Building on such approaches and Parrillo’s research, the Administrative Conference of the United States

Administrative guidance in the United States  243 (ACUS) recommends that agencies ‘afford members of the public a fair opportunity to argue for lawful approaches other than those put forward by a policy statement or for modification or rescission of the policy statement’ (ACUS 2017). ACUS further recommends that, when agencies do permit deviations from guidance, they consider disseminating their decisions and the reasons for them to other interested parties and to the public. Parrillo calls this approach ‘principled flexibility’ (Parrillo 2019, p. 181). It encourages agencies to develop an internal common law surrounding guidance, where the general policies might be adjusted over time in light of reasoned departures, usually made by higher ranking rather than front line officials. ACUS recommends a similar approach with regard to interpretive rules as well as policy statements, thus rejecting the sharp binary between the two that the courts often insist upon (Levin 2018; ACUS 2019; Emerson and Levin 2019). These approaches give both regulated parties and beneficiaries some assurance that the guidance does not operate as a legally binding requirement in the absence of the required process. 3.2

Democracy and Guidance

Guidance may be thought to pose a risk to democratic accountability. Guidance lacks the procedural-democratic controls of either ordinary legislation or of notice-and-comment rulemaking. As described in section 1, one of the supposed benefits of the US rulemaking process is that it enables participatory democracy to take place within administrative agencies. As imperfect as that process may be, it certainly does give affected members of the public an opportunity to weigh in on the rules by which they will be bound and, in some cases, to challenge such rules in court for failure to take their input into account. The same level of participation is not generally required with regards to guidance documents, because there is no general requirement that agencies solicit and respond to public comment before they issue guidance. This raises the concern that the use of guidance rather than regulation undermines public influence over agencies. The most direct reply to this legitimate worry is that, while guidance does not facilitate participatory democracy, it nonetheless enforces politically accountable officials’ hierarchical control over the bureaucracy. As Gillian Metzger (2015) argues, guidance enables agency leadership to supervise and direct agency staff at low-cost relative to the onerous and lengthy requirements of rulemaking or formal adjudication. Since these officials are appointed by the democratically elected president, they have an indirect popular warrant to govern affairs within their agencies. As a means of presidential control, guidance ensures that the significant discretion vested in executive branch officials and agencies is exercised consistent with the values and interests of an elected office. A salient example of this political use of guidance was the Obama Administration’s Deferred Action for Childhood Arrivals policy. Using an ‘enforcement memorandum’, the Department of Homeland Security (DHS) announced that it would not commence deportation proceedings against certain classes of undocumented immigrants who came to the US as children (DHS 2012). The memorandum was a means of controlling and centralizing the discretion exercised by immigration enforcement officers, so as to align with the political values of the incumbent administration (Motomura 2014). While the Trump Administration attempted to rescind this policy, they were unsuccessful. The Supreme Court concluded that the rescission was arbitrary, particularly because it failed to take into account ‘serious reliance interests’ engendered

244  Research handbook on soft law by the previous guidance (Case 18-587 Department of Homeland Security v. Regents of the University of California). Another procedural response to guidance’s supposed democratic deficit is to replicate some form of public participation in the formulation of guidance documents as well as binding rules. Proponents of this approach recognize, of course, that participation is costly and time consuming; they then aim for a more limited form of participation in the drafting of agency guidance (Mantel 2009). OMB’s Good Guidance Bulletin accordingly requires provision of public comment on ‘significant’ guidance documents, such as those with effects of 100 million USD or more or those that raise novel policy issues (72 FR 3432, p. 3437). But the Bulletin does not require pre-promulgation comment. Agencies commonly provide some opportunity for comment, even if not bound by OMB directives (Emerson and Levin 2019). Public participation can take the form of something as rudimentary as provision of an email address for post-promulgation feedback or a process that is just as formal as the notice-and-comment process required for legislative rules. In 2019, President Trump went further than these practices and issued an executive order generally requiring agencies to solicit and respond to public comments on significant guidance documents (Executive Order 13891 in 84 FR 55235). Given the antiregulatory posture of the Trump Administration, it is doubtful that the intent of the order was to promote genuine public engagement rather than to thwart aggressive actions outside the rulemaking process. The Order also exempted guidance related to homeland security, which would seem to encompass immigration—an area where the Trump Administration was eager to take such aggressive regulatory action without procedural constraint. If fully implemented, the Order would likely have encouraged decreased use of guidance, increased use of rulemaking and adjudication, and increased use of other exemptions to notice and comment requirements, such as the APA’s provision that an agency may avoid notice and comment when it finds that such process would be ‘impractical, unnecessary, or contrary to the public interest’ (5 U.S.C. § 551(b)(B)). This sea change did not come. President Biden swiftly rescinded this document at the start of his term. However, by the time Biden withdrew the Trump order, several agencies had already issued procedural rules committing themselves to follow the Trump order (Chambers 2021). While many agencies have rescinded their guidance regulations, the Trump Administration’s requirements may remain somewhat sticky so long as other agencies’ procedural rules remain in place.

4.

THEORIZING GUIDANCE BEYOND PROCESS: SOFT LAW’S MORAL AND POLITICAL CONTRIBUTION

American administrative law, and perhaps American public law broadly, is process-focused. Where there is fierce dispute over issues of political and moral substance, jurists can turn to questions of procedural fairness to find a neutral terrain and to lower the temperature of social conflict. So it has been with guidance. The political heat around the Obama Administration’s immigration guidance, as well as guidance on issues surrounding campus sexual assault and transgender rights, generated a raft of reform proposals and actions to place heightened procedural requirements on such actions. Some of those proposals are well-taken. In particular, it makes sense to give affected and interested parties opportunity to challenge the terms of guidance documents and to seek departures from them on terms of principled flexibility.

Administrative guidance in the United States  245 Furthermore, where the guidance concerns particularly weighty matters, and there is no serious harm in delay, the public ought to have opportunity to comment on the guidance at an early stage in its formulation. However, such concerns with procedure can sometimes stifle substantive concern with what guidance does, and whether the policies and interpretations adopted in guidance are well-justified (Seidenfeld 2011). Of course, it is beyond the scope of this chapter to discuss whether, for instance, it is appropriate as a matter of policy to defer deportation proceedings against undocumented immigrants who came to the US as children, or whether and when to wear masks as a precaution against COVID-19 transmission. Nonetheless, we ought to think about how such guidance on sensitive, contested topics impacts moral judgment and political life. The underlying phenomena that make guidance distinctive are its tentativeness, openness, and flexibility. As Jeremy Kessler and Charles Sabel have argued, those characteristic features of guidance are a response to profound problems of complexity, fallibility, and uncertainty that attend contemporary governance: ‘Guidance is the kind of law that uncertainty admits’ (Kessler and Sabel 2021, p. 191). US administrative law has migrated over time from focus on case-by-case adjudication—with its concern for fair hearing and impartial process—to legislative rulemaking—with its emphasis on public participation—to guidance—with its specialty in iterative learning and adaptive management. Ever increasing layers of social complexity arguably demand such nimble responses. Kessler and Sabel’s analysis seems mostly focused on issues of technical uncertainty that require guidance as an interim regulatory approach, developed in close coordination with regulated parties. This chapter shifts attention here to the specifically moral and political uncertainty that guidance interacts with. As a moral matter, guidance provides private parties with an official understanding of what their otherwise genuine obligations consist in, in cases where there is significant uncertainty about how to comply with such obligations. Guidance can thus shift moral reasoning, even where it lacks the power to bind, and even if its coercive force is minimal. As a political matter, guidance at once responds to and invites ideological and social conflict. It makes no claim to be authoritative and thus opens itself to continual challenge and rejection. It is therefore a natural response to politically fragmented landscapes where shared sources of authority are in short supply. But because guidance falls short of a claim to valid binding effect, it risks further undermining the already weak foundations of administrative legitimacy. 4.1

Guidance’s Moral Impact

Guidance often deals with the many mundane issues that are the stuff of administrative law, such as how obtain a construction permit, conduct a scientific test, or comply with a design or performance standard. Even these dry instances of bureaucratic papermaking often have moral dimensions (by ‘moral dimensions’, I mean features that are relevant to the content, implications, and performance of our genuine obligations, as distinct from solely legal, customary, or social obligations). For example, the USDA administers an organic-certified food program. Guidance documents explain how parties should make appropriate organic food labels (USDA 2014). Consumers may value organic food for moral reasons having to do with preserving their own health or the health of others or avoiding harm to animals or to the environment. Likewise, food producers have a genuine obligation not to mislead or deceive consumers as to

246  Research handbook on soft law products they will ingest. If the USDA offers useful guidance about labelling, it will help both consumers and producers comply with their otherwise binding moral obligations. Consumers will be able to purchase the organic goods that arguably advance various moral ends, and producers will know how to convey the contents of their products accurately. To take a more contentious example, consider the Obama Administration’s Dear Colleague Letter on transgender rights, which interpreted existing rules against sex discrimination by federal education grant recipients to require these recipients to let transgender students use school facilities that accorded with the students’ respective gender identities (USDE and DOJ 2016). Allow me to explain the guidance’s moral significance with a simple hypothetical that does not do full justice to the complexity of gender identity and its non-binary characteristics (Stryker et al. 2008). The guidance would instruct that a transgender person who identifies as a woman be permitted to use a women’s restroom (since previous regulations interpret sex equality to permit sex-segregated restrooms). Let us take it for granted that schools have a moral obligation not to discriminate on the basis of sex. Because sex encompasses gender identity, the government reasoned, it would be inconsistent with the obligation not to sex-discriminate for a school to deny a transgender woman access to a women’s restroom while permitting access to a cisgender woman. When a school administrator reads this Dear Colleague Letter, it gives them information relevant to determining what their underlying moral obligations not to sex-discriminate requires. It tells them that—in the view of the officials with an obligation to implement the law—sex equality requires institutionalized respect for the gender identity of trans persons amongst all others. The moral relevance of this information does not turn on an assumption that the law is morally binding simply because it is law. Rather, the guidance’s moral claims are based on several interlocking claims. First, some enacted laws do in fact reflect and specify existing moral commitments. They pick out from a number of different ways of implementing or interpreting moral obligations one particular way that we should all comply with (Greenberg 2017). Such laws thus help us to fulfil genuine obligations better than if there were no law available. Second, the considered views of officials who have the obligation to implement such moral-legal obligations deserve special weight in the moral reasoning of private parties. That is not just because these officials may have specialized knowledge. To the extent we respect the officials’ roles as authorized agents of our collectively binding commitments, we ought accordingly to privilege their views of the law’s requirements. This privileged status does not amount to deference, in which private parties must simply accept the moral position implied by official guidance. Rather, it requires people to give the guidance careful attention and, if the overall weight of reasons proves indecisive, to accept the position the guidance contemplates. Thus, a school administrator who remains uncertain about how to show respect for transgender students consistent with respect for others should know more about what to do after she receives the Dear Colleague Letter—for instance, let transgender women use the women’s restroom. Guidance can therefore have a moral impact, in the sense that it provides us with clarity about what our moral obligations and interests require. Guidance can have a moral impact on its beneficiaries as well as those whose conduct it addresses. Guidance may convey ‘official recognition’ that a person’s identity or status is worthy or respect or that the harms they have suffered deserve redress (Emerson 2019a, p. 2207). Inclusive transgender guidance thus conveys to transgender persons that the government sees them as they may understand themselves. Likewise, guidance that requires educational institutions to create offices to protect against sexual discrimination, harassment, and

Administrative guidance in the United States  247 assault acknowledges that these are genuine problems within educational institutions which require systematic response (USDE-OCR 2011). Because the guidance issues from officials entrusted with the authority to interpret and administer law, it carries with it the badges of trust, respect, and good faith that attach to the official role. 4.2

Guidance’s Political Impact

The foregoing indicates that guidance informs moral reasoning by clarifying existing obligations and recognizing the moral status of persons. This is not to say that guidance invariably plays this constructive role. Guidance may trigger adverse reactions, objections, and even categorical rejection. This may happen because the guidance simply misreads what moral obligations require. For instance, guidance that advises big cat owners to build fences only four feet high when eight feet is in fact necessary would actually impair cat owners’ performance of obligations of secure containment. Or private parties may not take the guidance genuinely to inform their conduct because they do not trust or have doubts about the legitimacy of the office that issues it. They may therefore ignore the guidance or even take it as an example of what not to do. In situations where private parties’ practical orientation towards the guidance, the issuing office, and the underlying norms they enforce diverge from one another in such ways, the moral impact of guidance may heighten political controversy. Because guidance deals with such morally fraught issues in an advisory rather than authoritative fashion, it invites contestation and critique. That is guidance’s benefit as well as its risk. While some may believe that the Dear Colleague Letter regarding transgender rights appropriately clarifies our existing obligation not to discriminate on the basis of sex, others may take the position that it gets the matter badly wrong. This latter group is apt to react by rejecting the guidance—not only in its terms, but also perhaps by rejecting the very effort to be told by way of guidance how they should act. Guidance may thus promote moral clarity for some even as it produces a sense of moral indignation and political alienation for others. There is good reason to believe, for instance, that the Obama Administration’s guidance on transgender rights has rendered the issue of bathroom use a flashpoint in the so-called ‘culture-wars’ between left and right as to traditional versus progressive conceptions of gender identity and gender relations. As a Washington Post headline succinctly put it, ‘[s]ome embrace Obama’s transgender directive, others vow to fight’ (Brown et al. 2016). To be sure, the same sort of moral factionalism might have arisen from a binding rule as well as from a nonbinding guidance document. But precisely because guidance does not make any claim to control conduct, but only to suggest, advise, and delimit, it by its terms invites political rather than legal resistance and contestation to a greater degree than rules with the force of law do. The public may interpret reliance on guidance rather than rules to mean that the issuing office is unsure of what they are doing, or unable to set down and justify general binding norms. That condition of ethical and empirical uncertainty may justify and is likely to catalyze probing questions, resistance, and adverse stances from members of the public who are not positively disposed towards the tentative positions the guidance states or the underlying moral claims it articulates. The beneficiaries of guidance are therefore likely to land on the threshold between exclusion and inclusion, protection and vulnerability, that corresponds to guidance’s ambivalent legal status. The recipient of deferred deportation action, for instance, does not obtain the status of citizenship or other entitlement, but rather only a ‘liminal’ legal

248  Research handbook on soft law status—a temporary reprieve from deportation, liable to shift with changing political priorities (Chacón 2015). The relationship between nonbinding guidance and political conflict has been on particularly vivid display during the COVID-19 pandemic. The CDC and various state and local agencies have repeatedly issued guidance on issues regarding the use of masks, vaccines, social distancing, and isolation after exposure or infection (Wiley et al. 2021). The country has been awash in a sea of guidance throughout this pandemic, as have EU member states (Eliantonio et al. 2021). For those who trust the issuing offices, or who trust their scientific expertise, such guidance helps to clarify a very important moral obligation: to protect others from harm and death. But for those who do not trust the issuing offices, the guidance becomes a rallying point for moral opposition and partisan resistance. For instance, when the CDC recommended that unvaccinated people wear masks indoors, some state and local governments implemented this guidance with binding mandates, while others not only ignored it but registered strenuous political objections to the guidance (Vestal 2021). These objections may sometimes have to do with factual disagreement about how best to incentivize vaccination. They may also involve the spread of misinformation and conspiracy theories which now play an outsized role in American politics. But they also bear the marks of a highly partisan dispute about governmental regulation of individual conduct, and about the degree to which one person ought to take into account the interests and welfare of others in their own behaviour. The situation has been exacerbated in the US by frequently shifting guidance and, at times, inconsistency between guidance issued by the White House and guidance issued by the CDC on issues such as social gatherings (Wiley et al. 2021). In part these inconsistencies are attributable to the difficulties of responding to a new and changing disease, as well as evolving scientific understanding of the disease. Political controversy then layers over and feeds upon the underlying medical and scientific difficulty. Economic pressures to open up businesses, social pressure to return to in-person schooling, pressure from teachers’ unions to keep schools closed, religious pressure to allow in-person worship, diverse risk levels for various populations, widely divergent knowledge and views about masks and vaccines—all of these factors impinge on the issuance of guidance by multiple actors at each layer of government. In that situation of crisscrossing and varying beliefs and interests, the moral impact of guidance is likely to become less certain, because private parties will be unsure what to do or who to trust. More broadly, as inconsistencies across time and institution proliferate, citizens’ faith in government institutions and in one another’s moral judgments may further erode. People may build personal identities around wearing a mask or not, or getting vaccinated or not, in ways that are untethered from scientific judgment or genuine moral obligations, and instead align with stark partisan divisions. As a form of law that is in its nature uncertain and flexible, guidance is at once indispensable in these sorts of situations and liable to deepen the crises of legitimacy that call for such informal governance strategies. Because it is easy to issue and amend, and because it does not legally require people to take specific actions, guidance is indispensable during periods of emergency, in which events unfold quickly and the government must have means of speedily advising the public about how to respond. At the same time, guidance’s posture of doubt and openness leaves space open for existing disagreements and conflicts to coalesce and fester. By disclaiming the authority to control, guidance may underscore the impression that the state is uncertain over whether its ‘monopoly of the use of physical force’ is indeed ‘legitimate’ (Weber [1918] 1946, p. 78). That is a particularly troubling risk in the US, where there are

Administrative guidance in the United States  249 increasingly mainstream arguments that administrative law as a whole is unconstitutional (Hamburger 2014). In a society with high levels of moral and political consensus, the uncertainty of guidance might not pose a serious problem. People would consider the advice given and accept it or not depending on a set of moral and political judgments they otherwise share. But in politically fractured societies such as the US, the necessary resort to informal and flexible action may simply buttress the existing and increasingly pervasive sense that the administrative state’s legitimacy is open to question. ‘If you, the government, knew what to do, why did you not actually tell us what we must do? Why do the requirements keep changing? Is it really science that is driving these decisions, or just politics?’ Such questions are invited by a form of government action that does not make the claim to bind, but explicitly permits departure. Guidance’s formal uncertainty and openness compounds deeper uncertainties about whether common and neutral ground exists on which to take collective action. This is not to suggest that guidance should not be used. In risky and fast-paced situations there may be no alternative. Rather, the necessary resort to guidance may be symptomatic of a society beset by such empirical and normative complexity that it has lost the capacity to reach binding judgment and instead spirals into ever deeper levels of self-doubt and recrimination.

5. CONCLUSION This chapter has examined the legal framework, normative tensions, and political controversies surrounding guidance in US administrative law. Guidance is a nonbinding form of law that need not be issued through the formal process required for legislative-type regulations. Such documents are beneficial for liberal democracy insofar as they help to provide fair notice of the government’s enforcement position and help to ensure hierarchical supervision by politically accountable agency heads. But they may also threaten liberal democracy by coercing private parties in an informal, opaque manner, and by avoiding public participation requirements. Guidance may, further, help to clarify pre-existing moral obligations. But because guidance is at once tentative and coercive, it invites political resistance, contestation, and fragmentation. This may prove perilous where, as in the United States, the administrative state’s legitimacy is already open to question.

REFERENCES 65 FR 56468., (2000). Administrative Practices and Procedures; Good Guidance Practices. Federal Register. 65(182), 56468–56480. 72 FR 3432., (2007). Final Bulletin for Agency Good Guidance Practices. Federal Register. 72(16), 3432–3440. 84 FR 55235., (2019). Promoting the Rule of Law Through Improved Agency Guidance Documents (Executive Order 13891). Federal Register. 84(199), 55235–55238. ACUS., (2017). Administrative Conference recommendation 2017-5: agency guidance through policy statements. 82 FR 61734. 82(249), 61734–61738. ACUS., (2019). Administrative Conference recommendation 2019-1: agency guidance through interpretive rules. 84 FR 38927. 84(153), 38927–38930. Anthony, R. A., (1992). Interpretive rules, policy statements, guidances, manuals, and the like—should federal agencies use them to bind the public? Duke Law Journal. 41(6), 1311–1384.

250  Research handbook on soft law Bagley, N., (2019). The procedure fetish. Michigan Law Review. 118(3), 345–401. Beals, M., (2021). Fauci defends new CDC isolation guidelines [online]. The Hill. [Viewed 5 April 2022]. Available from: https://​perma​.cc/​K3TW​-XNBF. Brown, E., Balingit, M., and Larimer, S., (2016). Some embrace Obama administration’s transgender directive, others vow to fight [online]. The Washington Post. [Viewed 21 February 2022]. Available from: http://​wapo​.st/​1OortaZ. Case No. 73-1766 United States v. Nixon, 418 U.S. 683 (1974). Case No. 91-1502 Franklin v. Massachusetts, 505 U.S. 788 (1992). Case No. 95-2571 Hoctor v. U.S. Department of Agriculture, 82 F.3d 165 (7th Cir. 1996). Case No. 97-7005 Paralyzed Veterans of America v. D.C. Arena L.P., 117 F.3d 579 (D.C. Cir. 1997). Case No. 99-5305 James V. Hurston Associates, Inc. v. Glickman, 229 F. 3d. 277 (D.C. Cir. 2000). Case No. 12-5310 National Mining Association v. McCarthy, 758 F.3d 243 (D.C. Cir. 2014). Case No. 18-587 Department of Homeland Security v. Regents of the University of California, 140 S. Ct. 1891 (2020). Chacón, J. M., (2015). Producing liminal legality. Denver University Law Review. 92(4), 709–767. Chambers, C., (2021). Agencies are rescinding guidance regulations at a rapid pace [online]. Washington, D.C.: George Washington University Regulatory Studies Center. [Viewed 5 April 2022]. Available from: https://​perma​.cc/​PTZ7​-RN6E. DHS., (2012). Memorandum from Janet Napolitano, Secretary of Homeland Security, to David V. Aguilar, Acting Commissioner, US Customs and Border Protection, et al. [online]. Washington, D.C.: US Department of Homeland Security. [Viewed 8 April 2022]. Available from: https://​perma​.cc/​ 5G4N​-VNRZ. EEOC., (2012). Enforcement guidance on the consideration of arrest and conviction records in employment decisions under title VII of the Civil Rights Act [online]. Washington, D.C.: US Equal Employment Opportunity Commission. [Viewed 8 April 2022]. Available from: https://​perma​.cc/​ STX5​-7BSU. Eliantonio, M., Korkea-aho, E. and Vaughan, S., (2021). EJRR special issue editorial: COVID-19 and soft law: is soft law pandemic-proof?. European Journal of Risk Regulation [online]. 12(1), 1–6. [Viewed 5 April 2022]. Available from: doi: 10.1017/err.2021.1. Elliott, E. D., (1992). Re-inventing rulemaking. Duke Law Journal. 41, 1490–1496. Emerson, B., (2019a). The claims of official reason: administrative guidance on social inclusion. Yale Law Journal. 128(8), 2122–2216. Emerson, B., (2019b). The public’s law: origins and architecture of progressive democracy. New York: Oxford University Press. Emerson, B. and Levin, R. M., (2019). Agency guidance through interpretive rules: research and analysis [online]. Washington, D.C.: Administrative Conference of the United States. [Viewed 5 April 2022]. Available from: https://​perma​.cc/​E4YX​-Q455. Freedman, J. O., (1978). Crisis and legitimacy: the administrative process and American government. Cambridge, UK: Cambridge University Press. Galbraith, J. and Zaring, D., (2014). Soft law as foreign relations law. Cornell Law Review. 99(4), 735–794. Gersen, J. E. and Posner, E. A., (2008). Soft law: lessons from congressional practice. Stanford Law Review. 61(3), 573–628. Gersen, J. and Vermeule, A., (2016). Thin rationality review. Michigan Law Review. 114(8), 1355–1412. Greenberg, M., (2017). The moral impact theory, the dependency view, and natural law. In: G. Duke and R. P. George, eds. The Cambridge companion to natural law jurisprudence. Cambridge: Cambridge University Press. pp. 275–313. Grisinger, J. L., (2012). The unwieldy American state: administrative politics since the New Deal. Cambridge: Cambridge University Press. Habermas, J., (1975). Legitimation crisis. Translated by T. McCarthy. Boston, M.A.: Beacon Press. Hamburger, P., (2014). Is administrative law unlawful?. Chicago: University of Chicago Press. Kant, I., (1797). Metaphysical elements of justice: part I of the metaphysics of morals. Translated by J. Ladd. Reprint, Cambridge: Hackett Publishing, 1999. Kessler, J. and Sabel, C., (2021). The uncertain future of administrative law. Daedelus. 150(3), 188–207. Klabbers, J., (1996). The redundancy of soft law. Nordic Journal of International Law. 65(2), 167–182.

Administrative guidance in the United States  251 Korkea-aho, E., Havu, K., Leino-Sandberg, P. and Tarkkila, L., (2021). The Finnish soft law puzzle: how can a rule-bound parliamentary democracy be reconciled with EU soft law? In: M. Eliantonio, E. Korkea-aho and O. Ştefan, eds. EU soft law in the member states: theoretical findings and empirical evidence. Oxford, UK: Hart Publishing. pp. 119–134. Levin, R. M., (2018). Rulemaking and the guidance exemption. Administrative Law Review. 70(2), 263–356. Mantel, J., (2009). Procedural safeguards for agency guidance: source of legitimacy for the administrative state. Administrative Law Review. 61(2), 343–406. Mashaw, J. L., (2018). Reasoned administration and democratic legitimacy: how administrative law supports democratic government. Cambridge, UK: Cambridge University Press. Mendelson, N. A., (2007). Regulatory beneficiaries and informal agency policymaking. Cornell Law Review. 92(3), 397–452. Metzger, G. E., (2015). The constitutional duty to supervise. Yale Law Journal. 124(6), 1836–1933. Metzger, G. E. and Stack, K. M., (2017). Internal administrative law. Michigan Law Review. 115(8), 1239–1307. Motomura, H., (2014). Immigration outside the law. New York: Oxford University Press. Parrillo, N. R., (2019). Federal agency guidance and the power to bind. Yale Journal on Regulation. 36(1), 165–271. Rose-Ackerman, S., (2021). Democracy and executive power: policymaking accountability in the US, the UK, Germany, and France. New Haven: Yale University Press. Sant’Ambrogio, M. and Staszewski, G., (2021). Democratizing rule development. Washington University Law Review. 98(3), 793–856. Seidenfeld, M., (2011). Substituting substantive for procedural review of guidance documents. Texas Law Review. 90(2), 331–394. Shepherd, G. B., (1996). Fierce compromise: the administrative state emerges from new deal politics. Northwestern University Law Review. 90(4), 1557–1683. Strauss, P. L., (2001). Publication rules in the rulemaking spectrum: assuring proper respect for an essential element. Administrative Law Review. 53(3), 803–850. Stryker, S., Currah, P. and Moore, L. J., (2008). Introduction: trans, trans-, or transgender. Women’s Studies Quarterly. 36(3/4), 11–22. Sunstein, C. R., (1983). Deregulation and the hard-look doctrine. The Supreme Court Review. 1983, 177–213. United States Department of Justice., (1947). Attorney General’s manual on the Administrative Procedure Act. Washington, D.C.: Government Printing Office. USDA., (2014). Labeling organic malt beverages [online]. US Department of Agriculture, National Organic Program, Agricultural Marketing Service. [Viewed 8 April 2022]. Available from: https://​ perma​.cc/​6784​-YXQP. USDE-OCR., (2011). Dear colleague letter from Assistant Secretary for Civil Rights, Russlyn Ali [online]. Washington, D.C.: US Department of Education, Office for Civil Rights. [Viewed 8 April 2022]. Available from: https://​perma​.cc/​8RKM​-7LQB. USDE and DOJ., (2016). Dear colleague letter from Assistant Secretary for Civil Rights, Catherine E. Lhamon and Principal Assistant Attorney General for Civil Rights, Vanita Gupta. Washington, D.C.: US Department of Education and US Department of Justice. [Viewed 8 April 2022]. Available from: https://​perma​.cc/​3BHZ​-2QP3. Vestal, C., (2021). Compliance and defiance: states react to new CDC mask guidelines [online]. PEW Stateline. [Viewed 7 April 2022]. Available from: https://​perma​.cc/​R543​-6KAV. Wagner, W. E., (2010). Administrative law, filter failure, and information capture. Duke Law Journal. 59(7), 1321–1432. Weber, M., (1919). Politics as a vocation. Munich: Duncker & Humblot. Reprint, In: H. H. Gerth and C. Wright Mills, eds. From Max Weber: essays in sociology. New York: Oxford University Press, 1946. pp. 77–128. Weil, P., (1983). Towards relative normativity in international law. American Journal of International Law. 77(3), 413–442. Wiley, L. F., Yearby, R. and Hammond, A., (2021). United States: legal response to Covid-19. In: J. King and O. L. M. Ferraz et al., eds. The Oxford compendium of national legal responses to Covid-19

252  Research handbook on soft law [online]. Oxford, UK: Oxford University Press. [Viewed 8 April 2022]. Available from: doi: 10.1093/ law-occ19/e24.013.24.

17. Soft law making at the European Commission: not much of a one-institution show Oana Ştefan

1. INTRODUCTION The first soft law instruments in European integration were the ‘Christmas notices’ from 1962, adopted by the Commission in the field of competition law (Senden and Prechal 2001, p. 181). Faced with an important number of requests to clear agreements, the Commission issued these notices to clarify the way in which it intended to interpret the antitrust provisions of the Treaty in agency contracts (European Commission 1962a) and patent licenses (European Commission 1962b). The instruments were thus issued both in the interest of transparency and in the interest of enforcement efficiency and decreasing the workload of the Commission in this regard. As such, they served the function of ‘regulation by information’ (Hofmann 2006, pp. 169–170) or ‘regulation by publication’ (Snyder 1994, pp. 199–201), allowing the Commission to explain the existing EU (hard) law, to present its own views on the law, and clarify those provisions of an open and indeterminate character (Gardeñes Santiago 1992, pp. 939–940). Until 1968, recourse to soft law instruments was exceptional. Afterwards, and especially after the accomplishment of the internal market, they became quite frequent in the activity of the European Commission (Leclerc 1998, p. 163). Classic examples include the Communication on the Cassis de Dijon judgment (European Commission 1980), restating the principle of mutual recognition previously laid down by the European Court of Justice (ECJ), or the 1985 White Paper on the Internal Market (European Commission 1985, para. 155), which announced the increased reliance on soft law instruments. Ever since, the Commission has been issuing a vast variety of instruments, going beyond the strict taxonomy of Article 288 of the Treaty on the Functioning of the European Union (TFEU 2012), which only refers to recommendations and opinions as non-binding EU legal sources. In their interpretative form, these instruments resemble the guidance issued by European national administrations, showing perhaps the maturity of the EU legal order. However, the purpose of other Commission soft law exceeds the interpretative goal sought by the first notices, and are sometimes part of structured governance processes, such as the Open Method of Coordination (OMC) (see also van Gerven and Stiller in this volume), or the post-accession Cooperation and Verification Mechanism (CVM). Commission soft law has not yet been quantified by academics, although work on this is carried out in certain sectors of European integration. An ongoing study shows that the Commission is the main producer of soft law in areas of EU exclusive competence, whilst the body of such instruments is more evenly spread between the various institutions, agencies, and technical bodies in areas of shared competence (Cappellina et al. 2022). Commission soft law reaches very far suggesting that it is best suited to deal with the complexity of European affairs, their diversity (Schäfer 2006, p. 84), and the bargaining constraints 253

254  Research handbook on soft law of sovereignty between the Union and the Member States (Sisson and Marginson 2001). The flexibility of soft law and its simplified adoption procedures are beneficial in sensitive sectors in which agreements are very hard to reach, in tackling issues of regulatory philosophy and broad policy, such as economic policy coordination (Hodson and Maher 2004, pp. 810–811), and in addressing situations where swift action is imperative, such as competition law (Cini 2001, p. 194; Cosma and Whish 2003, p. 33). In migration, the Commission has resorted in recent years to soft law, as an answer to the increased legislative costs associated with hard law making to substantiate a return policy (Slominski and Trauner 2021). During the COVID-19 pandemic, soft law provided a fast regulatory response to attenuate the effects of the lockdowns or to manage the spread of the virus (Ştefan 2020). The downside of this flexibility is a perceived lack of legitimacy of soft law and the potential for soft law to constitute backdoor legislation. Such potential is important, given that, as shown by empirical research, soft law is used at the national level by administrators and even judges (Eliantonio et al. 2021). Yet, we know little about how soft law is made, given that its flexibility means also that it escapes streamlined procedures. The regulation by publication method appears to suggest that only the end product will ever reach the public domain, whilst the laboratory of soft law making is tucked well within Commission’s own directorates, units, and services. Acknowledging that systematic empirical research is needed to unpack all this alchemy – which might very well be field specific – this chapter aims at gauging the broad lines of the procedural side of the soft law issued by the European Commission. Is the Commission performing a ‘one-institution show’ when issuing its soft law? By reviewing publicly available data such as legislation, case law, and the existing academic literature, the chapter unveils that Commission soft law emerges out of complex interactions and long-term relationships with EU institutions, national authorities, and stakeholders. Whilst most of these relationships are public, they are difficult to unpack, raising questions of accountability and transparency. The hope is that understanding these relationships might solve the ‘classic conflict between problem-solving/efficiency and democratic accountability’ of which EU soft law is a symptom in EU integration (Mörth 2021). The chapter proceeds as follows. Section 2 delves into general questions regarding the legal basis and the procedure of soft law making. Section 3 looks at the various inter-, intra-, and extra-institutional interactions at play in soft law making. Section 4 concludes.

2.

HOW TO FIT A SHAPE-SHIFTING PEG IN A SQUARE HOLE: APPLYING GENERAL EU LAW MAKING ORTHODOXY TO SOFT LAW

Referring specifically to Commission guidance documents, Van Dam (2017) reviews previous classifications of soft law and identifies five categories: interpretative guidance, that is soft law that further clarifies vague hard law; implementing guidance, such as recommendations indicating the best ways for the Member States to implement EU hard law; explanatory guidance, which explains what already is in the law and the case law of the courts; technical guidance, advising on technical measures and standards required by hard law; and the dissemination of good practices. The Commission soft law output was documented to have increased over time, especially in relation to interpretative instruments (also known as ‘complementary’ according to Zhelyazkova et al. 2015).

Soft law making at the European Commission  255 Yet, the boundaries between all these categories are rather fluid. Take for instance two ‘classics’ – the Cassis Communication and the White Paper on the Completion of the Internal Market. The former, a ‘communication’, is, in theory, supposed to interpret hard legislation, whilst also containing rules of conduct for the institutions and the Member States. Conversely, ‘White Papers’ are only ‘programmatic’ in that they set out courses of action, timetables, or action plans for the EU (Senden 2004, p. 126). Such categorization offers a quite narrow perspective on these two instruments, as the Cassis Communication contains both regulatory elements and elements of a planning nature. In its regulatory content, the Communication generalizes the principle of mutual recognition and states that an absolute prohibition on sale of products lawfully produced and marketed in another Member State would fail the proportionality test. Other paragraphs of the Communication are programmatic, announcing the intentions of the Commission regarding its harmonization work. What is more, the Cassis Communication and the White Paper on the Internal Market cannot be considered in isolation from each other and from their context. The Cassis Communication only summarized and generalized the Commission’s submissions in the Cassis de Dijon case, effectively introducing the Commission’s new approach to harmonization (European Commission 1980), whilst the White Paper further elaborated on this basis the plans for the internal market, including projects such as some 300 future legislative proposals (European Commission 1985). Even if, for the purpose of our exercise, we were to structure the soft law of the Commission according to one taxonomy or another, it is hard to say that each category would follow a clearly set rules of procedure. Flexibility in decision making appears to be the ground norm, and such flexibility stems directly from the non-legally binding character of the norms. There is no official EU Law of administrative procedure, as academic (ReNEUAL 2014) or Parliamentary initiatives (European Parliament 2016) have not borne fruit and have been resisted by the European Commission. Even so, it might be inappropriate for such rules to formally cover soft law rulemaking, although the Commission might be encouraged to have a principled approach both for its hard law and soft law making (Curtin et al. 2014). Publication is not necessarily a prerequisite for non-binding instruments, despite the wonderfully coined term of ‘regulation by publication.’ Similarly, choosing a particular legal basis or a specific competence for the adoption of soft law appears to be optional at best. 2.1

‘Regulation by Information’ but not Necessarily by ‘Official Publication’

Soft law does not have to be published by the European Commission, as Article 297 TFEU provides a publication obligation only for binding acts. However, the Commission usually published its guidance documents in the ‘C’ series of the Official Journal, and nowadays it is expected that Commission soft law would be available online. There is no obligation to translate soft law in all official languages (which is another point that talks to the flexible character of these instruments). In Polska Telefonia, the ECJ decided that Commission guidance addressed national regulators, and because it did not produce binding effects for individuals, it did not have to be translated. A Polish regulator could, therefore, apply such guidance published only in English to impose regulatory obligations on Polish undertakings (C-410/09 Polska Telefonia Cyfrowa, paras 31–34). Yet, this begs the question of whether consistent, well translated versions of soft law should be published in the interest of transparency in all EU languages, so that national authorities do not struggle with deciphering the meaning of guidance meant to assist in the application of EU law.

256  Research handbook on soft law The effectiveness of soft law can however be undermined by conflicting, or ambiguous language versions. In relation to the Recommendation on online gambling, for instance, it emerged in the proceedings of Belgium v Commission (C-16/16 P) that the different language versions contained terms suggesting various degrees of coercion, with the German, Spanish and Dutch versions drafted in the most mandatory terms. Whilst this is a point that is currently being researched in the academic literature (Andone and Greco 2018, p. 79), we note that neither the Court, nor the Advocate General seem to be preoccupied by such variation. Láncos (2019, p. 10) suggests that such interpretation of the reality is overly simplistic and reflects that the Court is not consistent in comparing language versions across the board. Indeed, if national authorities were to apply soft law measures linguistic differences might impact on the uniform interpretation of EU law. Beyond language versions and publication requirements, the drafting of soft law raises concerns. The Joint Practical Guide on the drafting of European Union legislation (European Commission 2016, point 17) offers extremely limited guidance in relation to non-binding instruments. It encourages the drafters not to use imperative forms or structures too close to that of a binding act (European Commission 2016, point 2.3.3), but this advice seems to be ignored in practice. Following a qualitative content analysis of soft law instruments issued in environmental law, Petropoulou Ionescu and Eliantonio (2023) note that the overwhelming majority employ ‘strongly prescriptive language that seems to leave little margin of choice for the addressees’.1 This creates some sort of incoherence between the non-binding character of the instrument, as laid out in the Treaty, and the prescriptive tone thereof, which might in turn create effectiveness problems at the national level. In this context, Andone and Coman-Kund (2022) argue that the key to an effective application of Commission soft law does not lie only in creating soft mechanisms of enforcement, such as reporting and monitoring, but also depends on the argumentation employed at the drafting stage. In particular, the Commission should explain more why a certain approach outlined in its soft law instruments is beneficial (Andone and Coman-Kund 2022, pp. 37–45). However, one may wonder whether delving into lengthy explanations is practical for instruments essentially meant to provide clear, concise interpretations of EU law and to help national authorities in their work. 2.2

Legal Basis and Commission Competence

The lack of legally binding force for EU soft law might also mean that the Commission can issue such instruments without being constrained by the list of competences provided in the TFEU. This point has been subject to academic controversy. As argued by Dawson (2017), soft law cannot ‘be controlled by a system of enumerated competences’, simply because it carries out a power different than that of hard law, in Foucauldian terms ‘bio-power’. As such, soft law does not aim at exercising legal power, yet it carries coercive effects for actors (Dawson 2017, p. 245). Conversely, Senden (2004, pp. 296–301) pleaded for the application of the principle of conferral to soft law, for rule of law reasons, as well as reasons related to the institutional balance. She considered that, for soft law, this principle materializes in an obligation to identify the correct legal basis for adopting certain instruments (be it in primary or secondary legislation). The legal basis of the Commission competence to adopt soft law can be traced to Article 17(1) of the Treaty on European Union (TEU 2008), providing in its first paragraph sweep-

Soft law making at the European Commission  257 ingly that ‘[t]he Commission shall promote the general interest of the Union and take appropriate initiatives to that end’. Another express legal basis is Article 292 TFEU, mentioning that ‘[t]he Commission, and the European Central Bank in the specific cases provided for in the Treaties, shall adopt recommendations’. It needs to be observed that the article only limits the ECB competence to adopt recommendations in specific cases, allowing the Commission a wide discretion in this regard. Other than this, multiple sectoral legal bases refer to the power of the Commission to adopt soft law, in matters related to the liberalization of services beyond the extent required in secondary legislation (Article 60 TFEU), or to encourage cooperation in public health (Article 168(2) TFEU). Other articles do not make specific reference to a soft law instrument, but have been considered as a legal basis in practice, such as Article 108 TFEU, requiring cooperation between the Member States and the Commission in State aid. Another source for specific legal bases is, of course, secondary legislation, which sometimes provides for follow-up Commission soft law. Such references in hard law should not alter the non-binding character of soft law instruments (European Commission 2016, point 17). Andone and Greco (2018) argue that the Commission ‘evades its burden of proof’ as it hardly references in its recommendations the legal basis employed or justifies its actions in relation to subsidiarity or proportionality. Furthermore, they observe that when the Commission does justify its soft law, it does so with arguments related for instance to the uniform application of the law, or the necessity to solve a regulatory problem. However, the Commission only enounces these justifications, and it does not go the extra mile to detail whether such goals are shared also by the Member States, or whether a regulatory problem exists. Whilst this might be an issue from a linguistics and argumentation perspective, it does not necessarily mean this is also an issue from a legal point of view. Brief justifications in the preamble that refer to principles of European law, such as ensuring transparency, or consistency, are generally accepted as part of the rule of law and of the constitutional traditions of the Member States. Requiring the Commission to further substantiate these claims is a very tall order, exceeding any reasonable expectations. This would only create supplementary hurdles which would get in the way of ensuring the much-sought flexibility of soft law instruments. Pandemic regulation offers a perfect example of why justifying a legal basis and a competence might defeat the very purpose of soft law instruments. Extreme urgency meant that interpretations of hard law were offered through multiple channels, possibly also as a reflection of the various overlapping competences in the field. Regulating for instance the use of data to control the spread of the virus post lockdown straddled between established EU policies (such as data protection or cross-border healthcare provision) and the unexplored territory of Article 168(5) TFEU (combatting major cross-border health scourges). Initially, the Commission adopted a recommendation on a common EU toolbox for the use of technology and data to address the COVID-19 crisis (the Commission Toolbox) (European Commission 2020a). The instrument provides for various levels of technical details to develop such apps and for various guiding principles in order to ensure compliance with the General Data Protection Regulation (2016) or the Patients’ Rights Directive (2011). Soft law responds thus to a need of speedy regulatory reaction in areas where swift action is needed, such as emergency situations, and circumscribing it to competence or publication and drafting requirements as with hard law would remove the flexibility benefit of these instruments. In this context, Dawson (2017, p. 250) sees the expansion of soft law as a symptom that the system of EU competences might need to be rethought. Whilst this is a powerful argument,

258  Research handbook on soft law changing the Treaties in current times is almost a mission impossible, and the focus might need to shift, in the meanwhile, in ensuring the accountability of soft law. The CJEU has refused systematically to review what they consider to be ‘true’ Commission soft law – that is non-binding instruments. This has been the topic of intense criticism by prominent academics (Eliantonio 2021, p. 286) and even members of the Court (Opinion of Advocate General Bobek in case C-16/16 P Belgium v Commission), suggesting that other channels of accountability might need to be sought for Commission soft law (see also Snyder in this volume, Eliantonio and Korkea-aho in this volume). The challenge by Belgium against the Commission Recommendation on online gambling, is extremely revealing. The Commission was accused that it breached the principles of conferral and the duty of sincere cooperation. The Court rejected the claim, noting the non-legally binding force of the document (C-16/16 P Belgium v Commission). This brings to the fore the question of whether practically the Commission could pass, unconstrained by procedures and legal challenges, any act it sees fit in a non-binding form. The reality of the matter is however more nuanced. The Communication on online gambling is the product of a three-year long soft law dialogue between various institutions and stakeholders. It built on public consultations on a Green Paper on online gambling in the internal market (European Commission 2011), and the resulting Commission Communication (European Commission 2012). Some accountability channels were there, although potentially unsystematic and not necessarily transparent. The next section traces these accountability channels in the various relationships between the Commission and its interlocutors in the moment of soft law drafting and implementation.

3.

INFORMAL PROCEDURES BETWEEN FLEXIBILITY, ACCOUNTABILITY, AND LEGITIMACY

There are no streamlined procedures for the adoption of EU soft law. Article 292 TFEU provides some rules of procedure only for the Council recommendations (which need to act on a proposal from the Commission and at unanimity in those cases where the Treaties provide for such procedures). Such lack of procedures might raise legitimacy concerns which were picked up by the European Parliament (1969) and national authorities (Conseil d’État 1993, pp. 22–23) since early days. More recently, the European Parliament (2007) declared that ‘the use of soft law is liable to circumvent the properly competent legislative bodies, may flout the principles of democracy and the rule of law under Article 6 of the EU Treaty, and also those of subsidiarity and proportionality’. Such objections were dismissed as ‘inter-institutional wrangling’ (Dawson 2009, p. 5), yet, others believe that the legitimacy issue is one of the fundamental flaws of soft law (Petropoulou Ionescu and Eliantonio 2021). What is certain is that creating heavy procedures for issuing soft law defeats the purpose of these instruments by removing their flexibility. The delicate balance between ensuring the legitimacy of soft law making and preserving its regulatory functions is difficult to strike. Perhaps for this reason, soft law emerges most of the times out of collaborations. Such collaborations are not new, with Leucht (2021, p. 93) tracing back the interactions between various officials the Commission, the Court, and the European Parliament, which, through informal instruments such as letters and communications, developed the new approach to internal market harmonization. In 1978, in the period between the submission of written obser-

Soft law making at the European Commission  259 vations in the Cassis de Dijon case (C-120/78), but before the oral hearing, the Commission circulated a letter to all the Member States in which it outlined that the common market agenda was under threat by various national protectionist measures. In the letter, the Commission was also outlining several plans to tackle these restrictions, preparing the ground for further action to dismantle protectionism. This was the genesis of the famous Cassis Communication, which rested on exchanges between Commission officials and directorates, also supported by debates within the European Parliament (European Commission 1980). As evidenced by Leucht (2021, p. 88), the 1978 letter of the Commission has been always the initial point for all these exchanges, at the Commission level or the European Parliament. Crucially, such debates occurred in the year of the first direct elections to the European Parliament, 1979, from Members eager to enhance their legitimacy by backing up European Commission’s initiatives (Leucht 2021, p. 96). The existence of a parliamentary debate on these issues is significant, given the critical position generally embraced by the European Parliament towards soft law. This historical account shows that the Commission seldom acts alone in the process of issuing of soft law, yet this has not been explored enough in the academic literature. This section is meant to provide an overview of what future research might want to investigate from this perspective, outlining five sets of interactions instrumental in the making and the transformation of EU soft law. First, one needs to acknowledge the interactions between the various Commission internal structures, that is, the intra-institutional interactions. Second, the soft law of the Commission is born from inter-institutional exchanges, between the Commission and the other EU institutions, such as the European Parliament itself, or EU agencies. Third, the Commission closely collaborates with the Member States in the drafting of soft law, often in areas of exclusive competence such as State aid. Fourth, many soft law instruments are submitted to public consultations. Fifth, soft law is never static, and it transforms following its application at the national level, which constitutes potentially the most important accountability channel for such instruments. 3.1

Intra-institutional Exchanges at the Level of the European Commission

As documented in relation to the Cassis Communication, the drafting of soft law starts at the level of the Commission directorates and services (European Commission 1980). The documents are generally prepared by the various units in the relevant Directorates-General (DGs), but we have little information of how this drafting takes place or who is instructed to take charge of specific instruments. Van Dam (2017, p. 88) observes that guidance with a technical nature is generally prepared by the Joint Research Centre, the European Commission’s scientific and knowledge service. Other examples can be gauged by tracing particular versions of various soft law. On the backdrop of the travel difficulties generated by the pandemic, a spontaneous Information on the Package Travel Directive in connection with the COVID-19 was issued by the DG for Justice and Consumers (2020). The main part of the document is interpreting the term ‘unavoidable and extraordinary circumstances’ of Article 12 of the Package Travel Directive (2015) and confirming that COVID-19 could ‘usually’ qualify as such. This document is difficult to place within usual soft law categories, as it contains a disclaimer that it was, unsurprisingly, neither formally adopted but also, rather puzzlingly, ‘not endorsed’ by the European Commission. However, it did clarify the interpretation of rules of conduct, and it did serve a transparency purpose, although in the crudest of its forms, i.e., explaining the law to consumers.

260  Research handbook on soft law The Commission followed up with a more streamlined instrument, the Recommendation on travel vouchers (European Commission 2020b), meant to offset the consequences of the pandemic for the tourism sector by making vouchers an attractive and reliable alternative to reimbursement in money for travel packages that were not consumed because of COVID. The recommendation refers to the text of the Information issued by DG for Justice and Consumers, thus rubberstamping the previous document. In relation to these instruments, the Director of the European Consumer Centre Network declared that they brought much-needed certainty over consumer rights for tourist services, despite certain gaps related to booking intermediaries or actually receiving reimbursement. This is a significative stakeholder endorsement, given that the amount of consumer protection complaints received by the Network soared during the pandemic (Marcus et al. 2021, p. 52). This example shows how soft law can evolve from weaker forms into something potentially more legitimate (or at least to an instrument actually named in the Treaty) – though the procedure through which this Recommendation was issued is not immediately apparent. 3.2

Inter-institutional Interactions in the Making of Commission Soft Law

As we have seen from the brief historical account of the Cassis Communication, the European Parliament is surprisingly not always absent from Commission soft law making. Another example is the above-mentioned Commission Recommendation on online gambling (European Commission 2014a). This was issued by the European Commission to answer several regulatory, societal, and technical challenges of this type of activity by protecting consumers and prevent minors from gambling online. The Recommendation comes on the backdrop of serious difficulties in harmonizing the gambling sector in the EU. The topic was not included in the Services Directive (2006), even though the Commission had pressed for it. Tagged as a directive-like recommendation – that is to say, Commission recommendations carrying a clause on implementation, deadlines, and a reporting mechanism (Láncos 2019, p. 10) – this Recommendation was at the core of controversy and unsuccessful legal challenge as mentioned in the concluding remarks to the section above. What is probably less known is that the Recommendation answered calls from the European Parliament to increase the safety of online gambling in the EU and raise higher awareness towards the risks of gambling addiction. Thus, in a Resolution from 2013, the European Parliament (2013) lists very specific instructions for the Commission, such as drawing up enforcement measures against illegal gambling, setting up specific controls to exclude minors from playing, limiting advertising, and so on. Of course, such example neither makes Parliamentary involvement streamlined, nor transparent, not always present. As shown by a textual analysis of the soft law issued by the Commission during the COVID-19 pandemic (Eliantonio and Ştefan 2021, p. 159), the EU Parliament is largely absent from any procedures, or at least the procedures that could be grasped by reading the various preambles of the instruments. This cannot be explained by the lockdowns in place in Europe, as the Parliament created, from late March, platforms for video-conferencing for large multilingual meetings as well as for remote voting (Feeley 2020). These conflicting two examples go to show that more research is needed before concluding that the Parliament is absent from all Commission soft law making processes, and also drawing relevant conclusions as to the legitimacy deficit of soft law.

Soft law making at the European Commission  261 One particularly controversial interaction in soft law making is the interaction with the Court of Justice. Whilst in the strictest legal terms the Court does not ‘cooperate’ with parties, such as the European Commission, when settling cases, research has shown that policy making through soft law in the EU works only if the Commission and the Court reciprocally rubberstamp their respective interpretations of hard law (Snyder 1994, pp. 199–201). Take for instance the Cassis Communication itself, which generalized the tenets of a single Court judgment beyond alcohol markets (Alter and Meunier-Aitsahalia 1994, p. 535), or the 2006 Interpretative Communication on contracts falling outside of the public procurement directives (European Commission 2006) which codifies judgments showing that such contracts might actually have cross-border interest and could be caught by the general internal market rules. As demonstrated elsewhere, the argumentation frameworks used by the Court to recognize legal effects to competition and State aid soft law, based on general principles of law, are used by the Commission as a justification for subsequent soft law instruments, in some sort of constitutionalizing virtuous circle (Ştefan 2013). The cooperation between the Commission and the agencies in soft law making is regulated in various hard law legal bases (see also Vaughan in this volume). One example can be drawn from the energy sector, in particular the drafting of network guidelines and network codes. These are essentially rules necessary to eliminate technical barriers to the flows of energy across the borders, pertaining to the way in which the grid and trade are operated. The procedure starts with the European Union Agency for the Cooperation of Energy Regulators (ACER) drawing non-binding framework guidelines in collaboration with the Commission and following consultation of the European networks of transmission system operators. The network codes and guidelines are then elaborated by the European networks of transmission system operators and should be approved by the Commission through comitology or remain non-binding soft law.2 Not all interactions are productive, of course. The recent example of the Rule of Law Conditionality Guidelines (European Commission 2020d) is an unusual yet revealing occurrence in this regard. Faced with the rule of law backsliding in the European Union, a Rule of Law Conditionality Regulation (2020) was passed back in 2020, meant to offer an additional layer of protection for the Union budget in cases where breaches of the rule of law can affect EU’s financial interests. Although the Regulation was operational as soon as it was adopted, the European Council ordered the Commission to issue application guidance (European Council 2020, para. 2(c)). This was the outcome of a tortuous budget negotiation on the backdrop of Budapest and Warsaw threatening to block the Next Generation EU package. The end result was that the Regulation was effectively blocked until the Commission finally issuing the guidance in 2022 (European Commission 2022). Such process shows that Commission soft law can sometimes delay the application of hard law, with potential consequences for the rule of law itself as documented by academics (Alemanno and Chamon 2020) and a legal challenge by the Parliament (C‑657/21 Parliament v Commission).3 Pandemic regulation also raised some challenges from the point of view of institutional cooperation, showing again the reverse side of the flexibility of soft law, bringing problems related to the fluid use of legal bases as described in Section 2. Multiple, yet not defined, legal bases means that many actors might be competent to act and adopt soft law. The Commission Toolbox on the use of data to control the spread of the virus was immediately complemented by several other instruments, issued by the Commission (European Commission 2020c) itself, the eHealth Network (2020), and the European Data Protection Board (EDPB 2020a). It

262  Research handbook on soft law appears that rather than cooperating, these bodies acted alone. The European Data Protection Board (EDPB 2020b) was planning on issuing its own guidance on apps when it received a request for feedback from the Commission in relation to a similar document the Commission prepared. Given they are issued by various bodies within a tight time span, it is difficult to establish whether some of these pandemic soft laws are a follow up to other instruments, and what are the interlinkages between them, creating a high potential for duplication. This is problematic not only in light of efficiency, vital in times of pandemic, but also in terms of coherence (Ştefan 2020). 3.3

Member States Interactions

Soft law is prepared often in collaboration with the Member States. One example is the handbook on marriages of convenience, providing for the criteria that need to be considered when assessing whether a marriage falls under this category. This was drafted, according to its preamble ‘in close cooperation’ (European Commission 2014b, cited in Van Dam 2017) with the Member States, yet no details are given as how that cooperation materialized. Such cooperation is sometimes structured within Commission’s own expert groups, and in these cases the cooperation can be gauged from the publicly available minutes of group meetings. On migration, following an invitation from the European Council to set up a European Return Programme in 2015, the Commission adopted in 2017 the revised Return Handbook (European Commission 2017a; Slominski and Trauner 2021). This was drafted, under the direction of DG for Migration and Home Affairs, more specifically Directorate C: Migration and asylum (Unit C1: Irregular migration and return), by the Contact Group Return Directive (CGRD) in three sessions, whilst involving national and EU policymakers.4 The CGRD is an informal temporary group, set up in order to exchange views with Member States experts on the application of the Return Directive (2008). However, the consultations or the negotiations for new soft law instruments might be sometimes problematic for certain countries, with smaller Member States feeling ‘neglected and ploughed over in this process’ (Aldestam 2004, p. 14). The Commission can also force acceptance of soft law through the threat of opening formal investigations into State aid measures covered by newly adopted soft law (Blauberger 2009, pp. 729–730). Of course, ‘only if a broad majority of Member States agree with its soft law can the Commission credibly threaten individual Member States and force them into a final approval’ (Blauberger 2009, p. 730). These examples show that the cooperation with the Member States in preparing soft law is neither streamlined for all instruments, nor accurately documented. One needs to plough through preambles or various meetings of Commission’s working groups to understand these relationships. Perhaps for these reasons soft law has been identified as a major source of competence creep, with the Open Method of Coordination constituting some sort of a poster child, as objectives and benchmarks appear to be set through a largely untransparent cooperation between the EU and the national executives (Garben 2019, p. 210). Against this backdrop, it appears that the ways in which cooperation occurred for crisis regulation is not necessarily exceptional, but business as usual. The Commission tailored its interpretation of the State aid rules of the Treaty to cater for the two recent emergencies, the COVID-19 pandemic and the war in Ukraine. This was done through Temporary Frameworks (TFs), presumably issued following talks with the Member States, as mentioned in various

Soft law making at the European Commission  263 press releases. In both cases, however, there is little transparency as to how these conversations took place and who was involved (Ştefan 2013, 2022). 3.4

Stakeholder Involvement

Conducting public consultations is an obligation for the Commission in accordance with Article 11(3) TEU. Accordingly, the Commission needs to conduct broad consultations with the parties concerned in the interest of coherency and transparency. Yet, in the absence of a clear recognition of participatory rights, expressly in the text of the Treaty or through the ECJ case law, the Commission remains free to organize these consultations and take them into consideration as it sees fit. As expected, the Commission appreciates flexibility in the consultation processes very much. Explaining the non-binding character of its guidance on consultations, the Commission pointed out that ‘over-legalistic’ approaches on procedure are ‘incompatible with the need for timely delivery of policy’ (European Commission 2002). Numerous studies were undertaken regarding consultations for hard law. For instance, Alemanno (2020, p. 122) notes that there is a systemic inequality in the access to public consultations, with negative consequences for the principle of political equality. There is not much empirical analysis delving into the consultations undertaken specifically for soft law making, but the consensus appears to be that the relevant procedures are unsystematic, a problem which is neither new, nor EU specific (Baldwin 1995, p. 284). The informality of stakeholder participation in OMC processes can result in an uneven representation, disproportionately favouring technocratic and bureaucratic views as opposed to substantive political or functional ones (Kröger 2007, p. 565). Consultation for interpretative guidance to environmental hard law appears to be happenstance, and untransparent procedures and outcomes are often the norm (Scott 2011, p. 336). The tide seems to change in recent years, with the Commission (European Commission 2021a) publishing a first consultation on a guidance document for the Habitats Directive (1992). This reflects potentially a new approach towards more legitimate soft law making by the Commission, in response to increased reliance on environmental guidance documents at the national level, but also to revamped better regulation initiatives at the EU level (Petropoulou Ionescu 2021, pp. 861–862). With regard to pandemic soft law, research into the first emergency instruments issued up to the Summer of 2020 shows that reference to consultations and participation was made in barely 15 per cent of the overall soft law output. Out of these, only six instruments give clear evidence that consultation of stakeholders and national authorities was undertaken, bringing the levels of citizen participation in the drafting of pandemic soft law very close to nil (Eliantonio and Ştefan 2021). Success stories do seem to exist. In (non-emergency) State aid, although procedures for consultations are not necessarily put in place, the practice is to publish all consultations online. Even so, it does not necessarily mean that transparency is fully ensured as it is difficult to understand what comments were taken on board by the Commission in the final version of guidance documents (Biondi and Ştefan 2018). This brief evaluation of soft law making shows that soft law is often born out of a dialogue between various EU institutions, as well as stakeholders. While this approach is the right one, it cannot be denied that there are shortcomings at least in relation to the transparency and the accountability of these processes. Interestingly, the European Commission has, in recent times, called for a more streamlined procedure of soft law adoption in EU agencies. In the financial sector, Marjosola (2022) observes that upon the setting up of the Lamfalussy

264  Research handbook on soft law Committees, the Commission did not set any restrictions with regards to soft law acts. The situation changed on the revision of the European Supervisory Authorities Regulations in 2017 when the Commission called for a higher standard of stakeholder consultation (European Commission 2017c), although this proposal was not retained following trilogue. Obvious solutions include to generalize the publication of the consultations, as it happens for instance in State aid. Digging deeper, however, the consultations process can be reformed too, with suggestions made along the lines of redesigning public consultations, implementing political equality by design, diversifying consultations via deliberative mechanisms, strengthening oversight, and ensuring an equalization of resources, by investment and also regulation (Alemanno 2020). Specifically for soft law, Petropoulou Ionescu and Eliantonio (2021, p. 59) believe that consultation mechanisms alone are not necessarily conducive to high level of participation and suggest being imaginative and relying on different formats stimulating interaction, reflection, and proper argument. In that regard, one solution might be the drafting of soft law through expert groups constituted by the European Commission, which subsequently consults stakeholders on drafts. This is the case of the Ethics Guidelines on Trustworthy Artificial Intelligence (AI), prepared by a High-Level Expert Group on AI (AI HLEG 2019), set up by the Commission to advise on the implementation of the Strategy on Artificial Intelligence. The Group revised their first draft considering more than 500 submissions from stakeholders, and the Commission subsequently supported the Guidelines in 2019. Another solution is to engage with the stakeholders more even after the publication of soft law instruments. As argued elsewhere (Ştefan, forthcoming), soft law gathers dynamic instruments, thriving on interactions, be these prior to its publication or after implementation. The next subsection will offer some reflections on how soft law is altered by the interactions had with the national levels of governance. 3.5

Revision/Transformation of Commission Soft Law

The flexibility of soft law instruments means that these instruments are dynamic by their definition and can be easily altered by the Commission. A good example are the instruments issued in the areas of competition and State aid, when oftentimes a revision deadline is embedded in the very text of the instrument. Take the ‘Almunia package’ of guidance within the State Aid Modernisation reform, currently being revised less than ten years after they were first issued (European Commission 2022). Soft law can be revised also if it no longer corresponds to the regulatory needs. The pandemic soft law has been altered frequently, corresponding to the changes in the spread of the disease, the rate of immunizations, and so on. Perhaps the best example in this regard is the Temporary Framework for State aid, amended six times in less than two years. Such a quick turnaround can be seen both as a sign that soft law does act as a flexible tool suitable to deal with emergency, but also as a sign that stakeholders were much less consulted, as mentioned above. Soft law is revised, or sometimes even transformed into hard law, as a response to problems encountered when applying it at the national level. This shows how soft law operates as an instrument of dialogue between various levels of governance. The de minimis notice in competition law (European Commission 2001) was revised following case law from the French courts and competition authority. The notice provided for market thresholds under which agreements between competitors were not considered harmful from the point of view of the Treaty. The notice was at the heart of a cartel case of the French Autorité de la Concurrence,

Soft law making at the European Commission  265 which discovered that the Notice allowed serious uncompetitive behaviours to go unpunished and challenged its application. Following a preliminary ruling, the ECJ sided with the Autorité de la Concurrence in deciding that agreements that have an anticompetitive object should not be allowed to use the de minimis defence (C-226/11 Expedia). The Commission followed up shortly after with a revised de minimis notice along the tenets of the Court judgment (European Commission 2014c). Soft law can be transformed into hard law following unsuccessful application at the national level, although the process tends to take time and be prompted by persistent national institutional resistance. In the telecommunications sector, a Commission Recommendation on termination rates (European Commission 2009) was applied unevenly in the Member States. The biggest resistance came from Dutch courts which criticized the regulator for relying on criteria provided for in the non-binding Commission recommendation in a series of cases which led to the landmark Koninklijke (C-28/15) preliminary reference. Conversely, the Bulgarian Supreme Administrative Court considered that the national regulator had to take into account the Commission Recommendation since the Commission was mandated, by the European regulatory framework, to issue such soft law and since such approach would be the only one respectful to the duty of sincere cooperation enshrined in Article 4(3) TFEU (Sulev 2020, p. 1562).5 This long dialogue between national authorities, national and EU courts, showing disagreement as to the application of EU soft law determined the European Commission to orchestrate a change in the legal framework (Georgieva 2021). New hard law, the European Electronic Communications Code (2018), incorporated various Commission soft law instruments, such as the main provisions of the 2009 Recommendation on termination rates. What is more, the Code mandated the Commission to adopt a delegated regulation on termination rates in order to replace this Recommendation (European Commission 2021b). This quick review of how soft law transforms shows that through multi-level exchanges various problematic issues with soft law can become apparent and can be remedied. This ranges from completely abandoning certain instruments, revising the soft law, or even the transformation of soft law into hard law.

4. CONCLUSION This chapter argued that soft law is born and changed through intricate relationships and complex dialogues between various levels of governance. This suggests that more work might be needed to substantiate claims that soft law is non-democratic, the perfect site of competence creep, and unaccountable decision making. At the same time, as shown in the chapter, many of the interactions described are not always part of the public domain, or easily retrievable, which raises important issues as to the transparency of the underlying processes. With the ECJ systematically refusing to review soft law instruments in direct actions, time has come to exploit the other channels which could make the Commission accountable for its own soft law. The links with other EU institutions, the national levels of governance, and general stakeholders if properly – and transparently – construed can provide a proper alternative to judicial review. Calls to understand and streamline these links more systematically have been made for more than a decade (Korkea-Aho 2009). It is time we acted on them.

266  Research handbook on soft law

NOTES 1. See also Láncos et al. (2023). 2. For instance, Article 59 of Regulation 2019/943 (2019). 3. The case C‑657/21 Parliament v Commission was withdrawn by the Parliament once the Commission adopted the guidance. 4. Minutes of the 24th meeting of the Contact Group ‘Return Directive’ (2008/115/EC) are available from: European Commission (2017b). 5. Bulgarian Supreme Administrative Court, judgment of 8 June 2012, no. 8244 confirmed in appeal in the Bulgarian Supreme Administrative Court, judgment of 21 November 2012, no. 14573, cited in Sulev (2020, p. 1562).

REFERENCES AI HLEG., (2019). Ethics guidelines for trustworthy AI [online]. Brussels: High-Level Expert Group on Artificial Intelligence. [Viewed 27 September 2022]. Available from: https://​ec​.europa​.eu/​futurium/​ en/​ai​-alliance​-consultation​.1​.html. Aldestam, M., (2004). Soft law in the state aid policy area. In: U. Mörth, ed. Soft law in governance and regulation: an interdisciplinary analysis. Cheltenham, UK: Edward Elgar. Alemanno, A., (2020). Levelling the EU participatory playing field: a legal and policy analysis of the Commission’s public consultations in light of the principle of political equality. European Law Journal. 26(1–2), 114–135. Alemanno, A. and Chamon, M., (2020). To save the rule of law you must apparently break it [online]. VerfBlog. [Viewed 27 September 2022]. Available from: https://​verfassungsblog​.de/​to​-save​-the​-rule​ -of​-law​-you​-must​-apparently​-break​-it. Alter, K. and Meunier-Aitsahalia, S., (1994). Judicial politics in the European Community: European integration and the pathbreaking Cassis de Dijon Decision. Comparative Political SI. 26(4), 535–561. Andone, C. and Coman-Kund, F., (2022). Persuasive rather than ‘binding’ EU soft law? An argumentative perspective on the European Commission’s soft law instruments in times of crisis. The Theory and Practice of Legislation. 10(1), 22–47. Andone, C. and Greco, S., (2018). Evading the burden of proof in European Union soft law instruments: the case of Commission Recommendations. International Journal for the Semiotics of Law. 31(1), 79–99. Baldwin, R., (1995). Rules and government. Oxford: Clarendon Press. Biondi, A. and Ștefan, O., (2018). The notice on the notion of state aid: every light has its shadow. In: B. Nascimbene and A. Di Pascale, eds. The modernisation of EU state aid for economic and social development. Cham: Springer. pp. 43–61. Blauberger, M., (2009). Of ‘good’ and ‘bad’ subsidies: European state aid control through soft and hard law. West European Politics. 32(4), 719–737. Cappellina, B., Ausfelder, A., Eick, A., Mespoulet, R., Hartlapp, M., Saurugger, S. and Terpan, F., (2022). Ever more soft law? A dataset to compare binding and non-binding EU law across policy areas and over time (2004-2019). European Union Politics. 0(0). Case C-120/78 Rewe-Zentral AG v Bundesmonopolverwaltung für Branntwein ECLI:EU:C:1972:42. (Cassis de Dijon) Case C-410/09 Polska Telefonia Cyfrowa sp. z o.o. v Prezes Urzędu Komunikacji Elektronicznej ECLI:EU:C:2011:294. Case C-226/11 Expedia Inc. v Autorité de la concurrence and Others ECLI:EU:C:2012:795. Case C-28/15 Koninklijke KPN NV and Others v Autoriteit Consument en Markt (ACM) ECLI:EU:C:2016:692. Case C-16/16 P Kingdom of Belgium v European Commission ECLI:EU:C:2018:79. Case C-16/16 P Opinion of Advocate General Bobek delivered on 12 December 2017, Kingdom of Belgium v European Commission ECLI:EU:C:2017:959.

Soft law making at the European Commission  267 Case C‑657/21 Order of the President of the Court of 8 June 2022, European Parliament v European Commission. Cini, M., (2001). The soft law approach: Commission rule-making in the EU’s state aid regime. Journal of European Public Policy. 8(2), 192–207. Conseil d’État., (1993). Rapport public 1992, la documentation française. collection études et documents, vol. 44. Paris: Conseil d’État. Consolidated version of the Treaty on European Union (2008). Official Journal C115, 9 May, pp. 13–45. (TEU) Consolidated version of the Treaty on the Functioning of the European Union (2012). Official Journal C326, 26 October, pp. 47–390. (TFEU) Cosma, H. A. and Whish, R., (2003). Soft law in the field of EU competition policy. European Business Law Review. 14(1), 25–56. Council directive 92/43/EEC on the conservation of natural habitats and of wild fauna and flora (1992) Official Journal L206, p. 7. (Habitats Directive) Curtin, D., Hofmann, H. and Mendes, J., (2014). ReNEUAL model rules on EU administrative procedure book ii – administrative rulemaking [online]. Research Network on EU Administrative Law (ReNEUAL). pp. 40–64. [Viewed 28 September 2022]. Available from: http://​www​.reneual​.eu/​ images/​Home/​BookII​-Admi​nistrative​Rulemaking​_individualized​_final​_2014​_09​_03​.pdf. Dam, J. C. A. van, (2017). Guidance documents of the European Commission: a typology to trace the effects in the national legal order. Review of European Administrative Law. 10(2), 75–91. Dawson, M., (2009). Soft law and the rule of law in the European Union: revision or redundancy?. EUI Working Papers (RSCAS) 2009/24 [online]. [Viewed 15 August 2022]. Available from: https://​ssrn​ .com/​abstract​=​1415003. Dawson, M., (2017). Integration through soft law: no competence needed? Juridical and bio-power in the realm of soft law. In: S. Garben and I. Govaere, eds. The division of competences between the EU and the member states: reflections on the past, the present and the future. Oxford: Hart Publishing. pp. 235–250. DG for Justice and Consumers., (2020). Information on the Package Travel Directive in connection with the Covid-19 [online]. Brussels: Directorate-General for Justice and Consumers, European Commission. [Viewed 15 August 2022]. Available from: https://​ec​.europa​.eu/​info/​sites/​default/​files/​ coronavirus​_info​_ptd​_19​.3​.2020​.pdf. EDPB., (2020a). Guidelines 04/2020 on the use of location data and contact tracing tools in the context of the COVID-19 outbreak [online]. Brussels: European Data Protection Board. [Viewed 28 September 2022]. Available from: https://​edpb​.europa​.eu/​our​-work​-tools/​our​-documents/​guidelines/​ guidelines​-042020​-use​-location​-data​-and​-contact​-tracing​_en. EDPB., (2020b). Letter from EDPB to Head of Unit European Commission DG for justice and consumers [online]. Brussels: European Data Protection Board. [Viewed 28 September 2022]. Available from: https://​edpb​.europa​.eu/​sites/​default/​files/​files/​file1/​edp​blettereca​dvisecodiv​-appguidance​_final​.pdf. eHealth Network., (2020). Mobile applications to support contact tracing in the EU’s fight against COVID-19. Common EU toolbox for member states [online]. Version 1.0. Brussels: eHealth Network. [Viewed 28 September 2022]. Available from: https://​ec​.europa​.eu/​health/​sites/​default/​files/​ehealth/​ docs/​COVID​-19​_apps​_en​.pdf. Eliantonio, M., (2021). Judicial review of soft law before the European and the national courts: a wind of change blowing from the member states?. In: M. Eliantonio, E. Korkea-aho and O. Ștefan, eds. EU soft law in the member states: theoretical findings and empirical evidence. Oxford: Hart Publishing. pp. 283–302. Eliantonio, M. and Ștefan, O., (2021). The elusive legitimacy of EU soft law: an analysis of consultation and participation in the process of adopting COVID-19 soft law in the EU. European Journal of Risk Regulation. 12(1), 159–175. Eliantonio, M., Korkea-aho, E. and Ștefan, O., eds., (2021). EU soft law in the member states: theoretical findings and empirical evidence. Oxford: Hart Publishing. European Commission., (1962a). Commission notice on exclusive dealing contracts with commercial agents. Official Journal 139, p. 2921. European Commission., (1962b). Commission notice on patent licensing agreements. Official Journal 139, p. 2922.

268  Research handbook on soft law European Commission., (1980). Communication concerning the consequences of the judgment given by the Court of Justice on 20 February 1979 in case 120/78 (‘Cassis de Dijon’). Official Journal C256, p. 2. European Commission., (1985). Completing the internal market: white paper from the Commission to the European Council (COM/85/310 final). Brussels: Commission of the European Communities. European Commission., (2001). Commission notice on agreements of minor importance which do not appreciably restrict competition under Article 81(1) of the Treaty establishing the European Community (de minimis). Official Journal C368, p. 13. European Commission., (2002). Communication towards a reinforced culture of consultation and dialogue – general principles and minimum standards for consultation of interested parties by the Commission (COM/2002/704 final) [online]. Brussels: European Commission. [Viewed 29 September 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​?uri​=​celex:​52002DC0704. European Commission., (2006). Interpretative communication on the Community law applicable to contract awards not or not fully subject to the provisions of the Public Procurement Directives. Official Journal C176, p. 2. European Commission., (2009). Commission recommendation on the regulatory treatment of fixed and mobile termination rates in the EU. Official Journal L124, p. 67. European Commission., (2011). Green paper on on-line gambling in the internal market (COM/2011/128 final) [online]. Brussels: European Commission. [Viewed 29 September 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​?uri​=​celex:​52011DC0128. European Commission., (2012). Communication towards a comprehensive European framework for online gambling (COM/2012/596 final) [online]. Strasbourg: European Commission. [Viewed 29 September 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​?uri​=​celex:​ 52012DC0596. European Commission., (2014a). Recommendation on principles for the protection of consumers and players of online gambling services and for the prevention of minors from gambling online (2014/478/ EU). Official Journal L214, p. 38. European Commission., (2014b). Communication on helping national authorities fight abuses of the right to free movement: handbook on addressing the issue of alleged marriages of convenience between EU citizens and non-EU nationals in the context of EU law on free movement of EU citizens (COM/2014/604 final) [online]. Brussels: European Commission. [Viewed 29 September 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​ALL/​?uri​=​celex:​52014DC0604. European Commission., (2014c). Notice on agreements of minor importance which do not appreciably restrict competition under Article 101(1) of the Treaty on the Functioning of the European Union (de minimis notice). Official Journal C291, p. 1. European Commission, Legal service., (2016). Joint practical guide of the European Parliament, the Council and the Commission for persons involved in the drafting of European Union legislation [online]. Luxembourg: Publications Office. [Viewed 3 October 2022]. Available from: doi: 10.2880/5575 European Commission., (2017a). Recommendation (EU) 2017/2338 establishing a common ‘Return Handbook’ to be used by Member States’ competent authorities when carrying out return-related tasks (Return Handbook). Official Journal L339, p. 83. European Commission., (2017b). Minutes of the 24th meeting of the Contact Group Return Directive [online]. Brussels: European Commission. [Viewed 29 September 2022]. Available from: https://​ ec​.europa​.eu/​transparency/​expert​-groups​-register/​screen/​expertgroups/​consult​?do​=​groupDetail​ .groupDetail​&​groupID​=​2232. European Commission., (2017c). Communication reinforcing integrated supervision to strengthen Capital Markets Union and financial integration in a changing environment (COM/2017/542 final) [online]. Brussels: European Commission. [Viewed 29 September 2022]. Available from: https://​eur​ -lex​.europa​.eu/​legal​-content/​EN/​TXT/​?uri​=​celex:​52017DC0542. European Commission., (2020a). Recommendation (EU) 2020/518 on a common Union toolbox for the use of technology and data to combat and exit from the COVID-19 crisis, in particular concerning mobile applications and the use of anonymised mobility data. Official Journal L114, p. 7.

Soft law making at the European Commission  269 European Commission., (2020b). Recommendation (EU) 2020/648 on vouchers offered to passengers and travellers as an alternative to reimbursement for cancelled package travel and transport services in the context of the COVID-19 pandemic. Official Journal L151, p. 10. European Commission., (2020c). Guidance on apps supporting the fight against COVID 19 pandemic in relation to data protection. Official Journal C124I, p. 1. European Commission., (2020d). Guidelines on the application of the Regulation 2020/2092 on a general regime of conditionality for the protection of the Union budget (2022/C 123/02). Official Journal C123, p. 12. European Commission., (2021a). Guidance document on the strict protection of animal species of Community interest under the Habitats Directive (C/2021/7301 final) [online]. Brussels: European Commission. [Viewed 29 September 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​ EN/​TXT/​?uri​=​PI​_COM:​C(2021)7301. European Commission., (2021b). Delegated Regulation (EU) 2021/654 supplementing Directive (EU) 2018/1972 of the European Parliament and of the Council by setting a single maximum Union-wide mobile voice termination rate and a single maximum Union-wide fixed voice termination. Official Journal L137, p. 1. European Commission., (2022). Guidelines on State aid for climate, environmental protection and energy (2022/C 80/01). Official Journal C80, p. 1. European Council., (2020). European Council meeting (10 and 11 December 2020) – conclusions (EUCO 22/20) [online]. Brussels: European Council. [Viewed 4 October 2022]. Available from: https://​www​.consilium​.europa​.eu/​media/​47296/​1011​-12​-20​-euco​-conclusions​-en​.pdf. European Parliament., (1969). Résolution sur le format de la collectivité des États membres de la Communauté ainsi que les actes du Conseil non prévus par les traités. Official Journal C63, p. 18. European Parliament., (2007). Resolution of 4 September 2007 on the institutional and legal implications of the use of ‘soft law’ instruments (2007/2028(INI)). Official Journal C187E, p. 75. European Parliament., (2013). Resolution of 10 September 2013 on online gambling in the internal market (2012/2322(INI)). Official Journal C93, p. 42. European Parliament., (2016). Resolution of 9 June 2016 calling for an open, efficient and independent European Union administration (2016/2610(RSP)). Official Journal C86, p. 126. European Parliament and Council directive 2006/123/EC on services in the internal market (2006) Official Journal L376, p. 36. (Services Directive) European Parliament and Council directive 2008/115/EC on common standards and procedures in Member States for returning illegally staying third-country nationals (2008) Official Journal L348, p. 98. (Return Directive) European Parliament and Council directive 2011/24/EU on the application of patients’ rights in cross-border healthcare (2011) Official Journal L88, p. 45. (Patients’ Rights Directive) European Parliament and Council directive (EU) 2015/2302 on package travel and linked travel arrangements, amending regulation (EC) no 2006/2004 and directive 2011/83/EU of the European Parliament and of the Council and repealing Council directive 90/314/EEC (2015) Official Journal L326, p. 1. (Package Travel Directive) European Parliament and Council directive (EU) 2018/1972 establishing the European Electronic Communications Code (recast) (2018) Official Journal L321, p. 36. (European Electronic Communications Code) European Parliament and Council regulation (EU) 2016/679 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing directive 95/46/EC (2016) Official Journal L119, p. 1. (General Data Protection Regulation) European Parliament and Council regulation (EU) 2019/943 on the internal market for electricity (2019) Official Journal L158, p. 54. European Parliament and Council regulation (EU, EURATOM) 2020/2092 on a general regime of conditionality for the protection of the Union budget (2020) Official Journal LI433, p. 1. (Rule of Law Conditionality Regulation) Feeley, T., (2020). Commentary: European Parliament and Covid-19 [online]. UK in a Changing Europe. [Viewed 15 August 2022]. Available from: https://​ukandeu​.ac​.uk/​european​-parliament​-and​-covid​-19. Garben, S., (2019). Competence creep revisited. Journal of Common Market Studies. 57(2), 205–222.

270  Research handbook on soft law Gardeñes Santiago, M., (1992). Las comunicaciones interpretativas de la Comision: concepto y valor normativo. Revista de Instituciones Europeas. 19(3), 933–949. Georgieva, Z., (2021). Interactions through soft law in the EU multi-level space – the cases of competition enforcement and telecommunications regulation. SSRN [online]. [Viewed 15 August 2022]. Available from: doi: 10.2139/ssrn.3835780 Hodson, D. and Maher I., (2004). Soft law and sanctions: economic policy co-ordination and reform of the Stability and Growth Pact. Journal of European Public Policy. 11(5), 798–813. Hofmann, H. C. H., (2006). Negotiated and non-negotiated administrative rule-making: the example of EC competition policy. Common Market Law Review. 43(1), 153–178. Korkea-aho, E., (2009). EU soft law in domestic legal systems: flexibility and diversity guaranteed?. Maastricht Journal of European and Comparative Law. 16(3), 271–290. Kröger, S., (2007). The end of democracy as we know it? The legitimacy deficits of bureaucratic social policy governance. Journal of European Integration. 5, 565–582. Láncos, P., (2019). Piercing the soft law veil? The hard truth about directive-like recommendations. SSRN [online]. [Viewed 15 August 2022]. Available from: doi: 10.2139/ssrn.3670619. Láncos, P., Xanthoulis, N. and Arroyo Jiménez, L., eds., (2023). The legal effects of EU soft law: theory, language and sectoral insights into EU multi-level governance. Edward Elgar. Leclerc, S., (1998). Les communications de la Commission et le marché intérieur. a propos de l’arrêt rendu par la Cour de justice des Communautés européennes le 20 mars 1997 dans l’affaire C-57/95, Rec. 1997, p. I-1640 à I-1652. Cahiers de droit European. 34(1-2), 161–176. Leucht, B., (2021). The Cassis de Dijon judgment and the European Commission. In: A. Albors-Llorens, C. Barnard and B. Leucht, eds. Cassis de Dijon: 40 years on. Oxford: Hart Publishing. pp. 69–94. Marcus, J. S., Poitiers, N. F., Guetta-Jeanrenaud, L., Grzegorczyk, M., Röhner, N., Buckingham, S., Hortal Foronda, F. and Pelkmans, J., (2021). The impact of COVID-19 on the internal market [online]. Luxembourg: Policy Department for Economics, Scientific and Quality of Life Policies, European Parliament. [Viewed 15 August 2022]. Available from: https://​www​.europarl​.europa​.eu/​thinktank/​en/​ document​.html​?reference​=​IPOL​_STU(2021)658219. Marjosola, H., (2022). Shadow rulemaking: governing regulatory innovation in the EU financial markets. German Law Journal. 23(2), 186–203. Mörth, U., (2021). The national politics of EU soft law: Sweden and the balancing act between output and input legitimacy. In: M. Eliantonio, E. Korkea-aho and O. Ştefan, eds. EU soft law in the member states: theoretical findings and empirical evidence. Oxford: Hart Publishing. pp. 343–358. Petropoulou Ionescu, D., (2021). Habemus legitimacy? The European Commission opens public consultation for a guidance document. European Journal of Risk Regulation. 12(4), 861–865. Petropoulou Ionescu, D. and Eliantonio, M., (2021). Democratic legitimacy and soft law in the EU legal order: a theoretical perspective. Journal of European Contemporary Research. 17(1), 43–65. Petropoulou Ionescu, D. and Eliantonio, M., (2023). Words are stones: constructing bindingness through language in EU environmental soft law. In: P. Láncos, N. Xanthoulis and L. Arroyo Jiménez, eds. The legal effects of EU soft law: theory, language and sectoral insights into EU multi-level governance. Edward Elgar. Available from: https://​ssrn​.com/​abstract​=​4053532 ReNEUAL., (2014). ReNEUAL 1.0: ReNEUAL model rules on EU administrative procedure [online]. Research Network on EU Administrative Law (ReNEUAL). [Viewed on 15 August 2022]. Available from: http://​www​.reneual​.eu/​projects​-and​-publications/​reneual​-1​-0. Schäfer, A., (2006). A new form of governance? Comparing the open method of co-ordination to multilateral surveillance by the IMF and the OECD. Journal of European Public Policy. 13(1), 70–88. Scott, J., (2011). In legal limbo: post-legislative guidance as a challenge for European administrative law. Common Market Law Review. 48(2), 329–355. Senden, L., (2004). Soft law in European Community law. London: Hart Publishing. Senden, L. and Prechal, S., (2001). Differentiation in and through Community soft law. In: B. de Witte, D. Hanf and E. Vos, eds. The many faces of differentiation in EU law. Oxford: Intersentia. pp. 181–199. Sisson, K. and Marginson, P., (2001). Soft regulation – travesty of the real thing or new dimension?. One Europe or Several? Working Papers 32/01 [online]. [Viewed 3 October 2022]. Available from: https://​ideas​.repec​.org/​p/​erp/​oneeur/​p0032​.html. Slominski, P. and Trauner, F., (2021). Reforming me softly – how soft law has changed EU return policy since the migration crisis. West European Politics. 44(1), 93–113.

Soft law making at the European Commission  271 Snyder, F., (1994). Soft Law and institutional practice in the European Community. In: S. Martin, ed. The construction of Europe: essays in honour of Emile Noël. Dordrecht: Kluwer Academic Publishers. pp. 197–225. Ștefan, O., (2013). Soft law in court: competition law, state aid and the Court of Justice of the EU. Alphen aan den Rjin: Kluwer Law International. Ștefan, O., (2020). The future of European Union soft law: a research and policy agenda for the aftermath of COVID-19. Journal of International and Comparative Law. 7(2), 329–350. Ştefan, O., (2022). Entrenching emergency soft law [online]. European Law Blog. [Viewed 3 October 2022]. Available from: https://​europeanlawblog​.eu/​2022/​04/​12/​entrenching​-emergency​-soft​-law. Ștefan, O., (Forthcoming). Soft law as an effectiveness enhancing tool for internal market harmonization. In: T. Tridimas and R. Schütze, eds. Oxford Principles of European Union law: Internal Market, vol. II. Oxford University Press. Sulev, G., (2020). Soft law in EU electronic communications regulation: a Bulgarian case study. European Papers [online]. 5(3), 1555–1564. [Viewed 15 August 2022]. Available from: https://​www​ .europeanpapers​.eu/​en/​europeanforum/​soft​-law​-eu​-electronic​-communications​-regulation​-bulgarian​ -case​-study. Zhelyazkova, A., Börzel, T., Schimmelfennig, F. and Sedelmeier, U., (2015). Beyond uniform integration? Researching the effects of enlargement on the EU’s legal system. MAXCAP Working Paper Series 8 [online]. [Viewed 4 October 2022]. Available from: http://​ userpage​ .fu​ -berlin​ .de/​ kfgeu/​ maxcap/​system/​files/​maxcap​_wp​_08​.pdf.

18. Beyond norm entrepreneurs: civil society and the framing of the ‘legal’ through soft law Rene Urueña and Rafael Tamayo-Álvarez

1. INTRODUCTION A defining feature of the current process of global governance is the understanding that states are not the only entities with the capacity to determine how to address international concerns (Fakhri 2015, p. 70). Global governance entails that an increasing number of non-state actors can exert authority in many areas over different subjectivities (Klabbers 2020, p. 270). Consequently, the exercise of power has become distinctly diffused and dispersed as its deployment comes through a range of mechanisms that, such as soft law, do not fall within the scope of the ‘formal’ sources of international law (Fakhri 2015), in whose formation states still dominate. Thus, to the extent that the exercise of authority takes place beyond the orbit of the nation-state, many civil society organizations have developed the ability to influence the creation of international norms and standards conceived to tackle a plethora of local and transnational issues (Sikkink and Keck 2014; Törnquist-Chesnier 2004). In that context, soft law has been a crucial tool for civil society organizations ever since the latter’s turn to legalized professionalization in the 1980s. Civil society representatives work as norm entrepreneurs, proposing and developing standards that, if successful, are adopted by states, becoming legally binding laws (Finnemore and Sikkink 1998). This perspective is helpful in the sense that it provides a wider analytical lens to understand the operation of the law in society, as opposed to more state-centered approaches, which build their analysis on state-sanctioned legal sources, thus obscuring non-state legal processes that are of obvious relevance in the distribution of resources (Michaels 2009). However, the idea of ‘norm entrepreneurs’ seems to imply that the ultimate test for the existence of legal norms remains within the state’s domain. Even if the ‘norm entrepreneurs’ framework recognizes the importance of civil society in the emergence of norms, it still underscores formal lawmaking authority lying in state authorities (see generally Pauwelyn 2012). From this perspective, civil society organizations promote norms for states to adopt or reject, and only then do such norms become ‘law’. Accordingly, a successful norm entrepreneur is one who persuades or entices state authorities to adopt her proposal. In this chapter, we propose that civil society plays a much more significant role with regard to soft law than just persuading states to adopt a particular regulatory initiative. While certainly ‘norm entrepreneurs’, we suggest that civil society is, in fact, part of a community of practice, whose actors interact and, through such interaction, create legal meaning that ultimately stabilizes normative expectations, thus acting as law. Regardless of whether such legal utterances (be them ‘guidelines’, ‘resolutions’, ‘manuals’, or the like) are formally adopted by state authorities (i.e., regardless of whether they acquire a legally ‘binding’ status), they can potentially play a crucial role in society, precisely because they have emerged from the practice of the relevant community and become part of that very practice. In that sense, as they put 272

Beyond norm entrepreneurs  273 forward non-binding legal utterances (i.e., ‘soft law’) civil society organizations participate in the process of framing what is considered as a legal problem by the relevant community. To make that point, this chapter starts by describing the concept of community of practice and its role in the emergence of soft law – in particular, in the context of global governance. Then, we focus on two processes through which civil society, as actors in the community of practice, frame the ‘legal’: first, through an ‘appeal to normativity’ to community members; and second, by producing normative discourses in international politics through soft law. As such civil society participates in the social construction of global concerns and defines a particularly ‘legal’ response to them. The following section then turns to domestic courts, and explores the way in which civil society has been successful in framing the legal by ‘hardening’ soft law through strategic litigation before domestic courts. The final section concludes.

2.

SOFT LAW AND THE COMMUNITY OF PRACTICE

This section describes the concept of a community of practice and its potential for a broader account of civil society’s influence through soft law.1 This concept comes from education research. In 1991, anthropologist Jean Lave and computer scientist Etienne Wenger (1991) proposed the notion of situated learning. They argued that learning is fundamentally a social process and involves socialization. A community of practice, then, denotes a group of people that is defined by mutual engagement, a joint enterprise, and a shared repertoire, meaning ‘routines, words, tools, ways of doing things, stories, gestures, symbols, genres, actions, or concepts that the community has produced or adopted in the course of its existence and which have become part of its practice’ (Wenger 1998, p. 83). This notion was later taken up by international relations scholar Emanuel Adler (2005, p. 15), who asserts that: there is no reason … why we should not be able to identify transnational or even global communities of practice. The closer we get to the level of practices, the more we can take the international system as a collection of communities of practice—for example, communities of diplomats, traders, environmentalists, and human rights activists. Communities of practice cut across state boundaries and mediate between states, individuals, and human agency, on the one hand, and social structures and systems, on the other.

Focusing on a community of practice better portrays civil society’s role concerning soft law. Instead of being mere norm entrepreneurs, civil society organizations are a crucial part of the community and, as such, contribute to the practice that creates norms. How is this possible? Perhaps the point to start is the work of Jutta Brunnée and Stephen Toope (2010), who have applied the notion to analyze the problem of international legal obligation. For Brunnée and Toope, transnational communities of practice provided the space for interaction that created the emergence of such obligations: ‘legal obligation, then, is best viewed as an internalized commitment and not as an externally imposed duty matched with a sanction for non-performance’ (Toope and Brunnée 2010, p. 115). This insight is critical to understanding civil society’s role in creating soft law. Brunnée and Toope’s interactive approach implies that the commitment emerges from within the community and not from without. That is, obligation derives not from whether the particular norm is legally binding but rather from whether the community internalizes a commitment to that specific norm, regardless of its legal status.

274  Research handbook on soft law To be sure, the process of creating such an internal notion of commitment among the community of practice might involve using juridical concepts as signalling devices. For example, adopting a norm as a binding treaty (and not as ‘soft law’) might signal that that particular norm is important and thus might facilitate the emergence of the internal commitment of the community to the norm. But the opposite process may also occur: the binding nature of a norm could hinder the emergence of internal commitment. This could happen where, for example, the binding norm was adopted or promoted by powerful actors who are rejected by the community of practice. Thus, from the community of practice perspective, the relevant division is not hard law versus soft law but rather internalized versus non-internalized commitments. A normative pull comes into play when a commitment is internalized, even if it derives from formally non-binding norms. In contrast, a formally binding obligation will not create a legal obligation if it fails to generate an internal commitment. Time, iterative interactions, and learning are crucial for this practice. It is hard to understand the community if one conceptualizes the interaction between actors as discrete encounters. Iteration creates an interaction that is qualitatively different from that of discrete contact, as the constant process of interaction and learning shapes expectations within a community of practice. Iteration, in other words, changes not only the strategies within a game but the game itself (see Frischmann 2003). Constructivist international relations have also explored this idea: while the interaction between agents develops the structure, being part of the structure impacts the interests and strategies of the agents (Wendt 1987, pp. 5–6; Adler 2005, pp. 5–6). This constant interaction process within the relevant community of practice helps explain compliance with soft law. Traditional readings of compliance with international norms view both norms and the context of implementation as static, and compliance is understood as a rather mechanical process in which ‘leverages’ are activated to achieve the demanded behaviour from the addressees (Howse and Teitel 2010). The paradigmatic form is domestic criminal law: compliance with a judicial decision is reached by activating specific sociopolitical mechanisms (judicial enforcement, for instance) to ‘force’ the addressee of the decision to do something (comply with an obligation). This view informs the dominant understanding of compliance in international legal scholarship (for a survey, see Kingsbury 1998; for a critique, see Howse and Teitel 2010), in which the critical problem seems to be how to incentivize compliance when there is little political leverage to compel states to change their behaviuor (Posner and Sykes 2013). Given the scarcity of enforcement mechanisms, compliance appears an almost discretionary choice for states, particularly in the context of soft law (see e.g., Posner 2014). Our reading rebuts that understanding in two directions. First, it rejects the idea that a soft law norm is static in the sense of being fully crystallized or carved in stone. Indeed, soft law norms are often vague since their precise contours only become apparent in the implementation process, hence in dialogue with the involved state authorities. In other words, a soft law norm is just one step (though certainly an essential one) in a long process. Moreover, the context of implementation is rarely static. Therefore, the community of practice creates a continuum between the norm and the conditions for its implementation. The political leverages that could lead to compliance are dynamic, and they can be influenced by the very decision the implementation of which is sought. Thus, facing a soft law norm, the question of compliance is not merely whether there are tools to coerce actors into compliance. Instead, the issue is also how soft law mobilizes and even generates such tools and pushes actors toward compliance. Consequently, compliance with soft law is not an end in itself but part of a much larger process

Beyond norm entrepreneurs  275 that involves a series of practices geared toward the emergence and consolidation of internal commitments to a norm by community members. In this context, the internalization of commitment relates to a learning process. The community of practice establishes cognitive frameworks created and circulated for interventions in concrete conflicts and the purpose of governance.2 Mainly, such internal commitment emerges through two forms of epistemic practices: First, the community of practice incentivizes the production of knowledge – a social process that involves sharing experiences, exerting and gaining influence, and developing networks that produce specifically legal knowledge. Specific interpretations of the law are taught, defended, and made dominant in the community of practice. This process of knowledge creation and circulation not only occurs in academic fora but also represents a core activity of many actors in the community. In this regard, civil society organizations draft reports and gather data, national human rights institutions create human rights tutorials, and governments, last but not least, publish their own reports and support the circulation of their own practices of knowledge. The second aspect relates to how ‘problems’ are cognitively framed to allow for a ‘legal’ answer. In many instances, the legal dimension is not evident in the early stages of a conflict, often because the problem is too large or complex to be processed as a distinctively legal dispute. To frame a complex social problem as a legal issue, the community must often intervene to reconstruct these problems in a profound way, which implies cognitive framing. Such descriptions are not neutral; they indicate normative choices. Cognitive framing goes beyond the mere ‘translation’ of domestic realities and helps organize actors’ interpretation of their contexts (see Jasanoff 2004). The remainder of this chapter analyses how civil society organizations play a significant role in the international lawmaking process as actors of the community of practice underlying such a process. First, we describe instances in which civil society participation was an essential factor in what Bradlow and Hunter (2020) refer to as the ‘linear relationship’ between international soft and hard law. This means the progression of specific soft law instruments so widely accepted that they become part of a ‘hardening’ process. Indeed, among the advantages credited to soft law is that its non-binding character confers flexibility to policymakers by enabling them to experiment with regulatory initiatives in areas where no international consensus has been reached (Boyle 1999). Hence, soft law instruments perform a foreshadowing function in a learning process geared toward enacting international rules with a legally binding force. In this regard, they allow civil society to push the normative frontier by presenting policymakers with a ‘dress rehearsal’ of possible regulation. Furthermore, we will argue that soft law not only allows civil society to encourage the production of legally binding regulations for specific issue areas. Indeed, civil society’s involvement in the international lawmaking process is more structural. This is because soft law allows civil society to shape and redefine our understanding of the social issues requiring global law-based responses. The core argument is that civil society organizations can frame our way of thinking about certain issue areas and, correspondingly, delimit the range of desirable and possible action upon them. Put another way, civil society endows with meaning the international lawmaking process. Specifically, it informs the social construction of states’ interests and influences the conduct of other relevant actors whose interests coalesce around a given issue area of international relations. As a result, civil society can expand the limits of the international community’s legal consciousness, especially by bringing social problems

276  Research handbook on soft law into existence and rendering some normative responses to act upon them more politically acceptable than others. Second, we show that civil society contributes to the internalization of specific international soft law norms – and by doing so to the collapse of the line between the national and the international – by litigating before local courts. Towards this end, we take constitutional adjudication in Colombia as a case study to show how strategic litigation can allow civil society organizations to produce normative implications by ‘importing’ international soft law norms into a domestic legal system with legally binding repercussions. In the Colombian case, we argue that this has taken the form of a constructive dialogue between the Constitutional Court and citizens regarding the relevance of international soft law norms as a hermeneutic criterion to be applied when discerning the scope of constitutional provisions.

3.

BETWEEN MAKING APPEALS TO NORMATIVITY AND CONSTRUCTING SOCIAL GLOBAL PROBLEMS

Civil society produces a type of knowledge that fulfils two main functions within the community of practice underlying the international lawmaking process. On the one hand, it makes what we call an ‘appeal to normativity’ to community members. This means putting international soft law norms and standards in circulation to encourage the subsequent adoption of international regulations based on them or position these as an ideal tool for regulating behaviour and stabilizing actor expectations in a particular field. On the other hand, civil society produces normative discourses in international politics through soft law. This means creating influential narratives that determine what counts as acceptable behaviour in a particular field of action. As such, civil society participates in the social construction of global concerns and determines the adequate normative response to them. 3.1

Making Appeals to Normativity

In a community of practice, civil society employs soft law to produce legal knowledge with the purpose of either influencing subsequent state practice or stabilizing the normative expectations of states and other stakeholders concerning a specific issue area. To substantiate this proposition, we start highlighting the role civil society organizations have had in the development of environmental law. For instance, the Rio Declaration on Environment and Development contains several principles that have progressively acquired a prominent status in ecological policymaking and state practice, such as providing access to the relevant information regarding environmental risks (Principle 10), undertaking an environmental impact assessment for proposed activities that are likely to have a significant adverse impact on the environment (Principle 17), and the precautionary approach (Principle 15) (Hernández 2019). Notably, civil society organizations, especially NGOs, were crucial participants in negotiating key political instruments that resulted from the 1992 UN Conference on Environment and Development (UNCED) (Gómez Lee and Maxfield 2017). Specifically, they used the space opened to them by the Preparatory Committee (PrepCom), which operated in anticipation to UNCED, to approach state delegations and provide inputs later used in drafting Agenda 21 and the Rio Principles (Island Press 1992).

Beyond norm entrepreneurs  277 To appreciate the extent of civil society’s influence during the 1992 Earth Summit, we need to go back to the World Conservation Strategy (WCS) prepared by the International Union for Conservation of Nature and Natural Resources (IUCN) in 1980 upon commission by the United Nations Environment Programme (UNEP) and supported by the World Wildlife Fund (WWF). The WCS resulted from a consultation process and pursued a consensus on policy and conservation efforts in the context of world development. Accordingly, a final draft was submitted to governments, governmental and non-governmental international organizations, individuals, and the scientific community. The WCS aimed to help advance the achievement of sustainable development through the conservation of living resources (see IUCN 1980). In doing so, this soft law instrument anticipated the linkage between environmental conservation and development perceived in the Brundtland Report and Principles 3 and 4 of the Rio Declaration (Gómez Lee and Maxfield 2017). Subsequently, the linkage between conservation and development has been codified by various international treaties, most notably the Convention on Biological Diversity (1992) negotiated during the Earth Summit in Rio de Janeiro. The influence of civil society organizations was also relevant regarding another international environmental treaty negotiated during the 1992 Earth Summit: The United Nations Framework Convention on Climate Change (UNFCCC 1992). In this case, two soft law instruments that included civil society participation acted as precursors, especially in the sense of positioning climate change as a global governance problem that required a suitable solution through an international treaty. The first is the Declaration that resulted from the World Climate Conference convened in 1979 by the World Meteorological Organization (1979). Indeed, this document reflects the expert opinion of many members of the scientific community who expressed their views on the implications of climate change during the conference’s proceedings. Also decisive for the coming into fruition of the UNFCCC was the First Assessment Report published in 1990 by the Intergovernmental Panel on Climate Change, which advocated for an international convention to tackle the growing concern raised by climate change (Intergovernmental Panel on Climate Change 1990). On the other hand, civil society can produce normative effects in the global community through soft law instruments without these necessarily becoming mandatory regulations. Indeed, Kaufmann-Kohler (2010) has identified specific soft law instruments elaborated and codified in the area of international arbitration, including the Rules on Taking Evidence in International Commercial Arbitration and the Guidelines on Conflicts of interests in international arbitration developed by the International Bar Association (IBA). The process of elaboration of both instruments by IBA working groups was open to the arbitration epistemic community (Kaufmann-Kohler 2010, p. 13). Through considering the inputs by private associations and professionals, both devices sought to reflect a consensus on best practices concerning the issues they aim to tackle. For instance, commenting on the Rules on Taking Evidence, Kaufmann-Kohler asserts that they were produced to remedy uncertainty about how evidence was gathered in international arbitration (Kaufmann-Kohler 2010, p. 7). Kaufmann-Kohler (2010, p. 13) further argues that these soft law instruments influence arbitral practice despite not being perceived as mandatory, referring to this effect as ‘soft normativity’. Indeed, Alvarez (2018) describes the IBA Rules on Taking Evidence’s receptivity among international investment tribunals, including some rendering those rules essential for their decision-making process. Likewise, tribunals and litigants constantly elaborate on the Guidelines on Conflicts of Interests in matters such as disqualifying an arbitrator (Alvarez

278  Research handbook on soft law 2018, p. 160). Accordingly, without necessarily becoming legally binding, these instruments have a normative appeal allowing members of the arbitration community to regulate their behaviour. 3.2

Constructing Global Social Problems

We now turn to a second aspect of our argument about civil society’s influence through soft law in the international lawmaking process. Specifically, we will see that civil society engages in the production and framing of global social problems that demand a normative response by the international community through shaping and dictating the terms of emerging normative discourses. For instance, Epstein (2008) notes that until the 1960s, the dominant narrative considered whales as a resource to be exploited as a matter of national security, to the extent that being a modern, advanced, industrialized state involved being a whaling state. From then onwards, a shift towards an anti-whaling global order where whaling is precluded has taken place. This shift, Epstein argues, resulted from the rise of a worldwide anti-whaling discourse produced by environmental activists that became hegemonic after being mixed with the Cold War discourse on capitalism and democracy and the emerging global ecological discourse (Epstein 2008, pp. 95 and 112). Once it had become dominant, this discourse strengthened the position of environmental NGOs as powerful actors in world politics with the ability to influence the policies of the International Whaling Commission (IWC). The anti-whaling discourse effected a normative change in international whaling politics (Epstein 2008, pp. 91, 150). Notably, it has specified the scope of what is considered legal and politically accepted about whaling, for instance, by contributing to defining categories such as ‘commercial’ and ‘cultural’ whaling employed to manage whaling at the IWC and encouraging many formerly whaling states to pass national legislation protecting whales (Epstein 2008, pp. 21, 106, and 152). Epstein argues that the anti-whaling campaign promoted by environmental activists transformed identities and subject positions in international whaling politics. First, it positioned NGOs as actors in the international system with the capacity to pressure states and persuade the global public opinion to mobilize against whaling (Epstein 2008, pp. 93 and 109). Second, the discourse created whales as subjects worthy of protection because they have a moral right to life, thus precluding any possibility to exploit them even in the absence of depletion in whales’ stocks (Epstein 2008, p. 206). Third, the discourse allowed NGOs to rearticulate state identities from whaling to anti-whaling states. It did so by infusing with a new meaning the idea of a modern, advanced, civilized, and environmentally responsible state by creating a binary distinction between anti-whaling states on the one hand and ‘barbarian’ states practicing whaling on the other (Epstein 2008, pp. 173 and 250). According to Epstein, this explains the behaviour adopted later by third-world states that have never whaled to enter the IWC and engage in anti-whaling politics to cast themselves as a developed and cooperative state (Epstein 2010, pp. 175–193 and in particular p. 188). Another example concerns the current global anti-corruption regime. The 1990s saw the ‘eruption’ of a worldwide anti-corruption crusade (Naím 1995). The renewed global interest in corruption led to many international agreements tackling the issue (see e.g., Convention on Combating Bribery of Foreign Public Officials in International Business Transactions 1999). Transparency International (TI) and other anti-corruption NGOs played a crucial role in this process (see Wang and Rosenau 2001; Hindess 2005; Carr and Outhwaite 2011).

Beyond norm entrepreneurs  279 TI is an international non-governmental organization whose stated mission is the ‘relief of poverty, suffering and distress in any part of the world caused directly or indirectly by corruption’. TI’s governance structure has two key components: the first is the Secretariat, based in Berlin, which is accountable to the International Board and to the wider TI movement, through the Board of Directors and the Annual Membership Meeting. The Secretariat is led by an Executive Director, and has an operating budget of about 25 million euros (TI 2022). The second key component is the national chapters, which are local non-governmental organizations that have been accredited by the Board of Directors, under the National Chapter Accreditation and Individual Member Appointment Policy, which imposes a set of procedures and requirements for the accreditation process. There are currently more than 100 national chapters, which vary in size, expertise and access to funding. TI was founded in 1993 by Peter Eigen, a former senior World Bank official, together with other individuals of extensive international experience, such as Fritz Heimann, legal counsel from General Electric, and Frank Vogl, also a former official of the World Bank. While created by former World Bank staff, TI is not a spin-off of Bank policies. On the contrary, according to his own account, Eigen created TI because the World Bank was unwilling to push further in understanding corruption as a crucial part of its mandate. The Bank’s position was that corruption was a political issue, and that the Bank should not get involved in such debates. In fact, the Bank’s initial approach was to deny funding to TI, and to block efforts to create an ‘island of integrity’, one of TI’s initial initiatives, around Bank-funded projects (Eigen 2002; Lundberg 2002). TI emerges in the context of the current global anti-corruption movement, which may be explored in function of the crystallization of the link between corruption and economic development, which became part of the common sense in academic and multilateral financial institutions in the 1990s. By the early years of that decade, the idea that ‘institutions matter’ for development gained momentum (North 1991, p. 97; see also North 1990). And, if institutions were crucial for economic activity, their failure would undermine economic development. And a central institutional failure is, of course, corruption. Following this rationale, the 1990s saw the emergence of a robust body of literature that correlated corruption with slow economic growth (see three influential studies at that time, Shleifer and Vishny 1993; Mauro 1995; Mauro 1996). TI played a crucial role in translating this research into a particular ‘legal’ problem that could be tackled through legal means. Indeed, through the Corruption Perception Index (CPI) and other regulatory initiatives, not only did this NGO contribute to establishing corruption as a global governance problem, but it also started ranking governments according to their ability or willingness to combat it (Urueña 2014). Furthermore, the measurement of corruption provided a vantage point to international financial institutions to present this phenomenon as an impediment to development (McCoy and Heckel 2001). Notably, this created a dominant narrative whereby combating corruption, perceived as a public sector problem, blended with the emerging normative discourse on good governance to catalyse neoliberal reforms worldwide (Brown and Cloke 2004; Brown and Cloke 2011). From then onwards, such a narrative informs most international treaties devoted to combating corruption by criminalizing bribery. Accordingly, TI has contributed to reconfiguring our understanding of corruption through its anti-corruption campaign based on measuring perceptions on corruption (Urueña 2018). In the 1970s, there were two competing conceptions about corruption in world politics. On the one hand, the third world identified corruption as a corporate influence on politics, recom-

280  Research handbook on soft law mending that it should be addressed systemically through a code of conduct for transnational corporations. On the other, the United States understood corruption from the paradigm of an individual actor offence, emphasizing the anti-bribery component (Katzarova 2018). Hence corruption as a global governance problem experienced a transformation from corporate abuse of power in the 1970s to government abuse of power in the 1990s (Katzarova 2018, p. 2). In this context, TI provided a narrative for blaming governments as the primary source of corruption. In doing so, it contributed to narrowing the conditions of possibility for international anti-corruption regulation that focuses on root causes instead of focusing on criminalizing certain corrupt practices (Bukovansky 2006). This intellectual and political shift put anti-corruption squarely in the mandate of international financial institutions, particularly the World Bank (see Polzer 2001). Since the 1990s, corruption and good governance became central to the agenda of multilateral financial institutions, which made the anti-corruption campaign an important aspect of the overall program governance transformation championed since then. TI is, therefore, part of a wider institutional complex, which has a particular ideological dimension, in the sense that it mobilizes meaning in order to affect the distribution of resources in the direction it favours. In our case, the anti-corruption campaign shares the ideological bend of a specific set of policies to foster development that was implemented in the late 1980s, and shares an institutional milieu with the producers of such policies, particularly multilateral financial institutions, such as the World Bank and regional development banks.

4.

HARDENING SOFT LAW BY LITIGATING BEFORE LOCAL COURTS

This section proceeds in the following sequence: first, we offer an overview of international soft law norms’ positive reception by the Colombian Constitutional Court. Second, we will show that, in turn, the Court’s receptivity has encouraged citizens to rely on soft law norms to substantiate their arguments when they recur before its constitutional jurisdiction. We contend that this is a co-constitutive dialogue between the Colombian Constitutional Court and citizens as members of a community of practice, which results in international soft law norms entering the Colombian legal system as part of a ‘hardening’ process. 4.1

The Constitutional Court’s Incorporation of Soft Law into the Internal Legal System

The Colombian Constitutional Court’s receptivity of international soft law norms primarily results from incorporating them into its doctrine on the ‘constitutional block’. As we will see, this move has been enabling in the sense of encouraging citizens to strategically deploy soft law as part of their legal arguments when they recur to the Court’s jurisdiction. The block of constitutionality is premised on Article 93 of the Constitution of Colombia. This Article operates as a referral clause, effectively expanding the scope of the Constitution (Bonilla 2015). It provides that international treaties and conventions on human rights ratified by Colombia and which prohibit their derogation in a state of emergency prevail in the internal order. Similarly, it provides that the constitutional rights and responsibilities shall be interpreted following these international human rights instruments.

Beyond norm entrepreneurs  281 The constitutional block enables harmonizing Article 93, which provides that human rights treaties prevail in the internal legal order, with Article 4, which defines the Constitution as a ‘supreme norm’ (Judgement C-146/2021, para. 91). Specifically, it fulfils two functions: an ‘integrating function’ whereby it offers a parameter to determine whether other norms within the national legal system conform to the Constitution, and an ‘interpretative function’, according to which it contains provisions that assist in the identification of the content and scope of constitutional clauses (Judgement C-291/07, para. 42). Hence, the block allows human rights treaties to be applied in the context of a constitutional review, which it does by taking them as a parameter of constitutionality despite not being formally part of the Constitution; and as an interpretive tool to be used to discern the scope of constitutional rights and responsibilities. Dating back to the 1990s, the Colombian Constitutional Court developed a line of jurisprudence distinguishing between a constitutional block understood ‘in the strict’ and ‘in the broad’ sense. The difference between the two is that only the norms that are part of the block in the strict sense have a constitutional rank (Judgment C-146/21, para. 93). However, according to the Court, international human rights treaties incorporated into the block of constitutionality, whether in the strict or broad sense, serve to interpret constitutional provisions and perform the constitutionality review of other norms within the national legal system (Judgment C-146/21, para. 95). Within this context, the Court uses international soft law norms as an interpretative tool to discern the extent of the obligations contained in human rights treaties concluded by Colombia (see generally López-Medina and Sánchez-Mejía 2008; Quinche Ramírez 2019), which, in turn, comprise the constitutional block. For instance, the Court has recurred to soft law standards when evaluating the normative content of human rights provisions in matters such as gender equality (see e.g., Judgment C-659/16), the rights of disabled persons (see e.g. Judgment C-458/2015), and the human right to water (see e.g. Judgment T-740/11). Accordingly, whereas soft law standards are not part of the constitutional block, they have an interpretative utility regarding the international human rights treaties in which Colombia is a party. Nevertheless, the Colombian Constitutional Court has also considered that international soft law norms can be part of the constitutional block in the broad sense. In doing so, the Court converts soft law instruments of international law to hard law at the local level (López-Medina and Sánchez-Mejía 2008, p. 344). Specifically, it integrates soft law into the Colombian constitutional order, thus giving it weight in interpreting constitutional provisions and determining whether other norms within the national legal system conform with the Constitution. For instance, the Court has attributed a binding force to the Guiding Principles on Internal Displacement since it has considered them part of the block of constitutionality to resolve specific cases (see generally Judgment T-821/07). Notably, referring to these principles, the Court declared that as they fundamentally reflect and contribute to filling gaps in the content of human rights treaties, and considering that they enjoy wide acceptance, they should be parameters in the creation and interpretation of norms and regulations about internally displaced persons (Judgment SU.1150/00, para. 38). Similarly, the Court has ascribed normative force to the Pinheiro Principles on Housing and Property Restitution for Refugees and Displaced Person within the Colombian constitutional order by considering them part of the constitutional block in the broad sense (Judgment C-035/16, paras 41–43). Specifically, the Court asserted that since the principles constitute a development of the international doctrine on the fundamental right to integral reparation, they should be parameters for the constitutionality of internal norms that develop this right (Judgment C-035/16, para. 44).

282  Research handbook on soft law 4.2

Soft Law as a Tool in Strategic Litigation

Although not the only local jurisdiction in the world where a high court has referred to international soft law standards as authoritative guidance (see e.g., Debevoise & Plimpton LLP 2021), Colombia is a relevant example because, through the constitutional block doctrine, the Colombian Constitutional Court has developed a line of jurisprudence that provides consistency to the incorporation of soft law into the internal legal system. Moreover, by rendering that international soft law norms can be part of the block of constitutionality in the broad sense, the Court effectively confers a legally binding effect to them at the local level. Within this context, it should not come as a surprise that citizens and civil society organizations alike have learned to cite international soft law instruments to enhance the persuasiveness of their legal arguments when litigating before the Court. Before elaborating on this argument, it is essential to clarify how citizens engage the Court’s jurisdiction. The Colombian Constitutional Court’s two-fold jurisdictional scope oversees the constitutionality of internal norms and guarantees the fundamental rights recognized in the Colombian Constitution. The former function is conducted through a constitutional review process. Meanwhile, the latter is performed through the ‘Tutela’, a legal remedy allowing citizens to bring claims for fundamental and human rights violations.3 In this regard, international soft law norms, as constituent parts of the block of constitutionality, serve as a specific parameter for determining a norm’s constitutionality or for interpreting which human rights are constitutionally protected and to what extent. Notably, it seems to be a trend in that citizens’ reliance on soft law norms is more prevalent in a constitutional review process than in ‘Tutela’ proceedings. We believe that this has to do with the latter usually being triggered by people living in vulnerable material conditions who can bring a claim directly without professional legal assistance. Conversely, legal expertise or advice is required to adequately substantiate a norm’s unconstitutionality in a constitutional review process. Furthermore, these are open to interventions by civil society representatives, thus allowing universities, academics, professional associations, guilds, unions, NGOs, etc., to better understand existing soft law standards and file amicus curiae briefs. When litigating before the Colombian Constitutional Court, citizens and civil society organizations have recurred to many international soft law norms. Examples include a ‘Tutela’ proceeding in which an indigenous community complained that a multinational mining corporation operating an open-pit mine infringed their fundamental rights. Specifically, the applicants argued that the mine produced particulate matter above the limits contained in the air quality guidelines of the World Health Organization (see Judgment T-614/19). In the same vein, Colombian unions constantly refer to the recommendations by the ILO Committee on Freedom of Association to shut down regulations that they considered regressive for association rights (see Judgment C-391/19). However, one salient example concerns how Deng and Pinheiro’s principles have been extensively relied on to substantiate charges of unconstitutionality against Colombia’s Victims and Land Restitution Law, the cornerstone of the land restitution program aimed at repairing the harm suffered by victims dispossessed from their property by being forced to flee amid the internal armed conflict. In most cases, the applicants contend that the unconstitutionality of the Victims and Land Restitution Law resides in this being at odds with human rights encapsulated in the Deng and Pinheiro Principles. For example, in one instance, the actor claimed that the law’s treatment of secondary occupants, whose right to compensation is contingent upon demonstration of good

Beyond norm entrepreneurs  283 faith, falls below the standards set forth by the Pinheiro principles (see Judgment C-330/16). Interestingly, the Court’s judgment features a debate on the position of secondary occupants in a condition of vulnerability that revolved around the Pinheiro Principles and was animated by intervening NGOs. Also, in one of the cases, the applicant elaborated on the Court’s jurisprudence recognizing the Pinheiro principles as constituting elements of the constitutional block, to conclude that they contain mandatory guidelines for the State and thus should be a critical parameter for evaluating the law’s constitutionality (see Judgment C-715/12). These examples show that a constructive dialogue is taking place between the Court and civil society around the normative value of soft law, underpinned by the block of constitutionality doctrine.

5. CONCLUSION This chapter explores how civil society influences the international lawmaking process beyond the gaze of norm-entrepreneurship. In particular, we have argued that, as actors in a community of practice, civil society organizations put forward unbinding legal utterances (that is, ‘soft law’) that contribute to the process of framing what is considered a legal problem by the relevant community. We drew on the concept of a community of practice to highlight that civil society promotes the internalization of commitment concerning soft law norms, thus expanding the frontier of the international legal consciousness as part of a learning process among community members. Furthermore, we described that the internalization of commitment takes place through a series of epistemic practices on which civil society relies to influence the regulatory process. First, we contended that through soft law, civil society produces knowledge that subsequently encourages the adoption of international regulations or conveys the adequacy of specific standards for regulating behaviour and stabilizing normative expectations in a particular field despite not having a legally binding status. Likewise, we discussed that soft law allows civil society to frame our thinking and act upon specific global social problems. In this regard, civil society organizations dictate the terms of normative discourses that shape actors’ identities, interests, and overall behaviour. Finally, we took constitutional adjudication in Colombia to illustrate that strategic litigation allows civil society organizations to ‘import’ international soft law norms into domestic systems, where they become binding under domestic law. In that context, civil society becomes part of a community of practice through a co-constitutive dialogue with local tribunals that fosters the internalization of commitment of certain soft law norms. These are three forms of expansion of the current legal consciousness advanced by civil society organizations as part of a community of practice that show how they have gained the ability to trigger international and performative processes of legality beyond the formal binding quality of norms. This is how civil society becomes an agent of social change in world politics, taking advantage of the decentralizing of authority characteristic of current global governance.

284  Research handbook on soft law

NOTES 1. The description of a community of practice is drawn from von Bogdandy and Urueña (2020, pp.402–403). 2. This understanding applies methodologies of literary critique. See Singh (2014, pp. 307–13): Diego López (2004, pp. 22–70) has, in turn, applied Harold Bloom’s ‘anxiety of influence’ to the appropriation of transnational legal theories in Latin America. 3. The Court acknowledges the first instance as an abstract control and the second as a concrete constitutionality control (see Judgment C-146/21, para. 104).

REFERENCES Adler, E., (2005). Communitarian International Relations: The Epistemic Foundations of International Relations. London; New York: Routledge. Alvarez, J. E., (2018). Reviewing the Use of ‘Soft Law’. Investment Arbitration. European International Arbitration Review. 7(2), 149–200. Bogdandy, von A. and Urueña, R., (2020). International Transformative Constitutionalism in Latin America. American Journal of International Law. 114(3), 403–42. Bonilla, D. M., (2015). International Law, Cultural Diversity, and the Environment: The Case of the General Forestry Law in Colombia. In: Gonzalez, C. G. et al. eds. International Environmental Law and the Global South. Cambridge: Cambridge University Press. pp. 508–28. Boyle, A. E., (1999). Some Reflections on the Relationship of Treaties and Soft Law. International & Comparative Law Quarterly. 48(4), 901–13. [viewed 20 June 2023] Available from: doi​.org/​10​.1017/​ S0020589300063739. Bradlow, D. D. and Hunter, D. B., (2020). Introduction: Exploring the Relationship between Hard and Soft International Law and Social Change. In: Bradlow, D. D. and Hunter, D. B. eds. Advocating Social Change through International Law. Leiden: Koninklijke Brill. pp. 1–15. [viewed 20 June 2023] Available from: doi​.org/​10​.1163/​9789004417021​_002. Brown, E. and Cloke, J., (2004). Neoliberal Reform, Governance and Corruption in the South: Assessing the International Anti-Corruption Crusade. ANTIPODE 36(2), 272–94. [viewed 20 June 2023] Available from: doi​.org/​10​.1111/​j​.1467​-8330​.2004​.00406​.x. Brown, E. and Cloke, J., (2011). Critical Perspectives on Corruption: An Overview. Critical Perspectives on International Business. 7(2), 116–24. [viewed 20 June 2023] Available from: doi​.org/​10​.1108/​ 17422041111128203. Bukovansky, M., (2006). The Hollowness of Anti-Corruption Discourse. Review of International Political Economy. 13(2), 181–209. [viewed 20 June 2023] Available from: doi​ .org/​ 10​ .1080/​ 09692290600625413. Carr, I. and Outhwaite, O., (2011). The Role of Non-Governmental Organizations (NGOs) in Combating Corruption: Theory and Practice. Suffolk University Law Review 44(3), 615–64. Colombian Constitutional Court Judgment SU.1150/00 (Cases T-186589, T-201615, and T-254941), Plenary Chamber, 30 August 2000. Colombian Constitutional Court Judgment C-291/07 (Case D-6476), Plenary Chamber 25 April 2007. Colombian Constitutional Court Judgment T-821/07 (Case T-1642563), Eight Revision Chamber, 5 October 2007. Colombian Constitutional Court Judgment T-740/11 (Case T-2.438.462), Eight Revision Chamber, 3 October 2011. Colombian Constitutional Court Judgment C-715/12 (Case D-8963), Plenary Chamber, 13 September 2012. Colombian Constitutional Court Judgment C-458/2015 (Case D-10585), Plenary Chamber, 22 July 2015. Colombian Constitutional Court Judgment C-035/16 (Case D-10864), Plenary Chamber, 8 February 2016. Colombian Constitutional Court Judgment C-330/16 (Case D-11106), Plenary Chamber, 23 June 2016.

Beyond norm entrepreneurs  285 Colombian Constitutional Court Judgment C-659/16 (Case D-11364), Plenary Chamber, 28 November 2016. Colombian Constitutional Court Judgment C-391/19 (Case D-12516), Plenary Chamber, 28 August 2019. Colombian Constitutional Court Judgment T-614/19 (Case T-6.518.300), Nine Revision Chamber, 16 December 2019. Colombian Constitutional Court Judgment C-146/2021 (Case D-13.933), Plenary Chamber, 20 May 2021. Convention on Biological Diversity. 1760 United Nations Treaty Series 79, entered into force 29 December 1993. Convention on Combating Bribery of Foreign Public Officials in International Business Transactions [online]. 1997. Entered into force on 14 February 1999. [Viewed 24 March 2022]. Available from: https://​www​.oecd​.org/​daf/​anti​-bribery/​ConvCombatBribery​_ENG​.pdf. Debevoise & Plimpton LLP, (2021). UN Guiding Principles on Business and Human Rights at 10. The Impact of the UNGPs on Courts and Judicial Mechanisms [online]. Office of the High Commissioner for Human Rights. [viewed 20 June 2023] Available from: https://​www​.ohchr​.org/​sites/​default/​files/​ Documents/​Issues/​Business/​UNGPsBHRnext10/​debevoise​.pdf. Eigen, P., (2002). Interview with K. Lundberg for a TI case study at the Kennedy School of Government, at Harvard University. March 4, 2002, Berlin. Epstein, C., (2008). The Power of Words in International Relations: Birth of an Anti-Whaling Discourse. MIT Press. Epstein, C., (2010). Moby Dick or Moby Doll? Discourse, or How to Study the ‘Social Construction of’ All the Way Down. In: Abdelal, R., Blyth, M. and Parsons, C. eds. Constructing the International Economy. Ithaca, London: Cornell University Press. pp. 175–93. Fakhri, M., (2015). Food as a Matter of Global Governance. Journal of International Law and International Relations. 11(2), 68–83. Finnemore, M. and Sikkink, K., (1998). International Norm Dynamics and Political Change. International Organization. 52(4), 887–917. Frischmann, B., (2003). A Dynamic Institutional Theory of International Law. Buffalo Law Review. 53(3), 679–809. Gómez Lee, M. I. and Maxfield, N., (2017). El papel de la sociedad civil en la Cumbre de la Tierra: gobernanza compleja. OPERA. No. 21, 135–56. [viewed 20 June 2023] Available from: https://​doi​ .org/​10​.18601/​16578651​.n21​.08. Hernández, G. I., (2019). International Law. Oxford: Oxford University Press. Hindess, B. (2005). Investigating International Anti-Corruption. Third World Quarterly. 26(8), 1389–98. [viewed 20 June 2023] Available from: doi​.org/​10​.1080/​01436590500336864. Howse, R. and Teitel, R., (2010). Beyond Compliance: Rethinking Why International Law Really Matters. Global Policy. 1(2), 127–36. Intergovernmental Panel on Climate Change (IPCC)., (1990). The First Assessment Report. Cambridge: Cambridge University Press. [viewed 20 June 2023] Available from: https://​www​.ipcc​.ch/​report/​ar1/​ wg1/​. Island Press., (1992). Earth Summit Bulletin, 20 April 1992. IUCN., (1989). World Conservation Strategy: Living Resource Conservation for Sustainable Development. Gland: IUCN. [viewed 20 June 2023] Available from: https://​doi​.org/​10​.2305/​IUCN​ .CH​.1980​.9​.en. Jasanoff, S., (2004). The Idiom of Co-Production. In Jasanoff, S. ed. States of Knowledge: The Co-Production of Science and Social Order. London, New York: Routledge. pp. 1–12. Katzarova, E., (2018). The Social Construction of Global Corruption: From Utopia to Neoliberalism. Cham: Palgrave Macmillan. Kaufmann-Kohler, G., (2010). Soft Law in International Arbitration: Codification and Normativity. Journal of International Dispute Settlement. 1(2), 283–99. [viewed 20 June 2023] Available from: doi​ .org/​10​.1093/​jnlids/​idq009. Kingsbury, B., (1998). The Concept of Compliance as a Function of Competing Conceptions of International Law. Michigan Journal of International Law. 19(2), 345-372.

286  Research handbook on soft law Klabbers, J., (2020). The Cheshire Cat That Is International Law. European Journal of International Law. 31(1), 269–84. [viewed 20 June 2023] Available from: doi​.org/​10​.1093/​ejil/​chaa018. Lave J., and Wenger, E., (1991). Situated Learning: Legitimate Peripheral Participation. Cambridge, New York: Cambridge University Press. López-Medina, D. and Sánchez-Mejía, A., (2008). La Armonización Del Derecho Internacional de Los Derechos Humanos Con El Derecho Penal Colombiano. Revista Colombiana de Derecho Internacional. 12, 317–51. López, D. E. M., (2004). Teoría Impura Del Derecho: La Transformación de La Cultura Jurídica Latinoamericana, 1. Bogotá: Legis. Lundberg, K., (2002). High Road or Low? Transparency International and the Corruption Perception Index. Kennedy School of Government, Harvard University, C-15-02-1658.0. Mauro, P., (1995). Corruption and Growth. Quarterly Journal of Economics. 110(3), 681–712. Mauro, P., (1996). The Effects of Corruption on Growth, Investment, and Government Expenditure. IMF Working Paper No. 96/98. [viewed 20 June 2023] Available from: dx​.doi​.org/​10​.2139/​ssrn​.882994. McCoy, J. L. and Heckel, H., (2001). The Emergence of a Global Anti-Corruption Norm. International Politics. 38(1), 65–90. [viewed 20 June 2023] Available from: doi​.org/​10​.1057/​palgrave​.ip​.8892613. Michaels, R., (2009). Global Legal Pluralism. Annual Review of Law and Social Science. 5, 243–62. Naím, M., (1995). The Corruption Eruption. The Brown Journal of World Affairs. 2(2), 245–61. North, D. C., (1990). Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press. North, D. C., (1991). Institutions. Journal of Economic Perspectives. 5(1), 97– 112. Pauwelyn, J., (2012). Informal International Lawmaking: Framing the Concept and Research Questions. In: Pauwelyn, J., Wessel, R. and Wouters, J. eds. Informal International Lawmaking. Oxford: Oxford University Press. pp. 13–34. Polzer, T. T., (2001). Corruption: deconstructing the World Bank discourse. DESTIN Working Paper Series, 1–18. [viewed 20 June 2023] Available from: https://​www​.files​.ethz​.ch/​isn/​138135/​WP18​.pdf. Posner, E. A. and Sykes, A. O., (2013). Economic Foundations of International Law. Harvard University Press. Posner, E. A., (2014). The Twilight of Human Rights Law. Oxford, New York: Oxford University Press. Quinche Ramírez, M. F., (2019). Bloque de Constitucionalidad, Soft Law y Control de Convencionalidad. In: Medina, M. R. ed. Crisis Del Estado Nación y de La Concepción Clásica de La Soberanía. Editorial Universidad del Rosario. [viewed 20 June 2023] Available from: https://​doi​.org/​10​.12804/​ tj9789587843538. Shleifer, A. and Vishny, R. W., (1993). Corruption. The Quarterly Journal of Economics. 108(3), 599–617. Sikkink, K. and Keck, M. E., (2014). Activists beyond Borders: Advocacy Networks in International Politics. Ithaca: Cornell University Press. Singh, S., (2014). Narrative and Theory: Formalism’s Recurrent Return. British Yearbook of International Law. 84(1), 304–43 TI., (2022). Current Operating Budget [online]. Transparency International. [Viewed 24 March 2022]. Available from: http://​www​.transparency​.org/​whoweare/​accountability/​current​_operating​_budget/​1/​. Toope, S. J. and Brunnée, J., (2010). Legitimacy and Legality in International Law: An Interactional Account. Cambridge: Cambridge University Press. Törnquist-Chesnier, M., (2004). NGOs and International Law. Journal of Human Rights. 3(2), 253–63. United Nations Convention against Corruption. 2003. 2349 United Nations Treaty Series 41, entered into force 14 December 2005. United Nations Framework Convention on Climate Change. 1992. 1171 United Nations Treaty Series 107, entered into force 21 March 1994. Urueña, R., (2014). Indicadores, derecho internacional y el surgimiento de nuevos espacios de participación política en gobernanza global. Revista Colombiana de Derecho Internacional. 12(25), 543–84. Urueña, R., (2018). Activism Through Numbers? The Corruption Perception Index and the Use of Indicators by Civil Society Organisations. In: Malito, D. V., Umbach, G. and Bhuta N., eds. The Palgrave Handbook of Indicators in Global Governance. Cham: Springer International Publishing. pp. 371–87.

Beyond norm entrepreneurs  287 Wang, H. and Rosenau, J. N., (2001). Transparency International and Corruption as an Issue of Global Governance. Global Governance. 7(1), 25–49. Wendt, A. E., (1987). The Agent-Structure Problem in International Relations Theory. International Organization. 41(3), 335–70. Wenger, E., (1998). Communities of Practice: Learning, Meaning, and Identity. Cambridge, New York: Cambridge University Press. World Meteorological Organization (WMO)., (1979). World Climate Conference: A Conference of Experts on Climate and Mankind. Declaration and Supporting Documents. Geneva: WMO. Available from [viewed 20 June 2023]: https://​library​.wmo​.int/​doc​_num​.php​?explnum​_id​=​3778.

19. The Open Method of Coordination (OMC): a hybrid tool of political leverage in the making Minna van Gerven and Sabina Stiller

1. INTRODUCTION The Open Method of Coordination (OMC) was institutionalized at the Lisbon European Council in 2000 as the European Union’s (EU) solution for policy coordination. Designed for areas where European member states agree to disagree, the OMC became an alternative to hard law regulation. The Treaty on the Functioning of the European Union (Article 153(2) TFEU) allows the EU to ‘adopt measures designed to encourage cooperation between Member States’ and with this mandate, the non-binding OMCs were introduced in diverse areas (such as employment, social inclusion, health, education and training, culture, youth, energy). However, the OMC process has only a legal basis in Article 148 TFEU regarding employment policy, whereas the other OMCs are based on more generic treaty authorizations for EU institutions to coordinate national policies (de Witte 2018, p. 209). Furthermore, the various OMCs differ regarding the participation of relevant stakeholders (i.e., the European Commission, European Parliament, civil society, interest representation partners) in the policy arena (Tholoniat 2010, p. 95). Despite this variation, the OMC can be generally characterized as a governance mechanism that aims to (summarized from Tholoniat 2010, p. 95 and Gornitzka 2018): 1. Identify and define general goals for the Union with specific timetables for Member State achieving them. 2. Establish EU level indicators and benchmarks for assessing (reported) member state progress towards the goals and to compare progress. 3. Translate common objectives to specific national and regional policies, strategies, plans, targets, measures, while being sensitive to national and regional differences. 4. Engage in periodic monitoring, joint evaluation, and peer reviews organized as mutual learning processes and to maintain peer pressure. The design of the OMC is led by the Commission together with a dedicated committee representing the member states (Tholoniat 2010, p. 95). Although its legal position remains to be debated (Dawson 2011; de Witte 2018), the OMC provides a common social space for European activities and ‘discourse regulation mechanisms’ (Jacobsson 2004, p. 356). Through these, the EU governance develops a shared understanding of problems and steers member state solutions in specific policy fields. The OMC’s central aspects include the development of joint definitions and concepts (e.g., employability, activation, active aging, lifelong learning), but also data sources (statistics, databases) and classification tools (indicators, targets), which can set standards for domestic policies. Moreover, launching the OMC was seen as a political 288

The Open Method of Coordination (OMC)  289 mandate to foster deliberative democracy (de la Porte and Nanz 2004). With this new eye to inclusiveness and transparency, the OMC involved different layers of government and included civil society and interest representation organizations in policymaking. Drawing on 20+ years of research on this soft governance instrument, this chapter views the OMC as a governance mechanism that includes and integrates both a policy and (mostly EU level) politics dimension and is aimed at coordinating and evaluating member state efforts, often in policy fields lacking EU legislative authority. As the chapter shows, it has become a force to be reckoned with in European policymaking as a result of gradual changes as a governance mechanism in the context of wider European Union developments. Scrutinizing the OMC over time and across different sectors offers interesting lessons for researching ‘soft law’ governance in the EU. It demonstrates the importance of understanding the OMC as a collection of hybrid and gradually changing policy tools and how this is linked to being a mode of governance that also generates policy leverage for both national and EU level actors. Second, the chapter illustrates the application of the OMC in two domains: social policy and energy and climate policy. Employment was the first sector in which the OMC was used in 2000, whereas energy and climate have turned successively into focal sectors for policy coordination to which the ‘hardened’ OMC has been applied more recently. Looking at these two sectors in comparison enables us to illustrate the development of the OMC over its lifespan. In addition, social policy and energy and climate policy have also been increasingly intertwined in the EU’s Green Deal strategy to ensure a just and inclusive transition to a sustainable future for EU citizens. Moreover, in the light of recent dramatic events in Ukraine (as of February 2022) and its already visible policy repercussions for EU economies and energy sectors, one may expect the need for policy coordination in these two domains to become more relevant for the EU. Through analysis of these OMCs, the chapter shows how the OMC has changed its position on the continuum between soft and hard law instruments. Thereby, it has become a tool of creative appropriation which continues to be reconfigured by the various EU-level and domestic actors that make use of it.

2.

TWISTS AND TURNS OF THE OMC

The Lisbon Summit introduced a new soft governance mechanism, OMC, as an alternative to the existing hard law instruments aiming to make the EU ‘the most competitive and dynamic knowledge-based economy in the world, capable of sustainable economic growth with more and better jobs and greater social cohesion’ (presidency conclusions, European Council 2000). The policy tools inherent in the OMC (including but not limited to guidelines, benchmarks, peer reviews) were, however, not novel. Rather, they were influenced by existing EU policy coordination strategies: the European Employment Strategy (EES) and the Broad Economic Policy Guidelines (BEPG). These, in turn, mimicked the soft coordination tools used by the Organization for Economic Cooperation and Development (OECD) to set non-binding prescriptions for national policies (van Gerven and Beckers 2009). The Lisbon Summit, however, equipped the OMC with ambitions to go beyond its predecessors: with a political mandate at the highest level, from the European Council, it set into stone management by objectives and targets for the coming decades. Doing so, it set course to converging standards via national and EU reporting and strengthening mutual learning at

290  Research handbook on soft law both levels. The first OMC was introduced in the field of employment policy, but the OMC procedures quickly extended to other social policy fields, including social inclusion, pension, health care, education, and culture policies. Each OMC included a different group of actors, e.g., EU actors (e.g., diverse Directorates-General (DG) of the Commission, national ministries, NGOs, interest representatives (trade unions, employers’ organizations)), and various networks engaged in the OMC processes at the EU and member state levels. Therefore, also, the exact toolset and (sometimes ad hoc) activities in new policy areas largely depended on the political interests and alignments in the specific policy area (Vanhercke 2020, p. 101), and the OMC was intentionally made as ‘a cookbook that contains various recipes, lighter and heavier ones’ (Vandenbroucke 2002, p. 9; Vanhercke 2020, p. 102). However, the OMC blueprint underlying the Lisbon strategy came under fire after only a few years. Being ‘about everything and thus about nothing’ (European Commission 2004, p. 16; Vanhercke 2020, p. 105), the Lisbon strategy was revised in 2005. Under the Lisbon two (European Commission 2005), the diverse guidelines underpinning various social OMC procedures at the time were integrated into a ‘meta-OMC’ (Dehousse 2004; Vanhercke 2020, p. 106). At the same time, new tools (including annual reporting on the integrated guidelines) were introduced, indicators were simplified, peer review procedures were extended to new areas, e.g., PROGRESS-peer reviews in social inclusion (Tholoniat 2010). These changes set the course towards the extended monitoring and intensified mutual learning capacities of the OMC. The design of this next big European strategy took place in the context of the global financial-economic crisis. The new blueprint of ‘Europe2020’, the European Semester, was centred around multilateral surveillance instruments and put forward a cycle of coordination of socioeconomic policies. It included three integrated pillars: (1) macroeconomic surveillance through the Macroeconomic Imbalance Procedure (MIP), (2) fiscal surveillance under the Stability and Growth Pact (SGP), and (3) thematic coordination, including the areas of innovation and research, employment, education, social inclusion (European Commission 2010). The OMC was integrated in the European semester in the third pillar on thematic coordination. To facilitate a coordinated exit from the crisis, the European Semester added a new tool to the existing toolkit of policy coordination. Country-level recommendations (CSRs) were now issued by the European Commission to specify how countries should or could improve their performance. Unlike the other two pillars (MIP and fiscal surveillance), the CSRs for thematic coordination (linked to the OMC) did not have the ability to result in preventive or corrective measures (e.g., imposing fines, structural reporting requirements). Yet, CSRs were tailored to highlight national challenges annually and set a more politically coercive stance to steer member states out of the crisis. For example, for employment, it has been found that the CSRs went much further than improving the status quo, calling for structural reforms of national employment systems (Bekker 2021), in the area where the EU formally had no competencies. Furthermore, new indicators and social scoreboards were launched to intensify the multilateral surveillance agenda within the EU (see e.g., Vanhercke 2020). Alongside the Europe 2020 strategy, in EU climate and energy policy a ‘2020’ Framework was developed, which consisted of a mix of hard law and policy coordination instruments with the potential to give Commission recommendations more ‘bite’. It comprised three headline benchmarks or targets for 2020: 20 per cent greenhouse gas (GHG) emission reduction, a 20 per cent share of renewable energy in final energy consumption (both binding), and a 20 per cent improvement in energy efficiency. The framework was implemented through three direc-

The Open Method of Coordination (OMC)  291 tives: on the EU emissions trading system (ETS) (Directive 2018/410), on renewable energy (Directive 2018/2001) and energy efficiency (Directive 2018/2002); and an ‘effort-sharing’ decision on reduction targets for member states’ GHG emissions outside the ETS (Regulation 2018/842). The subsequent framework includes updates of headline targets for 2030 and relies on the previous three, partially updated, directives (on the ETS, renewable energy, and energy efficiency) and three regulations (effort-sharing in GHG emission reduction; inclusion of greenhouse gas emissions and removals from land use, land-use change, and forestry (Regulation 2018/841)) and on Governance of Energy Union and Climate Action (Regulation 2018/1999). The more ambitious targets included: reducing GHG emissions by at least 40 per cent (from 1990 levels), a renewable energy target for 2030 of 32 percent of final energy consumption, and a new indicative energy efficiency target of 32.5 per cent (Oberthür 2019, p. 18). Importantly, the 2018 Governance Regulation established an integrated framework of policy coordination for planning, reporting, and review related to the 2030 Framework (including new elements, such as energy security). It requires each member state to prepare ten-year National Energy and Climate Plans (NECPs), that include existing and planned policies and measures, including plans to the EU-wide renewable energy and energy efficiency targets (de la Porte and Stiller 2020). From 2015 onwards, the Juncker Commission has relied on the OMC or OMC-like procedures in the adoption of various flagship initiatives. For instance, OMC has been formally linked to the European Pillar of Social Rights (EPSR) launched at the 2017 Gothenburg Summit (Vanhercke 2020). At the same event, a new social scoreboard was adopted with the intention to give a more ‘social colour’ to the EMU and to ‘socialize’ the Semester (Zeitlin and Vanhercke 2018). Moreover, since late 2018, OMC-like procedures as part of the European Semester have been incorporated in the Regulation on the Governance of the EU Energy Union and Climate Action (de la Porte and Stiller 2020; Knodt et al. 2020). Next to setting common rules for member states for planning, reporting, and monitoring, they also ensure that EU planning and reporting are synchronized with climate action cycles for submitting national plans under the UN (UNFCCC) global Paris Agreement on the reduction of greenhouse gases (European Commission, no data). The Governance Regulation, along with previously mentioned EU legislation, became an important part of the EU’s comprehensive Green Deal Strategy announced in 2019 by the Von der Leyen Commission. The European Green Deal involves a broad roadmap to make the EU economy and society carbon neutral by 2050 and is already seen by some as a ‘building block to the European economic model, alongside the single market and economic and monetary union’ (Bongardt and Torres 2022, p. 170). The Covid-19 outbreak led to major delays in EU governance since policy priorities shifted to mitigating the health, economic, and social consequences of the pandemic. However, after the initial shock, EU activities were resumed and initiatives such as the implementation of action plans under the EPSR and an update of the social scoreboard were back on the agenda of the Porto Summit in 2021 (Vanhercke and Spasova 2022). That summit also defined three new targets for Social Europe (monitored through the new social scoreboard): at least 78 per cent of the population (aged 20–64) should be employed by 2030; at least 60 per cent of the adult population should participate in education or training each year, and the number of people threatened by poverty or social exclusion should be reduced by at least 15 million by 2030. Importantly, the unprecedented crisis provoked ambitious policy

292  Research handbook on soft law responses. In 2020, the Commission introduced a new long-term (but temporary) EU budget strategy called NextGenerationEU, providing the largest stimulus packet ever financed in the EU (totalling 1.8 trillion Euros) to build a greener, more digital, and more resilient Europe (Bruyninckx et al. 2022, p. 76). At the core of this is the Recovery and Resilience Facility (RRF) that comes with great carrots, but also sticks for member states to stimulate reforms that are directly and indirectly linked to OMC, as explained below.

3.

LESSONS LEARNED ABOUT THE OMC

The evolution of the OMC as presented in the preceding section provides us with interesting lessons. First, it illustrates five aspects of how the OMCs have changed regarding their underlying policy tools and how these tools have gained importance in overall policy coordination, monitoring, and surveillance at the EU level in Section 3.1. Second, it reflects an evolving understanding of the OMC as a governance mode that generates policy leverage for both national and EU-level actors, as discussed in Section 3.2. 3.1

OMC as a Hybrid Toolset for Soft Governance

The spread of the OMC over time constitutes a convincing case of ‘a spillover effect’ of European integration. Following the experiences from the EES from the Amsterdam Treaty (1997), the Lisbon Summit introduced the first formal template of OMC in this area, which quickly extended to several related social areas, such as social inclusion, pensions, and health care. In 2010, OMCs were introduced to youth policy and education policy to foster convergence among member states and later to energy and climate change policy, which were included in the Europe 2020 strategy alongside employment and social inclusion policy. The key aims and benchmarks on energy and climate change policy were first agreed for a ten-year time horizon in both areas, requiring member states to report, although briefly (Da Graça Carvalho 2012; Helm 2014). Later, softer steering and new governance modes have also been adopted in the area of climate change adaptation (Bauer et al. 2012) and in environmental policy (Twena 2012). 3.2

Maturation of the OMC as Governance Instruments

Research also illustrates various phases of the maturation of the OMC, ranging from initial experimentation to streamlining and maturing of the governance process (see e.g., Tholoniat 2010; Pochet 2020; Vanhercke 2020). In the early phase of the OMC in employment (2000–04), tools and activities were developed in an experimental manner. Later, integrated social and economic OMCs (2005–09) became more streamlined and had greater steering capacity. The intergovernmental governance of the Eurozone at the time of crisis (2010–14) was seen to bring an essence of more technocratization and alignment of the various OMCs. In the most recent period (2015–present), the OMC is seen to have stabilized as a policy instrument as part of other EU instruments such as European Semester and EPSR. The maturation of the OMC is typically connected to the scholarly (and political) discussion whether, and to what extent, the OMCs bring about domestic policy change. There is a large body of research engaging in this ‘believer vs non-believer’ discussion of the OMC’s effec-

The Open Method of Coordination (OMC)  293 tiveness (see Zeitlin et al. 2014). A generic response is that – as much as the various OMCs embrace diversity – they also vary as to their effectiveness. The goals of the employment OMC have been very visible in the European and domestic agendas and hence resonate in the numerous domestic debates on activation, flexicurity, and lifelong learning. In contrast, other social OMCs, lacking the same formal status of the employment OMC, have been seen as less visible and thus less effective (Copeland and Daly 2014). This difference in visibility is partially explained by the greater relevance of employment affairs in economically driven governance (Lammers et al. 2018), but also by the sensitivity of issues such as health care, education (see e.g., Gornitzka 2018) and poverty prevention (Daly 2008, 2012), all strongly considered as national competences. 3.3

Qualitative Change in Underlying Interests Driving OMCs

A lot of research focusing on the social OMCs observes a ‘hardening’ of the OMC during and in the aftermath of the economic crisis (2010–14) (e.g., Barbier 2010; Zeitlin and Vanhercke 2018). In the sense of its underlying interests, neoliberal values concurring with the dominance of right or centre-right governments both at the national and EU levels have been seen to shift the substance of OMC from social interests (fighting poverty, inequality) to prioritizing economic and employment-supply-related guidelines. The analyses of such developments are often confounded to the theorization of the fundamental struggle in European integration, as to the asymmetry between economic or neoliberal and social interests (Scharpf 2010). Or they are related to the (changing) political division of power in the member states (Schmidt 2020) and its influence on the political support for Social Europe. The heyday of the ideal of ‘Social Europe’ rebalancing the social and economic interest is often related to the introduction of the Employment Chapter in the Amsterdam Treaty in 1997. This was followed by the launch and expansion of OMCs under the hegemony of pro-welfare governments across Europe. A qualitative change towards the dominance of neoliberal values and fiscal priorities is prescribed to the post-Lisbon period when economic and social guidelines were integrated, and social interests were side-lined or at least marginalized (Barbier 2012; Copeland and Daly 2014). More recent research suggests that social interests have been rejuvenated (Copeland and Daly 2018; Zeitlin and Vanhercke 2018) through a closer link between the OMC and the European Pillar of Social Rights as well as the attempts to back up the OMC governance with significant EU funding through the RRF in 2021. The latter provides member states with grants and loans linked to detailed national reporting – including social indicators – in Recovery and Resilience Plans (Verdun and Vanhercke 2022, p. 103). Yet, the existence and survival of Social Europe remain questioned and provoked scholarly debate around the inherent imbalances of political power and interests (Scharpf 2008), also impacting the OMC. 3.4

Hardening of the OMC as a Governing Tool

As a governance tool, studies illustrate the gradual hardening of the OMC. Although still lacking sanctions, the current OMC gives the European Commission more leverage to push changes at the national level. A good example of gradual hardening is the annual reporting procedure inherent in the OMC. What started as voluntary reporting by the member states, causing at best ‘learning by irritation’ or ‘window dressing’ by the member states, the OMC has become a hard currency by strongly relying on quantified targets and ‘learning by

294  Research handbook on soft law numbers’ (Mabbett 2007). Member states are increasingly scrutinized publicly as part of the OMC review processes and the generic guidelines from the Lisbon strategy have evolved into more stringent targets pushing member states’ compliance by much more detailed reporting on their performance. Such intensified monitoring is driven by the Commission’s intentional development of indicators and scoreboards and facilitated by increased data available for comparison. Although the voluntary OMC was previously seen as being comparatively weak on surveillance and enforcement, the new thematic coordination since 2010 undoubtedly includes a stronger arsenal for this. The OMC today comes with a political twist that goes beyond soft coordination by shaming, naming, and faming. The progressive development of indicators (e.g., the scoreboard of key social and employment indicators as part of the European Semester, the new social scoreboard as part of the EPSR, and, recently, indicators in the Recovery and Resilience Scoreboard) has meant that, on the one hand, national policymakers must engage in the process of translation of EU policy to domestic ones, as they are required to compare indicators with national data, and, on the other hand, they must anticipate on those at the national level, which leads to agenda setting (Mabbett 2007). The two-level policy dialogue strengthens the ownership of the OMC among administrative actors in the Commission and the member states. In the field of energy and climate, national reports (NECPs), for instance, are to provide a basis of dialogue among member states and the Commission, while the CSRs enable the Commission and member states to agree together on core policy measures to meet the EU’s strategic aims in this area for 2030 and 2050. This process entails ownership at EU and member state levels through a structured, yet contextualized policy dialogue (de la Porte and Stiller 2020). The targets and benchmarking exercises have led to constant monitoring and surveillance of member states, and the political interest for such intergovernmental scrutiny has grown specifically since the economic crisis in the 2010s. For example, with the 2018 Regulation on Governance a hardening of policy tools compared to the previous report in the European Semester has been observed in energy and climate policy (Knodt et al. 2020). Complementary aims and benchmarks are to be monitored on an on-going basis in the Energy Union as this ensures political visibility, enables the EU to compare member states in league tables, and sends a clear signal to member states about the strategic direction of the EU. In the Energy Union, member states are required to set national targets, as for instance in the EES, making national policy change to reach these aims more likely (de la Porte and Stiller 2020). Furthermore, it has been observed that the EU Commission has given more stringent feedback on CSRs in the social field (see also Bekker 2021; Verdun and Vanhercke 2022), particularly on how member states can balance their economic budgets in the face of EU fiscal discipline (Copeland and Daly 2018). At the same time, mutual learning procedures have matured over time, peer reviews have been formalized, and their impact has grown as instruments of mutual learning, but also of convergence (van Gerven and Ossewaarde 2018). 3.5 Hybridization The hardening of the OMC is often related to another development, namely an increased linkage of OMC guidelines and targets to other (more conditional) European governance instruments, such as the European Semester, the EPSR, or the Energy Union’s reporting obligations, as discussed above. In this way, the ‘softer’ EU governance instruments have become increasingly hybridized. EU social policy was traditionally constructed through three

The Open Method of Coordination (OMC)  295 EU level instruments: (1) legal regulation (a non-negotiated or negotiated form), (2) financial instruments such as the ESF, and (3) the OMC (Vanhercke et al. 2011). The OMC was applied to areas such as social policy where, because of lack of competence, formal EU legislation did not exist. This allowed the implementation of more ‘flexible’ elements, leaving member states considerable discretion to adapt to EU level goals. To us, this development illustrates a ‘hybridization’ of EU policy tools: since the creation of the ‘voluntary’ OMC, different policy instruments have become more interdependent and the OMC has been closely coupled to other instruments, giving it ‘more bite’ than previously. To give some examples, the social inclusion OMC has been closely linked to the Cohesion Fund and other Structural Funds by, for instance, making the European Social Fund (ESF) conditional on member states sticking to the priorities set in the OMC and by tying domestic use tightly to programmatic ESF-investments. Furthermore, the OMC has been formally linked to the European Semester, the EPSR, and, most recently, to the RRF strategies responding to economic turmoil in the wake of the Covid-19 pandemic (Verschraegen et al. 2011; van Gerven et al. 2014; Verdun and Vanhercke 2022). Finally, RRF funding is linked to CSRs and earmarked to facilitate green and digital transitions: climate-related expenditure was stipulated to make up at least 37 per cent of each national RRF strategy. Although no social targets are explicitly stipulated, the link of RRF to national reform programs and to social expenditure indicators are expected to drive domestic politics and reforms (Vanhercke and Spasova 2022, p. 158; see also Alcidi and Corti 2022). In sum, these developments constitute a shift from ‘as good as it gets’ coordination for politically sensitive no-go zones to a hardening of the available OMC toolset for increasing steering of member states’ policies. It illustrates the changing position of OMC governance on the continuum between soft and hard policy instruments, as will be further discussed below.

4.

OMC AS A POLITICAL (POLICY) LEVERAGE TOOL

In addition to its ability to generate policy alternatives and steer the substance of national policies, as discussed above, the OMC as a governance mode is deeply ingrained in multi-level decision-making. The evolution previously described also provides three lessons on its ability to generate policy leverage for both EU and national-level actors. 4.1

Reciprocal Influence of the OMC and Creative Appropriation by Multi-level Actors

OMCs have been found to affect member states’ policy ideals and reforms – in a reciprocal manner. It has been long acknowledged that the relationship between the EU and its member states is not ‘either bottom-up or top-down, rather it is a two-level game’ (Büchs 2008). Therefore, member states are both takers and shakers of EU-level policy (Stiller and van Gerven 2012). The OMC, as a reciprocal tool subject to periodic revisions, provides governments and other stakeholders with opportunities to influence its content in line with their preferences (López-Santana 2006). Similarly, OMC guidelines may be used to legitimize controversial domestic reforms otherwise not undertaken (Stiller and van Gerven 2012) and seek broader European societal support.

296  Research handbook on soft law Against this background, as Graziano et al. (2011, p. 9) have argued, member state actors appropriate and redefine EU resources to advance their own agendas. First, the OMC offers opportunities for member states to carry out reforms in line with their preferences and strategic objectives, as long as these are covered by the OMC’s substance (Stiller and van Gerven 2012). Heidenreich and Zeitlin (2009) call this process ‘creative appropriation’ by domestic actors for governments’ purposes. In that process, the open, deliberative, and negotiated character of the OMC serves as a ‘creative resource’ through providing formal and informal platforms, networks, and links to tangible and ideational resources at the EU and member state level (Heidenreich and Zeitlin 2009, p. 233). EU Council presidencies, for instance, are seen to play a significant role in launching new policy projects and agenda-setting. The Dutch presidency used the momentum to launch the EES in 1997 and to extend the activation policy paradigm into national social policies (Stiller and van Gerven 2012), while the German presidency oversaw the inclusion of the Employment, Social Policy, Health and Consumer Affairs (EPSCO) council in RRF decisions (Verdun and Vanhercke 2022), paving the way for a stronger presence of social policy indicators in EU surveillance and monitoring practices. Across EU institutions policy entrepreneurs continue to strive for a sound anchoring of the policy field (e.g., Energy Union) at the EU level to counter possible volatility in political priorities. The political commitment of national actors, i.e., prime ministers and ministers for energy and climate, who are central in prioritizing energy transitions towards carbon-free economies, is often seen as central in enabling a ‘harder’ impact of policy coordination in member states (De la Porte and Stiller 2020). Much of the political influence happens inside the OMC. In the case of the social OMC, for instance, several studies (Copeland and Daly 2018; Zeitlin and Vanhercke 2018; Vanhercke 2020) point to intense participation of EU-level stakeholders to reinvigorate the social dimension in the European Semester through the OMC. Research points to the heroism of certain EU committees, e.g., Vanhercke (2020) points towards the role of the Social Protection Committee, together with Commissioner Andor, in keeping Social Europe and social indicators on the (economically-driven) agenda after the financial-economic crisis in the 2010s, whereas Gornitzka (2018) highlights the role of the Commission’s DG Education and Culture in working around national sensitivities in education, and pushing forward those practices and routines in the OMC education that have bolstered the OMC’s legitimacy and impact on member states over time. Concerning the Energy Union, previous EU Commissioners engaged in challenging member states and the Council to develop more ambitious plans when working out first final NECPs assessed in 2020. With draft plans falling short in terms of renewables and energy efficiency contributions, Vice-President for the Energy Union Maroš Šefčovič called upon member states to give ‘national plans a solid final push’ by showing ‘stronger ambition, more policy detail, better specified investment needs, or more work on social fairness’ at the service of clarity and predictability as competitive advantage for European energy and climate policy. Equally, Commissioner for Climate Action and Energy, Miguel Arias Cañete, in the face of the EU’s ambitious goal of climate neutrality in 2050, invited the Council ‘to open a debate around the main priorities identified by the Commission and help ensure that the final plans contain an adequate level of ambition’ (European Commission 2019). Partially because of stepped-up final national plans, the Commission’s State of the Energy Union Report for 2021 reported a few encouraging trends but also stressed that greater efforts will be required to reach

The Open Method of Coordination (OMC)  297 the 2030 goal of cutting net emissions by at least 55 percent and achieving climate neutrality by 2050.1 While examples of a ‘direct European impact’ on domestic agendas and reforms through the OMC are scarce, ample research effort has also been devoted to the indirect impact of OMC processes as catalysts of policy learning and collaboration. Zeitlin et al. (2014, p. 14) for instance suggest that the OMC has an impact through ‘capillary effects’, where domestic actors linked through European networks advocate ideas and knowledge to national policy-making arenas. Following this line of argument, people working on specific issues get exposed to new ideas and values, which are then diffused into national policy processes and adapted to national contexts (Zeitlin et al. 2014, p. 14). 4.2

Coercive Conditionality

The inclusion of the OMC in the European Semester and the European Pillar of Social Rights is expected to provide a stronger mandate for EU actors (such as the Commission’s DG Employment) to call for greater consideration of social and employment challenges in the wider EU governance (Vanhercke 2020, p. 114) Furthermore, the recent linkage to the RRF strengthens EU conditionality in the member states (Verdun and Vanhercke 2022; Bekker 2022). These factors build on the earlier alignment of the OMC with CSRs and structural funds (as discussed in section 3.1) and amplify OMCs’ potential Europeanization effect and, consequently, EU-level political steering power. With such coercive conditionality (e.g., in the OMC guidelines coupled to ESF usage, see Verschraegen et al. 2011; van Gerven et al. 2014), the European Commission can exercise influence on national governments through the requirements imposed on member states implementing EU policies. 4.3

Engaging Civil Society Actors and Keeping the OMC on Political Agendas

Despite the noble aim of inclusiveness set in the Lisbon Council decisions, the OMC is commonly seen to be biased towards (technocratic) expert deliberation and closed for public deliberation (see e.g., de la Porte and Nanz 2004; Radulova 2007; Radaelli 2008). The technocratic turn of governance is particularly seen to occur amid the financial crisis at the end of the 2000s and the launch of the European Semester in 2010. Certain stakeholders, such as civil society organizations and social partners, were intentionally included in the OMC in the design and monitoring of economic and social policies (see also Carella and Graziano 2022 for their intentional involvement in the design of EPSR). However, public engagement and social dialogue have had varying successes. Social partners have been formally involved in policymaking, whereas other civil society actors are often consulted in an ad hoc manner, and to a varying extent depending on the OMC in question (Verdun and Vanhercke 2022). Yet, engagement in the social dialogue has been troublesome, particularly during the Barroso Commission (2005–14) term, but been ‘revived’ in the Juncker (2015–19) and von der Leyen’s (2019–) Commission terms (Tricart 2020). In the area of energy and climate policy, for instance, the inclusiveness of the civic and social actors is being addressed by a public consultation process, as member states were required to ‘consult citizens, businesses and regional authorities in the drafting and finalization process (of final NECPs)’ (European Commission, no date-b). The Governance Regulation requires member states to give early and effective opportunities to participate in the process of

298  Research handbook on soft law preparing NECPs, including setting reasonable timeframes to allow the public to be informed, participate and express its views. Moreover, a summary of the public’s views needs to be attached to the NECP. Next to this, the Aarhus Convention (1998) ensures a number of rights of the public in decision-making on environmental matters. Especially, this concerns transparent and fair arrangements, necessary information to be provided to the public, ensuring reasonable timeframes, and taking due account of outcomes of public participation (European Commission, no date-b). In this way, as Jacobsson and Johansson (2009) suggest, OMCs can fuel micro-politics by creating new patterns of cooperation, e.g., between national ministries and social NGOs as well as within social NGO networks. Certain kinds of funding, such as ESF funding, are conditional on the partnership principle: the ESF should be implemented through a multi-level and integrated approach, bringing together a range of state and non-state actors on different levels of government (van Gerven et al. 2014). Therefore, the OMC promotes inclusiveness by steering towards certain kinds of partnerships.

5. CONCLUSION With the launch of the OMC, EU policymaking was extended to previous no-go zones of policy at the margins of European competencies. The European Commission was given a mandate to seek alternatives to hard law, and experimentation and mutual learning was promoted to steer policy convergence in the member states towards shared EU priorities. In the last twenty odd years, the OMC has become the ‘European way’ in finding agreements in policy areas where member states hold different views and ‘the recipe’ to substantive steering and nudging member states towards preferred policy directions. This chapter discussed the evolution of the OMC and its instruments and the role of different level (national and EU) actors herein. It highlights that the OMCs are governance mechanisms that include and integrate policy and (mostly EU-level) politics dimensions with the aim of coordinating and evaluating efforts of member states in the sensitive policy field. The OMC has undergone gradual changes following the EU developments. As the literature demonstrates, the OMC is a collection of hybrid and over-time changing policy tools, but it goes beyond being merely an instrument. It provides a mode of governance that fosters creative appropriation and policy leverage for both national and EU-level actors. By illustrating its application in two different policy domains, employment and energy and climate, the chapter has shown that the OMC has been changing position on the continuum between soft and hard law instruments, becoming a governance mechanism that continues to be reconfigured by the actors that are making use of it. In public policy studies, initially, the use of hard or soft instruments was discussed in a dichotomy of standard legislation on the one hand and soft governing tools on the other (Jordan et al. 2005). Here, hard instruments (often legislation) are seen as more authoritative, standing for ‘top-down’ power structures, whereas soft governing instruments, like the OMC, are seen as more deliberative or consensual and linked to a ‘bottom-up steering logic’. Legislators can delegate the implementation and enforcement of hard instruments to courts, which may lead to sanctions to ensure compliance; however, soft tools cannot lead to delegation of enforcement to others because a legal basis for this is lacking. As a result, hard policy tools are perceived as more precise and fixed in their content, while soft tools are vaguer and more flexible (Blomqvist 2016, p. 269). However, this

The Open Method of Coordination (OMC)  299 dichotomy of hard and soft policy tools proved ultimately unhelpful in grasping differences in tools and nuances of European governance, as EU legislation (‘hard law’) may be imprecise and allow room for manoeuvre for member states, while recommendations may lead to sanctions. This implies that the boundaries between the two types are already blurred, and leading scholars suggested to treat hardness and softness of governance tools rather as a scale (for an overview, see Knodt and Schoenefeld 2020, p. 764). While various forms of soft governance have been around for a long time, there is a growth of new policy elements to ‘harden’ soft governance arrangements. These new forms of ‘harder’ soft governance (HSG) vary in the degree of hardness in different settings. Knodt and Schoenefeld (2020) have drawn together several studies for this development of ‘hardening’ from diverse fields and policy instruments including the OMC, the European Semester, climate policy monitoring, the EU Energy Union, and in other international organizations. The OMC, being part of these governance architectures, has hardened over time and keeps on adjusting to the needs and pressures of the European Union’s political landscape, also in the future.

NOTE 1. While renewables overtook fossil fuels as major power source in the EU for the first time in 2020 (generating 38 per cent of electricity, compared to 37 per cent for fossil fuels), the notable 10 per cent drop in greenhouse gases in 2020 (amounting to an overall emission reduction of 31 per cent compared to 1990) will be critically evaluated in 2022 due to the post-Covid economy recovery. See European Commission (2021).

REFERENCES Alcidi, C., and Corti, F., (2022). The EU response to Covid-19: breaking old taboos?. In: B. Vanhercke and S. Spasova, eds. Social policy in the European Union: state of play 2021. Re-emerging social ambitions as the EU recovers from the pandemic. Brussels: European Trade Union Institute (ETUI) and European Social Observatory (OSE). pp. 39–60. Barbier, J.-C., (2010). ‘Stratégie de Lisbonne’: les promesses sociales non tenues. CES Working Paper, 2010.18 [online]. Paris: Centre d’Economie de la Sorbonne. [Viewed 6 May 2022]. Available from: https://​halshs​.archives​-ouvertes​.fr/​halshs​-00462456/​document. Barbier, J.-C., (2012). Tracing the fate of EU ‘social policy’: changes in political discourse from the “Lisbon Strategy” to “Europe 2020”. International Labour Review. 151(4), 377–399. Bauer, A., Feichtinger, J. and Steurer, R., (2012). The governance of climate change adaptation in 10 OECD countries: challenges and approaches. Journal of Environmental Policy & Planning. 14(3), 279–304. Bekker, S., (2021). Hardening and softening of country-specific recommendations in the European Semester. West European Politics. 44(1), 114–133. Bekker, S., (2022). The social dimension of EU economic governance after the Covid-19 pandemic: exploring new interlinkages. Italian Labour Law E-Journal, 15(1S), 1–14. Blomqvist, P., (2016). Soft and hard governing tools. In: C. Ansell and J. Torfing, eds. Handbook on theories of governance. Cheltenham: Edward Elgar. pp. 267–278. Bongardt, A., and Torres, F., (2022). The European Green Deal: more than an exit strategy to the pandemic crisis, a building block of a sustainable European economic model. Journal of Common Market Studies. 60(1), 170–185. Bruyninckx, H., Karadeniz, G. and Martin, J., (2022). The European Green Deal: opportunities and prospects after the Covid-19 crisis. In: B. Vanhercke and S. Spasova, eds. Social policy in the European

300  Research handbook on soft law Union: state of play 2021.Re-emerging social ambitions as the EU recovers from the pandemic. Brussels: European Trade Union Institute (ETUI) and European Social Observatory (OSE). pp. 61–80. Büchs, M., (2008). The open method of coordination as a ‘two-level game’. Policy & Politics. 36(1), 21–37. Carella, B. and Graziano, P., (2022). Back to the future in EU social policy? Endogenous critical junctures and the case of the European pillar of social rights. JCMS: Journal of Common Market Studies, 60(2), 374-390. Copeland, P. and Daly, M., (2014). Poverty and social policy in Europe 2020: ungovernable and ungoverned. Policy & Politics. 42(3), 351–365. Copeland, P. and Daly, M., (2018). The European Semester and EU social policy. Journal of Common Market Studies. 56(5), 1001–1018. Da Graça Carvalho, M., (2012). EU energy and climate change strategy. Energy. 40(1), 19–22. Daly, M., (2008). Whither EU social policy? An account and assessment of developments in the Lisbon social inclusion process. Journal of Social Policy. 37(1), 1–19. Daly, M., (2012). Paradigms in EU social policy: a critical account of Europe 2020. Transfer: European Review of Labour and Research. 18(3), 273–284. Dawson, M., (2011). New governance and the transformation of European law: coordinating EU social law and policy. Cambridge: Cambridge University Press. De la Porte, C. and Nanz, P., (2004). The OMC – a deliberative-democratic mode of governance? The cases of employment and pensions. Journal of European Public Policy. 11(2), 267–288. De la Porte, C. and Stiller, S., (2020). Lessons about the ‘harder’ elements of OMC governance for the EU Energy Union. Journal of Environmental Policy and Planning. 22(6), 830–842. De Witte, B., (2018). The place of the OMC in the system of EU competences and sources of law. European Papers – A Journal on Law and Integration. 3(1), 207–213. Dehousse, R., (2004). La méthode ouverte de coordination. quand l’instrument tient lieu de politique. In: P. Lascoumes and P. Le Galès, eds. Gouverner par les instruments. Paris: Presses de Sciences Po. pp. 331–356. European Commission., (2004). Facing the challenge: the Lisbon strategy for growth and employment: report from the High Level Group chaired by Wim Kok [online]. Luxembourg: Publications Office. [Viewed 6 May 2022]. Available from: https://​op​.europa​.eu/​s/​v9qV. European Commission., (2005). Working together for growth and jobs: a new start for the Lisbon Strategy. Communication to the Spring European Council from President Barroso in agreement with Vice-President Verheugen (COM/2005/24 final) [online]. Brussels: Commission of the European Communities. [Viewed 11 February 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​ EN/​ALL/​?uri​=​CELEX:​52005DC0024. European Commission., (2010). EUROPE 2020: a strategy for smart, sustainable and inclusive growth (COM/2010/2020 final) [online]. Brussels: European Commission. [Viewed 11 February 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​en/​ALL/​?uri​=​CELEX:​52010DC2020. European Commission., (2019). Energy Union: Commission calls on Member States to step up ambition in plans to implement Paris agreement [online]. Press release, IP/19/2993. Brussels: European Commission. [Viewed 2 May 2022]. Available from: https://​ec​.europa​.eu/​commission/​presscorner/​ detail/​en/​IP​_19​_2993. European Commission., (2021). State of the Energy Union 2021 – contributing to the European Green Deal and the Union’s recovery (COM/2021/950 final) [online]. Brussels: European Commission. [Viewed 3 May 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​?uri​=​COM:​ 2021:​950:​FIN. European Commission., (no date-a). Governance of the Energy Union and Climate Action [online]. European Commission. [Viewed 2 May 2022]. Available from: https://​ec​.europa​.eu/​clima/​eu​-action/​ climate​-strategies​-targets/​progress​-made​-cutting​-emissions/​governance​-energy​-union​-and​-climate​ -action​_en. European Commission., (no date-b). National energy and climate plans (NECPs) [online]. European Commission. [Viewed 3 May 2022]. Available from: https://​ energy​ .ec​ .europa​ .eu/​ topics/​ energy​ -strategy/​national​-energy​-and​-climate​-plans​-necps​_en.

The Open Method of Coordination (OMC)  301 European Council., (2000). Presidency conclusions, Lisbon European Council, 23 and 24 March 2000 [online]. Lisbon: European Council. [Viewed 11 February 2022]. Available from: https://​ www​ .europarl​.europa​.eu/​summits/​lis1​_en​.htm. European Parliament and Council Directive (EU) 2018/410 amending Directive 2003/87/EC to enhance cost-effective emission reductions and low-carbon investments, and Decision (EU) 2015/1814 (2018) Official Journal L 76, p. 3. European Parliament and Council Regulation (EU) 2018/841 on the inclusion of greenhouse gas emissions and removals from land use, land use change and forestry in the 2030 climate and energy framework, and amending Regulation (EU) No 525/2013 and Decision No 529/2013/EU (2018) Official Journal L 156, p. 1. European Parliament and Council Regulation (EU) 2018/842 on binding annual greenhouse gas emission reductions by Member States from 2021 to 2030 contributing to climate action to meet commitments under the Paris Agreement and amending Regulation (EU) No 525/2013 (2018) Official Journal L 156, p. 26. European Parliament and Council Regulation (EU) 2018/1999 on the Governance of the Energy Union and Climate Action, amending Regulations (EC) No 663/2009 and (EC) No 715/2009 of the European Parliament and of the Council, Directives 94/22/EC, 98/70/EC, 2009/31/EC, 2009/73/EC, 2010/31/ EU, 2012/27/EU and 2013/30/EU of the European Parliament and of the Council, Council Directives 2009/119/EC and (EU) 2015/652 and repealing Regulation (EU) No 525/2013 of the European Parliament and of the Council (2018) Official Journal L 328, p. 1. European Parliament and Council Directive (EU) 2018/2001 on the promotion of the use of energy from renewable sources (2018) Official Journal L 328, p. 82. European Parliament and Council Directive (EU) 2018/2002 amending Directive 2012/27/EU on energy efficiency (2018) Official Journal L 328, p. 210. Gerven, M. van and Beckers, M., (2009). Unemployment protection reform in Belgium, Finland, the Netherlands, and the UK: policy learning through open coordination?. In: M. Heidenreich and J. Zeitlin, eds. Changing European employment and welfare regimes: the influence of the Open Method of Coordination on national reforms. London: Routledge. pp. 61–83. Gerven, M. van and Ossewaarde, M., (2018). Beyond the neoliberal paradigm? Images of Social Europe in Open Method of Coordination Employment peer reviews. Social Policy & Administration. 52(7), 1354–1366. Gerven, M. van, Vanhercke, B. and Gürocak, S., (2014). Policy learning, aid conditionality or domestic politics? The Europeanization of Dutch and Spanish activation policies through the European Social Fund. Journal of European Public Policy. 21(4), 509–527. Gornitzka, Å., (2018). Organising soft governance in hard times – the unlikely survival of the Open Method of Coordination in EU education policy. European Papers – A Journal on Law and Integration. 3(1), 235–255. Graziano, P. R., Jacquot, S. and Palier, B., eds., (2011). The EU and the domestic politics of welfare state reforms: Europa, Europae. New York: Palgrave Macmillan. Heidenreich, M. and Zeitlin, J., (2009). Changing European employment and welfare regimes. The influence of the Open Method of Coordination on national reforms. London: Routledge. Helm, D., (2014). The European framework for energy and climate policies. Energy Policy. 64, 29–35. Jacobsson, K., (2004). Soft regulation and the subtle transformation of states: the case of EU employment policy. Journal of European Social Policy. 14(4), 355–370. Jacobsson, K. and Johansson, H., (2009). The micro-politics of the Open Method of Coordination: NGOs and the social inclusion process in Sweden. In: M. Heidenreich and J. Zeitlin, eds. Changing European employment and welfare regimes: the influence of the Open Method of Coordination on national reforms. London: Routledge. pp. 173–191. Jordan, A., Wurzel, R. K. and Zito, A., (2005). The rise of ‘new’ policy instruments in comparative perspective: has governance eclipsed government?. Political Studies. 53(3), 477–496. Knodt, M. and Schoenefeld, J. J., (2020). Harder soft governance in European climate and energy policy: exploring a new trend in public policy. Journal of Environmental Policy and Planning. 22(6), 761–773. Knodt, M., Ringel, M. and Müller, R., (2020). ‘Harder’ soft governance in the European Energy Union. Journal of Environmental Policy and Planning. 22(6), 787–800.

302  Research handbook on soft law Lammers, I., Gerven, M. van and Treib, O., (2018). Efficiency or compensation? The global economic crisis and the development of the European Union’s social policy. Global Social Policy. 18(3), 304–322. López-Santana, M., (2006). The domestic implications of European soft law: framing and transmitting change in employment policy. Journal of European Public Policy. 13(4), 481–499. Mabbett, D., (2007). Learning by numbers? The use of indicators in the co-ordination of social inclusion policies in Europe. Journal of European Public Policy. 14(1), 78–95. Oberthür, S., (2019). Hard or soft governance? The EU’s climate and energy policy framework for 2030. Politics and Governance. 7(1), 17–27. Pochet, P., (2020). Twenty years of the publication ‘Social policy in the European Union’: what have we learned?. In: B. Vanhercke, D. Ghailani, S. Spasova and P. Pochet, eds. Social policy in the European Union 1999-2019: the long and winding road. Brussels: European Trade Union Institute (ETUI) and European Social Observatory (OSE). pp. 13–35. Radaelli, C. M., (2008). Europeanization, policy learning, and new modes of governance. Journal of Comparative Policy Analysis. 10(3), 239–254. Radulova, E., (2007). The OMC: an opaque method of consideration or deliberative governance in action?. Journal of European Integration. 29(3), 363–380. Scharpf, F. W., (2008). Negative und positive integration. In: M. Höpner and A. Schäfer, eds. Die politische ökonomie der europäischen integration. Frankfurt/Main: Campus. pp. 49–87. Scharpf, F. W., (2010). The asymmetry of European integration, or why the EU cannot be a ‘social market economy’. Socio-Economic Review. 8(2), 211–250. Schmidt, V. A., (2020). Europe’s crisis of legitimacy: governing by rules and ruling by numbers in the Eurozone. New York: Oxford University Press. Stiller, S. and Gerven, M. van, (2012). The European Employment Strategy and national core executives: impacts on activation reforms in the Netherlands and Germany. Journal of European Social Policy. 22(2), 118–132. Tholoniat, L., (2010). The career of the Open Method of Coordination: lessons from a ‘soft’ EU instrument. West European Politics. 33(1), 93–117. Tricart, J., (2020). Once upon a time there was the European social dialogue. In: B. Vanhercke, D. Ghailani, S. Spasova and P. Pochet, eds. Social policy in the European Union 1999-2019: the long and winding road. Brussels: European Trade Union Institute (ETUI) and European Social Observatory (OSE). pp. 71–98. Twena, M., (2012). Networked environmental governance in the European Union: who participates and (how) do they learn? [online]. Doctoral thesis. University of East Anglia. [Viewed 5 May 2022]. Available from: https://​ueaeprints​.uea​.ac​.uk/​41973. Vandenbroucke, F., (2002). The EU and social protection: what should the European Convention propose?. MPIfG Working Paper, 02/6 [online]. [Viewed 6 May 2022]. Available from: http://​hdl​ .handle​.net/​11858/​00​-001M​-0000​-0012​-530A​-2. Vanhercke, B., (2020). From the Lisbon Strategy to the European Pillar of Social Rights: the many lives of the Social Open Method of Coordination. In: B. Vanhercke, D. Ghailani, S. Spasova and P. Pochet, eds. Social policy in the European Union 1999-2019: the long and winding road. Brussels: European Trade Union Institute (ETUI) and European Social Observatory (OSE). pp. 99–123. Vanhercke B., and Spasova S., (2022). Conclusions. dealing with the pandemic: re-emerging social ambitions as the EU recovers. In: B. Vanhercke and S. Spasova, eds. Social policy in the European Union: state of play 2021. Re-emerging social ambitions as the EU recovers from the pandemic. Brussels: European Trade Union Institute (ETUI) and European Social Observatory (OSE). pp. 145–170. Vanhercke, B., Verschraegen, G., Gehuchten, P.-P. van and Vanderborght, Y., eds., (2011). L’Europe en Belgique, la Belgique dans l’Europe: configuration et appropriation des politiques sociales. Gent: Academia Press. Verdun, A. and Vanhercke, B., (2022). Are (some) social players entering European recovery through the Semester back door?. In: B. Vanhercke and S. Spasova, eds. Social policy in the European Union: state of play 2021. Re-emerging social ambitions as the EU recovers from the pandemic. Brussels: European Trade Union Institute (ETUI) and European Social Observatory (OSE). pp. 103–126. Verschraegen, G., Vanhercke, B. and Verpoorten, R., (2011). The European Social Fund and domestic activation policies: Europeanization mechanisms. Journal of European Social Policy. 21(1), 55–72.

The Open Method of Coordination (OMC)  303 Zeitlin, J. and Vanhercke, B., (2018). Socializing the European Semester: EU social and economic policy co-ordination in crisis and beyond. Journal of European Public Policy. 25(2), 149–174. Zeitlin, J., Barcevičius, E. and Weishaupt, J. T., (2014). Institutional design and national influence of EU social policy coordination: advancing a contradictory debate. In: E. Barcevičius, J. T. Weishaupt and J. Zeitlin, eds. Assessing the Open Method of Coordination: institutional design and national influence of EU social policy coordination. Houndmills: Palgrave Macmillan. pp. 1–15.

20. Studying the EU soft law cycle: the role of domestic factors Anne Ausfelder, Adam Eick and Miriam Hartlapp

1. INTRODUCTION1 In 2016 the European Banking Authority (EBA) adopted guidelines for the supervision and governance of retail banking products (EBA 2016). In a notice published on its website in September 2017 the French Autorité de contrôle prudentiel et de résolution (Authority for Prudential Supervision and Resolution, ACPR), which is the national supervisory authority, stated that it complied with these guidelines (ACPR 2017). It clarified that as of that moment, the guidelines would apply to credit institutions, payment institutions, and electronic money institutions under its supervision. Consequently, all national financial institutions had to make efforts to comply with the guidelines and to ensure that their product distributors also complied. Yet, the Fédération bancaire française (French Banking Federation, FBF) sought the annulment of that notice on the ground of the alleged lack of competence on the part of the EBA to issue those guidelines. In substance, it questioned the EBA’s decision to transfer the concept of ‘product governance’ that had been established for financial products on to retail banking. Retail banking products, it argued, are less risky for consumers and thus the requirements are not justified. Consequently, the FBF took to court the ACPR that had rendered the soft EU guidelines legally binding in France. In the context of the examination of the action, the Conseil d'État (Council of State, France) called on the European Court of Justice to clarify whether the plea of invalidity raised by the applicant against the EBA guidelines was admissible and well-founded. Even though the Court ruled in its judgment (C-911/19 Fédération bancaire française) that while the guidelines cannot be object of an action for annulment, their (non-)bindingness does not affect the admissibility of a request for preliminary ruling on validity. Advocate General Bobek had gone even one step further and indicated in his opinion that EU soft law, once it has been used by national authorities becomes de facto binding for national addresses. Consequently, it ought to be justiciable like EU hard law (Gentile 2021b, pp. 995–997). While the EBA guidelines are specific in eliciting that EU soft law is complied with or national resistance explained (the comply-or-explain mechanism), the more general point is that justiciability at the supranational level is becoming less and less dependent on the bindingness of the act but increasingly more on the usage of EU soft law at the national level. This highlights the fact that our understanding of EU soft law will remain limited if research focuses exclusively on the making of EU soft law at the supranational level or its (normative) characteristics as a specific type of legal instrument. This chapter argues that in order to study the phenomenon of EU soft law the whole cycle in the EU multilevel system should be taken into consideration: the creation of soft law at the EU level, the use or non-use of soft law at the national level and possible feedback at the EU level. Consequently, research benefits from a broadening of the horizons and thinking beyond 304

Studying the EU soft law cycle: the role of domestic factors  305 the supranational level, i.e., from including national level factors, e.g., the role of actors such as national authorities, addressees or courts, and the role of processes such as transposition (Gentile 2021a) or litigation at the national level (Xanthoulis 2021). Put differently, we should pay attention to the political system, territorial set-up or to state-society system to understand the usage of EU soft law across the multilevel system. EU soft law recommends, gives guidance and suggests – but never in a binding form. A minimum definition states that soft law are ‘rules of conduct which have no legally binding force but which nevertheless may have practical effects and legal effects’ (Snyder 1994). The EU treaties recognize recommendations and opinions in Article 288 TFEU. Yet, many other instruments also exist. The Open Method of Coordination (OMC) in particular received much scholarly attention (Zeitlin and Trubek 2003; Jacobsson 2004; Trubek and Trubek 2005; van Gerven and Stiller in this volume). Beyond soft law instruments recognized in the EU treaties and used within the OMC, there exists a wide range of less studied soft law instruments, such as communications, notices, declarations, and agency soft law. A case in point are the EBA guidelines mentioned above. At the beginning of the 2000s, about 11 per cent of EU legislation in force were recommendations and opinions (Bogdandy et al. 2004). Numbers have increased over time (Zhelyazkova et al. 2015) with substantial differences being reported across policy fields (Korkea-aho 2015; Cappellina et al. 2022). Growing empirical relevance is met with an increasing scholarly attention for EU soft law. Legal scholarship (Snyder 1994; Beveridge and Nott 1998; Klabbers 1998; Abbott et al. 2000) was first to discuss how to delineate soft from hard law (d’Aspremont 2008) as well as to differentiate it from no law (Terpan 2015). It suggested distinguishing between preparatory/ informative, interpretative/decisional and steering instruments as different types of EU soft law (Senden 2004, p. 140). The process of soft law-making is debated in light of participatory and democratic standards (Dawson 2011; Korkea-aho 2015) and its usage by courts in light of general principles of law (Ştefan 2012). These studies highlight normative issues that gain in importance with the increasing use of EU soft law. More recently, research interest has broadened to include effects of EU soft law at the national level. Empirical studies, although preliminary and limited in scope, indicate that the usage of EU soft law is frequent (Georgieva 2015; Saurugger and Terpan 2015; Eliantonio et al. 2021). Studies have identified unique explanations and mechanisms of soft usage that come with its design, such as framing and norm diffusion (López-Santana 2006) or its connection to hard law (Hartlapp and Hofmann 2021). These explanations address EU level policy-making and instruments. The national context has, notwithstanding the few exceptions mentioned above, received little attention so far. Astonishingly so, as research on EU hard law implementation has emphasized the national context as a highly important explanation for the reception of EU norms (Falkner et al. 2005). This chapter seeks to contribute to closing this gap with a framework to analyze EU soft law across the EU multilevel system. This lens is theoretically informed by public policy research and therewith suggest avenues for future research at the intersection of legal and political science research. The next section discusses the different steps in the cycle and suggest conceptualizations as well as research designs and data sources to be employed (Section 2). We then zoom into a particular section of the cycle – national usage – to suggest how one factor, the role of the state, can be studied empirically, to explain the national usage of soft law. To this aim, two case studies explore the role of territorial structure of the state and of state society relations for the usage of EU soft law in the administrations of France and Germany (Section 3). We

306  Research handbook on soft law selected two EU soft law instruments from different policy fields: the Commission regional state aid guidelines and the Commission recommendations for the monitoring of acrylamide. We illustrate effects of these instruments in France and Germany as countries that have been described as rather reluctant towards EU soft law. Their Romano-catholic legal systems are based on formal law that is written down and codified. This leaves little room for a culture of proactive soft law usage (Hartlapp et al. 2021, p. 154; Rubio and Ştefan 2021, p. 138). Constituting unlikely cases of far-reaching usage of EU soft law, France and Germany shed light on the conditions under which EU soft law may nevertheless take effect at the national level. What is more, comparison across countries with different political systems – one federal and used to multilevel governance (Germany) and one more characterized by statism and centralization (France) – allows us to capture how differences in the role of the state matter for understanding the use of soft law. We conclude (Section 4) with a summary and suggest future perspectives to study the effect of EU soft law at the national level.

2.

RESEARCHING THE USE OF SOFT LAW ACROSS THE EU MULTILEVEL SYSTEM

EU soft policy-making takes place in a dynamic multilevel process. We suggest to study this process with a framework in form of a cycle that starts with the creation of soft law at the EU level, goes on with the use or non-use of soft law at national level and culminates in feedback effects at EU level.2 Compared to more formalized legislative procedures (i.e., the process of hard law-making), the soft law policy cycle is blurred. Agenda-setting and policy formulation phases are likely to be similar for both hard and soft law. At the early stages of many policy-making processes, actors may have no clear idea of the specific form that the instrument ultimately adopted will have. However, once adopted, we can clearly differentiate between policy-making processes of hard and soft norms. Given non-bindingness and absent implementation deadlines, we expect the effects of EU soft law at the national level to span larger time stretches compared to hard law measures and to take place before or in parallel to adoption of the EU instrument, e.g., when actors adapt their behaviour in anticipation of a soft law measure. Finally, once implemented, soft law may also feed back into the supranational policy-making and become EU hard law. Employing actor-centred institutionalism (Scharpf 1997), we assume that the usage of EU soft law depends on the interest and preferences of actors and is shaped by the institutional setting in which they operate. Thus, cognitive frames in which agents act, the salience they attach to a policy issue, or power relations among them are important analytically. Consequently, we argue that we can only explain the usage of EU soft law if we know how the characteristics of a EU soft law instrument are mediated by additional factors at each stage of the soft law cycle. 2.1

The Starting Point: EU Soft Law

Our framework assumes that policy fields differ in degrees of ‘softness’. EU policy fields have been developing based on the competences attributed to the EU by EU treaties (Börzel 2005). They usually combine hard and soft law, but the proportion of each kind of norm varies. Moreover, a policy area can evolve over time and be legalized or delegalized. Legalization

Studying the EU soft law cycle: the role of domestic factors  307

EU

EU

SOFTLAW

HARDLAW

Creation of soft law at the EU level

Feedback effects at the EU Level

NATIONAL

USAGE

Use or non-use of soft law at national level

Figure 20.1

Soft law cycle in the EU multilevel system

(soft law hardening) occurs when soft norms become binding and/or subject to stronger enforcement requirements. A prominent example is the ‘communautarization’ of Justice and Home Affairs since the Amsterdam and Lisbon Treaties, as well as the European Charter of Fundamental Rights, which was transformed into a binding instrument by the Treaty of Lisbon (Saurugger and Terpan 2021, p. 8). Delegalization (hard law softening) describes a situation where enforcement is no longer sought or hard norms are replaced by soft ones, e.g., the Council Directive 96/61/EC (1996) on Integrative Pollution Prevention and Control introduced non-binding requirements where previously there were hard obligations (Saurugger and Terpan 2021, p. 8). A systematic understanding of the amount and characteristics of soft law at the supranational level would benefit from large-N analysis that allows to capture the dynamics of soft law making over time and across policy fields. The EU official database (EurLex) forms a solid starting point for such as systematic assessment. Yet, as much of EU soft law is not captured in EurLex, such as guidelines or agency soft law, documentary searches for non-binding rules issued by the responsible Directorates General of the EU Commission, other EU institutions and decentralized agencies can provide a more complete picture. This enables precise evaluation of the respective proportion of soft law and hard law in a particular policy field, differences across policy fields over time, issuing authorities, or types of soft law instruments (Cappellina et al. 2022). Initial empirical studies measuring the share of soft law relative to hard law come to different conclusions due to different timeframes, definitions of soft law, and data collection methods. But all find that soft law makes up a considerable amount of 11 per

308  Research handbook on soft law cent (Bogdandy et al. 2004, p. 97), 40 per cent (Zhelyaskova et al. 2015) and 49 per cent or 43 per cent depending whether a wider or restrictive definition is applied (Cappellina et al. 2022, pp. 750–752) of the EU’s legal framework and that the number of soft law instruments has increased over time. But the degree of legalization, the share of soft law relative to hard law, strongly varies across policy fields (Cappellina et al. 2022, pp. 750-752). This kind of research suffers from assuming that EU instruments are homogenous units of analysis. Yet, individual acts can also be of a harder or softer nature. Other studies have assessed the hardness of individual acts (Knodt and Schoenefeld 2020; Bekker 2021). It has been suggested that such characteristics of policy fields and instruments at the EU level can affect the usage of soft law. Concerning policy fields, Hartlapp and Hofmann (2021) suggest that national usage is stronger in policy areas in which competences have been at the EU level for a longer period of time and where EU rules are denser. Saurugger and Terpan (2015, p. 60) expect that the national reception of soft law depends on the possibility of sanctions, especially of social sanctions. 2.2

Domestic Usage: When and Why do Soft Norms take Effect at the Domestic Level?

At the domestic level, soft law can be used by national governments and administrations, by private actors or the civil society more generally. Our dependent variable ‘usage’ captures the degree to which these actors refer to and implement soft law in their actions. We explicitly refer to the usage of soft law rather than to its implementation. As a term, implementation is suitable to explain the efficiencies and deficiencies of the multilevel system. Direct (formal) requirements and implementation deadlines formulated at the EU level specify what is expected of national political system and actors and in what timeframe. We can assess the timeliness and correctness of implementation and on this basis judge compliance with EU rules (Falkner et al. 2005). EU soft law, in turn, offers resources, references and policy frames that can, but need not, be used at the national level (Woll and Jacquot 2010, p. 114). Usage on the domestic level is triggered when the EU is perceived as a set of opportunities, leading subsequently to adjusted and/or redefined practices and political interactions. Actors create a context of mutual influence by moving between the different levels of the EU multilevel governance framework (Jacquot and Woll 2003, p. 9). The term usage should thus be suited to capture a wider empirical range of outcomes and processes at the national level that are causally linked to EU soft law. Accordingly, at this stage of the cycle our analytical framework focuses particularly on the factors that render usage at the national level more likely. In some cases, our dependent variable ‘usage’ can be further specified in different types of soft law usage. There are several choices possible for a Member State to use EU soft law. We analytically distinguish hard and soft usage: 1. A hard usage of EU soft law would occur where a Member State formally adapts to EU soft law (e.g., EU guidelines are included in a national piece of legislation); or where the soft norm is referred to by national judges in court rulings to decide a case. 2. A soft usage, in contrast, becomes visible where EU soft law is used by national administrations outside the framework of hard law (e.g., cited in national guidelines, manuals, guidebooks, databases or Q&As that inform decisions). This can take the form of repro-

Studying the EU soft law cycle: the role of domestic factors  309 ducing EU soft law in the form of more traditional national soft law or takes place in the form where other societal actors refer to EU soft law in their discourse, position-taking or daily actions (Hartlapp and Korkea-aho 2021, pp. 67–72).3 While analytically distinct, we expect that empirically mixed forms prevail. Such hybrid incorporations of hard and soft law can be used at the same time, e.g., when a new law is complemented by a guideline setting out how to apply the legislative standards in practice, or they can stretch out over time with soft and hard instruments being issued in a sequence. Whether soft law is incorporated into legislation, national soft law, some other form of non-binding guidance, or their combinations depends on national actors and institutions. National and, in particular, sectoral patterns of interest configurations and policy-making will affect the extent to which soft law is used at the national level and can help explain differences in the usage of EU soft law. In terms of actors, EU hard law is typically transposed via the legislative arena at the national level. Soft law, in turn, is used by a range of governance actors (see country chapters in Eliantonio et al. 2021). We differentiate between government, public officials in ministries and agencies, judges as well as organized interests. Government and other public officials have much leeway in deciding about the usage of EU soft law as part of their agenda-setting and decision-making powers. Other actors are more constrained. Judges will usually need a case that touches upon EU policies to refer to EU soft law as a means to give reason or to interpret related EU rules (Georgieva 2015; Eliantonio 2021). Organized interests need access and influence. Their usage of soft law ultimately depends on the concrete positions that these interests have vis-à-vis EU soft law – at least one actor has to have a substantial interest in using EU soft law in order for usage to occur at the national level. Usage is more likely when soft law reflects the preferences of major domestic stakeholders who form alliances and join forces with elected officials to oppose or foster compliance with European standards (Quaglia 2019). Actors use soft law if it benefits them. There are different benefits attached to soft law usage: to put new issues on the table or to strengthen one’s own position in an ongoing policy process. Strengthening can concern procedural or substantial positions in a policy process. And it can occur directly or indirectly for example by weakening opponents. How then to study the usage of EU soft law at the national level? Researching domestic usage of EU soft law can be challenging for both the complexity of the analysed processes and for the lack of publicly available data on the implementation of EU soft law and feedback effects in EU policy-making. We argue that case study analysis, based on primary and secondary sources, seems to be a suitable research instrument to obtain an in-depth understanding of processes and relevant causal factors. Bibliographical research and public documents from existing and ongoing legislative and regulatory processes, and the mapping of stakeholders, administrations, committees and other relevant actors are a good starting point. These sources, however, usually provide little insights into whether soft law is recognized and used by domestic actors in the political or administrative sphere, to bring out motivations of actors, potential conflicts and questions of power. Here, expert interviews are an important source of information (Aberbach and Rockman 2002; Littig 2008). A drawback of case study research is that it is difficult to find out whether the potential causal factors identified are specific to the case at hand, and whether the findings can be generalized to other cases. One way to assess whether or not we can generalize across cases or even levels of governance and policy fields in the EU multilevel system is to combine the

310  Research handbook on soft law case studies with a survey. A survey carried out among public officials and/or actors in the legal system offers broad and systematic insights into whether and how EU soft law is used at the domestic level. Reaching out to socio-legal methods of research similar approaches with expert interview-based case studies and surveys can be employed to understand the role of lawyers and the legal system for the usage of EU soft law (Hartlapp and Korkea-aho 2021). In addition, existing research has demonstrated the usefulness of official databases that offer concrete examples when and where EU soft law is used in national case law. It is also possible to use such databases to see in which policy fields judges refer (more frequently) to soft law usage when justifying a case or which EU soft law acts are particularly influential in the legal sphere of the EU multilevel system (Georgieva 2015; Eliantonio et al. 2021). 2.3

Feedback in EU Policy-making: When and Why does EU Soft Law Feed Back into EU Policy-making?

In the EU’s complex multilevel system, implementation processes at the national level feed back to the supranational level. Ideas and arguments travel through governance levels, and actors will seek to defend their interests in policy-making processes across the multilevel system (for a similar approach on the feedback of the case law of the European Court of Justice, see Martinsen 2015). In the case of EU soft law, a feedback effect can manifest in the ‘hardening’ of an act that was initially soft, e.g., where EU non-binding norms are transformed into (national or EU) rules that are obligatory and precise enough to be controlled. Or the feedback effect can occur when actors engage in oversight and control activities, including e.g., the sanction capacity of third parties. For example, the European Commission used the Code of Conduct for business Taxation (ECOFIN Council 1997), a soft law adopted by the Council in the area of business taxation, as a point of reference for its assessment of state aid, a case of hard enforcement (Traversa and Flamini 2015). The transformation of soft law into hard law can be brought about by different factors that capture the interplay between levels as well between policy-making actors when negotiating a revision of initially non-binding instruments. Here, the distribution of powers between EU actors (Holzinger and Biesenbender 2019) as well as positions and dynamics of negotiation and consensus building in the Council and the European Parliament are central explanations (Wratil and Hobolt 2019). Following the cycle perspective, EU actors are likely to be influenced by the usage of EU soft law at the national level. Where many Member States have implemented EU soft law instruments, this signals the general acceptance of relevant norms. In this case, ambitious agenda-setting by the European Commission should be more likely in the future. A good example of this is the recent Commission (2021) proposal for a Directive on Transparency in Equal Pay. It takes up many provisions that had been part of the non-binding Commission Recommendation 2014/124/EU (2014) with the same title. The recommendation has been widely used by national actors at the domestic level, preparing the ground for a more ambitious and in this sense, a harder approach to equal pay at the supranational level (Hartlapp et al. 2020). Researching the feedback of EU soft law in EU policy-making is faced with similar challenges as the study of domestic usage. Case study design allows to establish the causal relevance of theorized factors. Surveys or the systematic tracing of EU soft law in official databases allows for generalizations. In sum, combining different designs and data sources seems a promising avenue to research the EU soft law cycle. Having introduced the cycle as

Studying the EU soft law cycle: the role of domestic factors  311 framework to study soft law in the EU multilevel system, we now zoom into one stage of the cycle to explore usage at the national level in more detail.

3.

EXPLORING HOW FEATURES OF THE STATE SHAPE USAGE OF EU SOFT LAW

As an institution, the state is distinct but not separate from government and public institutions as its constituting parts (Hayward 2017, p. 44). We assume differences in the role of the state for France and Germany. Overall, the role of the French state is characterized by state elitism, dirigisme and bureaucratic statism which implies a constitutional philosophy of the state acting in the public interest. In Germany, state authority is relatively restrained. This has to do with the handling of power after the Second World War and the primacy of legality (‘Rechtstaat’) with checks and balances that come with federalism (Knill 2001, p. 62). In a nutshell, the state is likely to matter for soft law usage, but we can expect differences across France and Germany with respect to specific features of their state such as a territorial structure or interactions with interest groups. 3.1

State Structure: Federalism and State Aid

According to Article 107(3)(a) and (c) TFEU, Member States can grant state aid for the purposes of regional development. Such measures – if they exceed a certain monetary threshold – must be notified to the Commission, which reviews, authorizes, and monitors them. To specify its criteria of enforcement, the Commission (2013) has published Regional State Aid Guidelines for 2014–2020. Both policy formulation and enforcement are in DG Competition’s hands with Member States and stakeholders being consulted. The frequency of usage and intensity of effects of the Regional State Aid Guidelines are similar in Germany and France. All interviewees report that the Commission’s enforcement authority and wide discretion endow the guidelines with strong indirect legal and practical effects. Such effects have been confirmed in case law of the Court of Justice of the European Union (see e.g. Case T-27/2 Kronofrance SA v Commission). The guidelines provide that Member States shall draw up and notify to the Commission so-called regional aid maps in which they designate, according to a set of fixed indicators, geographical areas where and in what amount regional aid can be disbursed. Once the Commission has accepted these areas, again, the guidelines serve as the basis for the notification of all planned measures of regional state aid, which can range from individual subsidies to firms to broad regional funding framework laws. Member States must demonstrate the lawfulness of aid through rigorous economic assessments. Both German and French officials go to great lengths do so. Non-compliance is perceived to be too expensive, as the Commission systematically monitors and enforces regional aid based on the guidelines. Thus, the guidelines affect highly salient elements of national spatial planning policy and create a high administrative workload for national administrations (BMWi 2020; ANCT 2021). But the way in which administrations use the guidelines depends on the state structure. In German federalism, federal authorities and the regions (Länder) work together as equal partners in all stages of the policy cycle: from involvement in the formulation of the guidelines at the EU level to the implementation of regional development schemes on the ground (BMWi

312  Research handbook on soft law 2020; MWAE Brandenburg 2020). Constitutionally, regions may shape their economic policy according to Article 30 Grundgesetz (GG), but the federal government can shape spatial planning according to Article 75 GG. When drawing up the ‘regional aid maps’, Member States can categorize the regions in a relatively unfettered manner (‘non-predefined’ C areas). In this context, distributional conflicts can arise between the regions (StMWi Bayern 2020). The regions bargain within two political and administrative working groups in the institutional structure of the Gemeinschaftsaufgabe Verbesserung der regionalen Wirtschaftsstruktur (GRW), Germany’s main regional funding program. The federal government only takes a coordinating role. The negotiations are conflictual, long and complicated. Ultimately however, there is a strong focus on reaching compromise and consensus. According to the interviewees, this is partly to achieve a strong negotiating position vis-à-vis the Commission. But beyond that, institutional arrangements are geared towards consensus (BMWi 2020; MWAE Brandenburg 2020; Lijphart 2012). In sum, soft law usage is in line with established patterns of cooperative federalism, where institutions facilitate compromise, but actors still prioritize individual regional interests when distributional conflicts are at stake (Scharpf 2009, p. 29). In contrast, in centralized France, the central administration retains the upper hand at all stages of the policy cycle. Interactions are strictly hierarchical. The Commissariat général à l’égalité des territoires (CGET, since 2017 a part of the ANCT), an authority subordinate to the Ministry of Cohesion, leads the consultation, the regional map drafting and the management of the main funding program. During the formulation of the guidelines at the EU level, the authorities of the state draft a contribution to the public consultation organized by the Commission with a national vision of spatial regional planning in mind. Regions are not formally involved, although some organize informal dialogues with the state to make their views heard. The CGET organizes the regional map drafting and takes political decisions about distribution (ANCT 2021). The operational work of categorizing regions according to indicators of the EU guidelines are delegated to prefectural offices, which represent the central state in the regions. The regional maps are then assembled in Paris in décret (nº 2014-758 du 2 juillet 2014), i.e., in hard law. France has a long history of centrally designed spatial regional planning policy. In the past, territorial cohesion and national unity was the main goal. Today, attempts are being made to build multiple economically strong urban centres away from Paris (Augias 2017). A central investment aid scheme is the prime d’aménagement du territoire (PAT), which is notified to the Commission as a regional state aid scheme under the Guidelines. Although it is no longer a purely hierarchical instrument as it was 20 years ago, the basic planning is still carried out in Paris by an inter-ministerial federal commission (see Arrêté du 3 mars 2010). The distribution of individual regional aid measures, on the other hand, is strongly decentralized to départments, communes and cities. But these typically small aids are mostly below the threshold above which the guidelines become relevant (Régions de France 2020). In sum, the federal set-up in Germany supports wider usage of EU soft law, as there are both a wider range of actors that can demand usage while at the same time the centre has an interest in incentivizing uniform practices. France, in contrast, follows the general stance of a centralized regional policy that reduces demand for the usage of EU soft law.

Studying the EU soft law cycle: the role of domestic factors  313 3.2

State-society Relations: Investigations into the Levels of Acrylamide in Food

As a result of public concerns about possible carcinogenic effects of acrylamide in the early 2000s, the Commission published several recommendations on the monitoring of acrylamide levels as well as explicit indicative values (see Commission 2007; Commission 2010; Commission 2011). From 2011 onwards, Member States were asked to carry out investigations in cases where acrylamide levels in food exceeded the prescribed values. Even though the Commission Regulation (EU) 2017/2158 (2017) later set legally binding minimization measures and indicative values as a result of EU-wide heterogenous application of the recommended benchmarks, soft law still provides detailed information on procedural mitigation measures, affiliation of indicative values and reporting obligations. The recommendations produce effects in both countries. Yet, the different state-society relations influence and/or trigger different forms of usage of soft law. As a neo-corporatist system (Schmitter 1974), Germany exhibits features of a stronger participatory system than France. French state-society relations are widely believed to be distant with frequent confrontation in the streets and (in most sectors) pluralist interest organizations with little access to the state (McCauley 2017, p. 9). We observe extensive use of EU soft law on acrylamide in the French and German administrations, but especially the German industry relies on the standards that are laid down in the recommendations for their food production. German as well as French ministry and food authority officials perceive the benchmarks laid down in the recommendations as quasi-binding. Within the field of food safety, soft law is often used to ‘test’ benchmarks in the Member States which are later taken up in hard law (BMEL 2020; BVL 2020; French Administration 2021a). In the meantime, if Member States have the analytical method to detect contaminants and the resources to implement the mitigation measures, these indicative values are considered common practice (BMEL 2020; BVL 2020). In France, the Direction Générale de la Concurrence, de la Consommation et de la Répression des Fraudes (DGCCRF) carries out regular controls as part of its plan to monitor and control the contamination of food by newly formed compounds (including acrylamide), just as recommended by the Commission. The controls are centrally steered. DGCCRF has set up a working group within the Conseil national de la consommation (CNC), whose aim is to define precise recommendations for the preparation of food to reduce the contamination of food in private households and to make consumers aware of the danger of acrylamide. The EU recommendations were transposed into national guidelines by the DGCCRF (Dillmann 2021). Despite the recommendations’ non-binding legal nature, they influence national administrative practices and enforcement vary widely. In Germany, the recommendations and indicative values have been transposed into national guidelines, e.g., by the Bundesamt für Verbraucherschutz und Lebensmittelsicherheit (BVL), and resulted in wide effects for administration and industry. Even before the EU recommendations were published, Germany advocated for EU-wide uniform benchmarks for acrylamide. The BVL collected ‘signal values’ that should indicate when an acrylamide content can be hazardous to health from 2002 onwards, just after the organization was founded, because of several food crises in the late 1990s and specifically around the acrylamide issue in 2002. The food industry itself has demanded uniform regulation and even a hard usage: ‘The large retail companies […] usually also set values in their private law agreements with their suppliers and these values are then oriented towards such guideline values’, comments a German official (BMEL 2020). Sometimes the industry even voluntarily chooses harder benchmarks

314  Research handbook on soft law (BMEL 2020). Germany could thus use the recommendations to legitimize industry controls and the already ongoing mitigation measures, as well as to advocate for EU-wide harmonized market conditions. It may seem surprising that the impetus for uniform rules for the monitoring of acrylamide came from the manufacturers. Yet, in a corporatist system, state-society relations offer access and influence for organized interests. German food business representatives exerted influence at different regional levels as well as the federal level to push for an overall policy change. EU soft law was very much welcomed to further this interest. The strong tradition of organized interests participating in policy-making made the policy change even possible in the first place. In these established state-society relations EU soft law is more likely to be used and might more often take the form of hard law. The food business operators together with German officials strategically used the soft law instruments to push for hard benchmarks in all the EU. France, however, reacted rather passively. French officials report that interest groups within the field of food safety seek contact with political and legal decision-makers in more political issues, e.g. Bisphenol A than in issues related to reducing acrylamide levels (French Administration 2021b). In sum, the type of state-society relations is useful to explain how EU soft law is used. In Germany, the corporatist system has led to harder usage and a stronger cooperation with the industry, while the weaker state-society relations in France explain why usage has remained in the hands of the administration mostly. Regarding the scope of our findings, we suggest that even though some types of soft law (e.g., technical soft law in food safety to guide administrative actors in interpreting hard law) are quite common for specific policy fields, that does not mean that they are all used in the same way. We expect variance in usage to prevail.

4.

SUMMARY AND CONCLUSION

This chapter started from the observation that EU soft law is increasingly used in EU multilevel governance. Even though EU soft law ‘merely’ recommends, guides and suggests, it can have practical and legal effects in the Member States and feed back into EU policy-making. While the literature has so far focused on the characteristics of different soft law instruments to explain the usage of EU soft law in the EU multilevel system, we contribute to the emerging research agenda on EU soft law by introducing an EU soft law cycle that ranges from the creation of soft law at the EU level, the use or non-use of soft law at the national level to feedback effects at the supranational level. This framework allows us to direct our attention to specific actors and potential causal factors at the different stages. We therefore advocate for acknowledging relevant causal factors and employing research designs that are suitable to gain insights into this cycle. In addition to the existing research, this perspective allows integrating domestic factors that drive usage with implications for the whole cycle. These potential explanatory factors vary across countries and frequently also across policy fields, highlighting the complexity of studying soft law across the EU multilevel system. We illustrate the importance of such domestic factors by focusing on the features of the state as one potentially causal factor. In two case studies, we explore usage by focusing on the territorial set-up as well as state-society relations. We described these state features on the basis of existing public policy literature and show the effect EU soft law has in concrete implementation processes. A comparison between France and Germany enabled us to capture how

Studying the EU soft law cycle: the role of domestic factors  315 differences in state characteristics matter and thus provide for mechanistic insights (Hedström and Ylikoski 2010) on how the features of the state causally impact the usage of EU soft law: the federal set-up of the state as well as a higher number of actors that have access to public policy making and stronger state-society relations are conducive to (hard) usage of EU soft law in Germany, while strong the role of the state in France is decisive to understand why usage remains in the hands of the administration. Our cases also showed that the usage of soft law is very much driven by problem pressure and opportunity costs. Future research should thus pay particular attention to interest constellations in EU soft law usage at the national level. While EU soft law frequently contributes to solving concrete empirical problems, it might strengthen some actors more than others. Systematically considering domestic factors and the mechanisms through which they affect the usage of EU soft law could help us to understand who wins and who loses from EU soft law. By putting features of the state at the centre stage we do not claim that other factors do not matter. In particular, we expect the significance of the features discussed in this chapter to vary within different policy fields, e.g., the size and importance of the administration or the structure of state-society relations differ between Member States. Yet, we think that the importance of the state as a shaper of ideas, interests and an institutional setting and its impact on usage of soft law in the whole soft law cycle justifies detailed attention in future research.

NOTES 1. The chapter is part of a wider research project on Effects of Soft Law across the EU multilevel system (EfSoLaw) funded by ANR and DFG (project N° 411037244). We thank our project partners Bartolomeo Cappellina, Romain Mespoulet, Sabine Saurugger and Fabian Terpan for fruitful debates and the editors for valuable feedback. 2. This conceptualization stems from our broader project “Effects of EU Soft Law across the multilevel system” (EfSoLaw). See the EfSoLaw website for more information: [Visited 21 June 2023]. https://​wwhttps//​www​.efsolaw​.eu/​index​.html. 3. Terpan (2015) further differentiates different types of soft usage that can be considered to cover different degrees of softness on a continuum that ranges from binding instruments without enforcement, to non-binding norms within binding instruments and finally to non-binding instruments.

REFERENCES Abbott, K. W. et al., (2000). The Concept of Legalization. International Organization. 54(3), 401–420. Aberbach, J. D. and Rockman, Bert A., (2002). Conducting and Coding Elite Interviews. Political Science & Politics. 35(4), 673–676. ACPR., (2017). Avis ACPR Banque de France Mise en œuvre des orientations de l’Autorité bancaire européenne sur les politiques et les pratiques de rémunération liées à la vente et à la fourniture de produits et de services de banque de détail (EBA/GL/2016/06) [online]. 6 September 2017. [Viewed 25 March 2022]. Available from: https://​acpr​.banque​-france​.fr/​sites/​default/​files/​media/​2017/​09/​19/​ avis​-eba​-gl​-2016​-06​.pdf. Arrêté du 3 mars 2010. Journal officiel de la République française. N°0075, 30 mars 2010. Augias, D., (2017). De l’aménagement à l’égalité des territoires: quelle stratégie pour l’Etat ?, in N.Kada (ed.), Les territoires de l’Etat. Droit et gestion des collectivités territoriales 2017, GRALE-CNRS, Le Moniteur, 2017.

316  Research handbook on soft law Bekker, S., (2021). Hardening and Softening of Country-specific Recommendations in the European Semester. West European Politics. 44(1), 114–133. Beveridge, F. and Nott, S., (1998). A Hard Look at Soft Law. In: P. Craig and C. Harlow, eds Lawmaking in the European Union. London: Kluwer Law International. pp. 285–309. Bogdandy, A. von, Arndt, F. and Bast, J., (2004). Legal Instruments in European Union Law and Their Reform: A Systematic Approach on an Empirical Basis. Yearbook of European Law. 23(1), 91–136. Börzel, Tanja A., (2005). Mind the Gap! European Integration Between Level and Scope. Journal of European Public Policy. 12(2), 217–236. Cappellina, B. et al., (2022). ‘Ever More Soft Law? A Dataset to Compare Binding and Non-binding EU Law Across Policy Areas and Over Time (2004–2019)’, European Union Politics 23(4): 741–57. Case C-911/19 Fédération bancaire française (FBF) v Autorité de contrôle prudentiel et de résolution (ACPR) ECLI:EU:C:2021:599. Case T-27/2 Kronofrance SA v Commission of the European Communities ECLI:EU:T:2004:348. Opinion of Advocate General Bobek in Case C-911/19 Fédération bancaire française (FBF) v Autorité de contrôle prudentiel et de résolution (ACPR) ECLI:EU:C:2021:294. Commission., (2007). Commission recommendation of 3 May 2007 on the monitoring of acrylamide levels in food. Oficial Journal. L123, 12 May 2007, 33–40. Commission., (2010). Commission recommendation of 2 June 2010 on the monitoring of acrylamide levels in food. Official Journal. L137, 3 June 2010, 4–10. Commission., (2011). Commission Recommendation of 10 January 2011 on investigations into the levels of acrylamide in food [online]. C(2010) 9681 final. [Viewed 1 April 2022]. Available at: https://​ec​.europa​.eu/​transparency/​documents​-register/​api/​files/​C(2010)9681​_0/​de00000000646879​ ?rendition​=​false Commission., (2013). Guidelines on regional State aid for 2014–2020. Official Journal. C209, 23 July 2013, 1–45. Commission., (2021). Proposal for a Directive of the European Parliament and of the Council to strengthen the application of the principle of equal pay for equal work or work of equal value between men and women through pay transparency and enforcement mechanisms. COM/2021/93 final. Commission Recommendation 2014/124/EU on strengthening the principle of equal pay between men and women through transparency. Official Journal, L69, 7 March 2014, 112–116. Commission Regulation (EU) 2017/2158 establishing mitigation measures and benchmark levels for the reduction of the presence of acrylamide in food. Official Journal. L304, 21 November 2017, 24–44. Council Directive 96/61/EC concerning integrated pollution prevention and control (1996). Official Journal. L257, 10 October 1996, 26–40. d’Aspremont, J., (2008). Softness in International Law. A Self-Serving Quest for New Legal Materials. European Journal of International Law. 19(5), 1075–1093. Dawson, M., (2011). New Governance and the Transformation of European Law: Coordinating EU Social Law and Policy: Cambridge Studies in European Law and Policy. Cambridge: Cambridge University Press. Décret n° 2014-758 du 2 juillet 2014 relatif aux zones d’aide à finalité régionale et aux zones d’aide à l’investissement des petites et moyennes entreprises pour la période 2014–2021. Journal officiel de la République française. N°0152, 3 juillet 2014. Dillmann, P., (2021). L’acrylamide dans les aliments : une préoccupation de santé publique [online, viewed 5 January 2022]. Available from: https://​www​.economie​.gouv​.fr/​dgccrf/​lacrylamide​-dans​ -aliments​-preoccupation​-sante​-publique. EBA., (2016). Guidelines on remuneration policies and practices related to the sale and provision of retail banking products and services [online]. EBA/GL/2016/06, 13 December 2016. [Viewed 25 March 2022]. Available from: https://​www​.eba​.europa​.eu/​sites/​default/​documents/​files/​documents/​ 10180/​1693444/​000c55c6​-cbe4​-4904​-b43d​-2ca9294b6cd6/​Guidelines​%20on​%20sales​%20staff​ %20remuneration​%20​%28EBA​-GL​-2016​-06​%29​_EN​.pdf​?retry​=​1. ECOFIN Council., (1997). Resolution of the Council and the Representatives of the Governments of the Member States, meeting within the Council of 1 December 1997 on a code of conduct for business taxation - Taxation of saving. Official Journal. C2, 6 January 1998, 1–6. Eliantonio, M., Korkea-aho, E. and Ştefan, O., eds., (2021). EU Soft Law in Member States: Theoretical Findings and Empirical Evidence. Bloomsbury: Hart Publishing.

Studying the EU soft law cycle: the role of domestic factors  317 Falkner, G. et al., (2005). Complying with Europe. EU Minimum Harmonisation and Soft Law in the Member States. Cambridge: Cambridge University Press. Gentile, G., (2021a). ‘Verba Volant, Quoque (Soft Law) Scripta?’: An Analysis of the Legal Effects of National Soft Law Implementing EU Soft Law in France and the UK. In: M. Eliantonio, E. Korkea-aho and O. Ştefan, eds. EU Soft Law in Member States: Theoretical Findings and Empirical Evidence. Bloomsbury: Hart Publishing. pp. 79–98. Gentile, G., (2021b). To be or not to be (legally binding)? Judicial review of EU soft law after BT and Fédération Bancaire Française. Revista de Derecho Comunitario Europeo. 70, 981–1005. Georgieva, Z., (2015). Soft Law in EU Competition Law and its Judicial Reception in Member States: A Theoretical Perspective. German Law Journal. 16(2), 223–260. Hartlapp, M. et al., (2020). Studying EU Soft Law Effects in Social Policy. SoLaR Working Paper. Maastricht. Hartlapp, M. and Hofmann, A., (2021). The Use of EU Soft Law by National Courts and Bureaucrats: How Relation to Hard Law and Policy Maturity Matter. West European Politics. 44(1), 134–154. Hartlapp, M., Hofmann, A. and Knauff, M., (2021). ‘Soft Law in Germany. Still Opposing Dynamics in Status and Effect’, in M. Eliantonio, E. Korkea-aho and O. Ştefan, eds. EU Soft Law in Member States: Theoretical Findings and Empirical Evidence. Bloomsbury: Hart Publishing, pp. 153–67. Hartlapp, M. and Korkea-aho, E., (2021). ‘Whatever-Law’ and Teenage Member States? In: M. Eliantonio, E. Korkea-aho and O. Ştefan, eds. EU Soft Law in Member States: Theoretical Findings and Empirical Evidence. Bloomsbury: Hart Publishing. pp. 59–78. Hayward, J., (2017). The State Imperative. In: R. Elgie et al., eds. The Handbook of French Politics. Oxford: Oxford University Press. pp. 44–59. Hedström, P. and Ylikoski, P., (2010). Causal Mechanisms in the Social Sciences. Annual Review of Sociology, 36(1), 49–67. Holzinger, K. and Biesenbender, J., (2019). The Evolution of Legislative Power-Sharing in the EU Multilevel System. In: N. Behnke, J. Broschek and J. Sonnicksen, eds. Configurations, Dynamics and Mechanism of Multilevel Governance. Festschrift in honour of Arthur Benz' 65th birthday. London: Palgrave. pp. 331–349. Jacobsson, K., (2004). Soft Regulation and the Subtle Transformation of States: The Case of EU Employment Policy. Journal of European Social Policy. 14(4), 355–370. Jacquot, S. and Woll, C., (2003). Usage of European Integration – Europeanisation from a Sociological Perspective, European Integration online Papers (EIoP) 12(7) [Viewed 8 July 2023]. http://​eiop​.or​ .at/​eiop/​texte/​2003​-012a​.htm. Klabbers, J., (1998). The Undesirability of Soft Law. Nordic Journal of International Law. 67(4), 381–391. Knill, C., (2001). The Europeanisation of National Administrations. Patterns of Institutional Change and Persistence. Cambridge: Cambridge University Press. Knodt, M. and Schoenefeld, J. J., (2020). Harder Soft Governance in European Climate and Energy Policy: Exploring a New Trend in Public Policy. Journal of Environmental Policy & Planning. 22(6), 761–773. Korkea-aho, E., (2015). Adjudicating New Governance: Deliberative Democracy in the European Union. Oxon: Routledge. Lijphart, A., (2012). Patterns of Democracy. Government Forms and Performance in Thirty-Six Countries, New Haven/ London: Yale University Press. Littig, B., (2008). Interviews mit Eliten - Interviews mit ExpertInnen: Gibt es Unterschiede? Forum: Qualitative Research (FQS). 9(3). López-Santana, M., (2006). The Domestic Implications of European Soft Law: Framing and Transmitting Change in Employment Policy. Journal of European Public Policy. 13(4), 481–499. Martinsen, D. S., (2015). Judicial Influence on Policy Outputs? The Political Constraints of Legal Integration in the European Union. Comparative Political Studies. 48(12), 1622–1660. McCauley, D., (2017). Interest Groups: Moving Beyond State-centric Models. In: R. Elgie et al., eds. The Handbook of French Politics. Oxford: Oxford University Press. pp. 417–438. Quaglia, L., (2019). The Politics of State Compliance with International ‘soft law’ in Finance. Governance. 32(32), 45–62.

318  Research handbook on soft law Rubio, N. and Ştefan, O., (2021). ‘No Longer that Small to Fall Through the Cracks: A French Story of Adaptation to the Petites Sources of EU Law’. In: M. Eliantonio, E. Korkea-aho and O. Ştefan, eds., EU Soft Law in Member States: Theoretical Findings and Empirical Evidence. Bloomsbury: Hart Publishing, pp. 137–51. Saurugger, S. and Terpan, F., (2015). Studying Resistance to EU Norms in Foreign and Security Policy. European Foreign Affairs Review. 2(1), 1–20. Saurugger, S. and Terpan, F., (2021). Normative Transformations in the European Union: on hardening and softening law. West European Politics. 44(1), 1–20. Scharpf, F. W., (1997). Games Real Actors Play: Actor-Centered Institutionalism in Policy Research. Boulder: Routledge. Scharpf, F. W., (2009). Regieren in Europa: Effektiv und demokratisch?. Frankfurt a.M./New York: Campus. Schmitter, P., (1974). Still the Century of Corporatism? The Review of Politics. 36(1), 85–131. Senden, L. A. J., (2004). Soft Law in European Community Law. Oxford/ Portland: Hart Publishing. Snyder, F., (1994). Soft Law and Institutional Practice in the European Community. In: S. Martin, ed. The Construction of Europe: Essays in honour of Emile Noël. Dordrecht: Kluwer Academic Publishers. pp. 197–225. Ştefan, O., (2012). European Union Soft Law: New Developments Concerning the Divide Between Legally Binding Force and Legal Effects. The Modern Law Review. 75(5), 879–893. Terpan, F., (2015). Soft Law in the European Union—The Changing Nature of EU Law. European Law Journal. 21(1), 68–96. Traversa, E. and Flamini, A., (2015). Fighting Harmful Tax Competition Through EU State Aid Law: Will the Hardening of Soft Law Suffice. European State Aid Law Quarterly. 14(3), 323–331. Trubek, D. M. and Trubek, L. G., (2005). Hard and Soft Law in the Construction of Social Europe: the Role of the Open Method of Co-ordination. European Law Journal. 11(3), 343–364. Woll, C. and Jacquot, S., (2010). Using Europe: Strategic Action in Multi-level Politics. Comparative European Politics. 8(1), 110–126. Wratil, C. and Hobolt, S. B., (2019). Public Deliberations in the Council of the European Union: Introducing and Validating DICEU. European Union Politics. 20(3), 511–531. Xanthoulis, N., (2021). Legal Effects and Reviewability of EU Acts: Between Institutional Intentions and Perceptions of Concerned Parties. In: M. Eliantonio, E. Korkea-aho and O. Ştefan, eds. EU Soft Law in Member States: Theoretical Findings and Empirical Evidence. Bloomsbury: Hart Publishing. pp. 305–323. Zeitlin, J. and Trubek, D. M., eds., (2003). Governing Work and Welfare in a New Economy: European and American Experiments. Oxford: Oxford University Press. Zhelyazkova, A. et al., (2015). Beyond Uniform Integration? Researching the Effects of Enlargement on the EU´s Legal System (MAXCAP). Berlin: MAXCAP Working Papers Series.

Interviews ANCT (Agence nationale de la cohésion des territoires), (2021). Interview of 21 January 2021. BMEL (Bundesministerium für Ernährung und Landwirtschaft), (2020), Interview of 30 September 2020. BMWi (Bundesministerium für Wirtschaft und Energie), (2020). Interview of 17 September 2020. BVL (Bundesamt für Verbraucherschutz und Lebensmittelsicherheit), (2020). Interview of 12 November 2020. French Administration, (2021a). Interview of 18 January 2021 French Administration, (2021b). Interview of 19 January 2021. MWAE Brandenburg (Ministerium für Wirtschaft, Arbeit und Energie des Landes Brandenburg), (2020). Interview of 8 December 2020. Régions de France, (2020). Interview of 15 December 2020. StMWi Bayern (Bayerisches Staatsministerium für Wirtschaft, Landesentwicklung und Energie), (2020). Interview of 5 November 2020.

PART IV CRISES, CHALLENGES AND CHANCES

21. Fighting a hard battle with a soft weapon: is international climate change law softening? Kati Kulovesi and María Eugenia Recio

1.

INTRODUCTION: A NEW, SOFTER ERA OF INTERNATIONAL CLIMATE GOVERNANCE?

Is international climate change law softening? Our analysis of this question takes place against the backdrop of commentaries on the 2015 Paris Agreement as a hallmark of a new era whereby the legally binding structures of the top-down Kyoto Protocol are replaced by a softer, bottom-up framework for global climate governance. We link our analysis to scholarship on the ‘softening’ of international law more broadly. While the textbook account of public international law remains one of legal positivism, this description leaves out a considerable body of normative activity as states commonly collaborate based on soft law instruments and informal processes. Scholars have explored these phenomena through somewhat different terminology and emphasis, including informal international lawmaking (Pauwelyn et al. 2012), deformalization of international law (d’Aspremont 2011), the juxtaposition of a culture of legal formalism with a culture of dynamism (Koskenniemi 2001) and general discussion on the role and function of international soft law (Chinkin 1983; Weil 1983; Klabbers 1996; Abbott and Snidal 2000; Shaffer and Pollack 2009; Vihma 2012; Boyle 2019). Divergent views remain, including on whether these developments are fundamentally transforming the traditional, state-centric space of formal international law itself, or whether they are parallel developments that leave the core of public international law largely intact (Osofsky 2012, p. 179). While our ultimate conclusion is more critical and questions the popular view that the Paris Agreement is drastically softer than the previous climate treaties, the starting point for our analysis is that international climate change law reflects prima facie many of the characteristics associated with the softening of international law. The Paris Agreement has been described as a treaty ‘cast in very soft terms, setting distant targets but containing few hard and fast obligations’ (Klabbers 2018, p. 17). Moreover, a considerable proportion of norms of international climate change law can be found outside the three formal climate treaties, the 1992 United Nations Framework Convention on Climate change (UNFCCC), the 1997 Kyoto Protocol and the Paris Agreement. Hundreds of decisions adopted by the UNFCCC’s governing treaty body, the Conference of the Parties (COP) and equivalent bodies under the Kyoto Protocol and the Paris Agreement, constitute the most relevant examples. Our focus here is on the softening of international climate change law in terms of instruments and legal norms under the UNFCCC. However, changes have also been observed with respect to actors and processes. Indeed, the UNFCCC is increasingly seen as just one of the multiple sites of polycentric climate governance (Jordan et al. 2018). International climate change law, then, can be seen as a part of a broader body of global (Farber and Peeters 2016; van Asselt, et al. 2022) or transnational climate change law (Affolder 2021) in which diverse informal pro320

Fighting a hard battle with a soft weapon  321 cesses and norms as well as non-state actors play a role alongside with traditional international law originating from states. With this image in mind, we will examine in more detail the initial impression that international climate change law is softening. Section 2 contains a brief overview of general scholarly discussions on the softening of international law to provide doctrinal background and context for our analysis. In Section 3, we explore countries’ and experts’ views on the choice between hard and soft law instruments during negotiations that lead to the adoption of the Paris Agreement. In Section 4, we analyze how the legal design and nature of countries’ commitment to mitigate climate change have evolved from the UNFCCC through the Kyoto Protocol to the Paris Agreement. In Section 5, we present our conclusions.

2.

BRIEF OVERVIEW OF SCHOLARSHIP ON ‘SOFTENING’ OF INTERNATIONAL LAW

According to the positivist textbook account of international law, formal sources are the key for identifying its contents and distinguishing legal norms from other social norms. In the words of a prominent textbook, ‘if a candidate rule is attested by one or more of the recognized “sources” of international law, then it may be accepted as part of international law’ (Crawford 2012, p. 20). The doctrine of sources is hence ‘what allows norms and standards to be formally anchored in a legal order, generate bindingness and be eligible as materials for international arguments’ (d’Aspremont 2019, p. 71). Article 38.1 of the Statute of the International Court of Justice (ICJ) is the starting point for identifying sources of international law. It lists treaties between states, customary international law derived from the practice of states, and general principles of law recognized by civilized nations as the primary sources of law. The ICJ Statute further mentions judicial decisions and the teachings of the most highly qualified publicists of the various nations as ‘subsidiary means for the determination of rules of law’. Article 38.1 of the ICJ Statute is, however, commonly seen as a non-exhaustive list. ICJ jurisprudence, for example, has treated unilateral declarations (ICJ 1974, p. 253) and joint press communiqués (ICJ 2010, p. 14) as binding expressions of state consent. In international diplomatic practice, also a range of other instruments are commonly used. These include resolutions by international organizations, declarations by international conferences, voluntary codes of conduct as well as decisions by treaty bodies, such as COPs. Such instruments are commonly placed in the vast and diverse category of ‘soft law’. It is commonly accepted in both judicial practice and scholarship that soft law instruments can have legal effects, for example, by providing guidance on the interpretation of international law or paving the way for the emergence of new norms. In the words of Higgins, ‘legal consequences can also flow from acts which are not, in the formal sense, “binding”’ (Higgins 1994, p. 24). For example, in its Namibia Advisory Opinion, the ICJ found that UN General Assembly resolutions, ‘while manifestly not binding, were not without legal effect’ (ICJ 1971, p. 50). In international legal scholarship, soft law instruments are often described as being more flexible and easier to replace or amend than treaties, which is seen as an explanation as to why states have leaned towards more informal norms (Goldman 2012, p. 336). According to Chinkin (1983, p. 866):

322  Research handbook on soft law Soft law instruments allow for the incorporation of conflicting standards and goals and provide States with the room to manoeuvre in the making of claims and counterclaims. While this process inevitably causes normative confusion and uncertainty in terms of the traditional sources of international law, it is probably the inevitable consequence of unresolved pressures for change in international law.

Similarly, Boyle argues that soft law ‘allows states to tackle a problem collectively at a time when they do not want to shackle their freedom of action’ (Boyle 2019, p. 102). In the context of international environmental law, ‘soft law is often relied on to provide the detailed technical rules and standards’ (ibid.). States can also sometimes adopt soft law ‘with the intention that international institutions, courts, and states should apply them when relevant’ (ibid.). Scholarly views of soft law are far from uniform, however, and principled differences can be detected between political scientists and legal scholars in particular. Some legal scholars have opposed the very notion of ‘soft law’, finding it impossible or even dangerous to replace the traditionally binary legal thinking distinguishing between binding and non-binding norms as well as law and non-law with the idea that legal norms could come in shades of grey and bind to varying degrees. According to Weil’s famous article, the danger of diffusing the parameters of what is legally binding on states ‘might well destabilize the whole international normative system and turn it into an instrument that can no longer serve its purpose’ (Weil 1983, p. 423). According to Klabbers, then, ‘law can be more or less specific, more or less exact, more or less determinate, more or less wide in scope, more or less pressing, more or less serious, more or less far-reaching; the only thing it cannot be is more or less binding’ (Klabbers 1996, p. 167). Also Koskenniemi has criticized the shift in vocabulary ‘from rules to regulation, from government to governance, from responsibility to compliance, from legality to legitimacy and from legal expertise to international relations expertise’ (Koskenniemi 2005; as phrased in d’Aspremont 2011, p. 510). Koskenniemi opposes the tendency to replace the idea of formally valid law by ‘a vocabulary of empirical political science’ (Koskenniemi 2007, p. 8). The problem, according to him, is that ‘if law is defined as what causes compliance, then the distinction between law and power is lost’ (ibid., p. 18). Therefore, legal validity cannot be about cost-benefit analyzes and the usefulness of certain norms; it is about a legal claim being ‘right’ in the relevant legal system (ibid., p. 19). Thus, while formalism in the words of Koskenniemi (2001, p. 500): … may no longer be open as a jurisprudential doctrine of the black and white of legal validity (a position perhaps never represented by anyone), nothing has undermined formalism as a culture of resistance to power, a social practice of accountability, openness, and equality whose status cannot be reduced to the political positions of any of the parties whose claims are treated within it.

In contrast, especially in broader social science literature, the emphasis is on empirical effectiveness of international norms. For rationalist international relations scholars, hard and soft law have distinct attributes that States use in different contexts (Shaffer and Pollack 2009). In their seminal article, Abbott and Snidal place international norms on a continuum from hard to soft law and highlight the ‘great variety in the degrees and forms of legalization’, arguing that ‘international actors choose softer forms of legalized governance when those forms offer superior institutional solutions’ (Abbott and Snidal 2000, p. 421). Also the theory of interactional international law developed by Brunnée and Toope (using international climate change law as a prominent example) builds on the idea that the boundary between law and non-law is fluid (Brunnée and Toope 2010, p. 355). For them, interactional international law eschews formal

Fighting a hard battle with a soft weapon  323 indicators of bindingness and the notions of hard and soft law (Brunnée 2002, pp. 35–36). Instead, ‘legal norms are distinguished from non-legal norms through internal characteristics, which entail distinctive legal legitimacy and persuasiveness’ (ibid., p. 36). Overall, scholarly perceptions on soft law seem to have gradually become more positive. Klabbers observes that ‘it now seems well established among international lawyers that soft law exists, that it has its functional uses and, to a wide majority, is quite simply a good thing, and no longer solely the poorer cousin to hard law’ (Klabbers 2006, p. 1193). He notes, however, that there are important conceptual challenges and no clear definition of what constitutes soft law and how it can be distinguished from hard law (ibid.). Also Pauwelyn notes that most scholars seem to nowadays accept that legal normativity is ‘a matter of degree with varying scales of normativity and a large grey zone between what is law and what is not’ (Pauwelyn 2012, p. 127). In his view, however, ‘there must be a conceptual line – not a zone – separating law from other norms even if this may be difficult to apply in practice or be of little practical consequence’ (ibid., p. 128). Making a distinction between ‘being law and having legal effect’ can, in his view, be helpful in this regard (ibid.). International climate change law provides an interesting example to study these issues in light of recent state practice and literature – to what extent is international climate change law softening? If it is softening, are there signs of this being due to a new ideological preference in the international legal mindset for soft law?

3.

A NEW CLIMATE TREATY, COP DECISION(S) OR SOMETHING ELSE? VIEWS ON HARD AND SOFT LEGAL INSTRUMENTS DURING NEGOTIATIONS FOR THE PARIS AGREEMENT

3.1

Key Milestones and Concepts in International Climate Change Law

How were the respective roles of hard and soft law perceived during negotiations on long-term international climate change cooperation that took place under the UNFCCC between 2005 and 2015? Does the outcome of these negotiations lend support to the view that international climate change law is softening? These questions are the focus of this section. We start, however, with a brief overview of the evolution of international climate change law to provide some background for our analysis. As a framework convention, the UNFCCC sets out the main goals, principles, and institutions for international climate change collaboration. It lacks, however, detailed climate change mitigation commitments. COP 1 in 1995 therefore launched negotiations for ‘a protocol or another legal instrument’ (UNFCCC 1995, preamble) to establish mitigation targets for developed countries while not introducing any new commitments for developing countries (UNFCCC 1995, para. 2(b)). The mandate reflects the fundamental importance of the principle of common but differentiated responsibilities and respective capabilities (CBDRRC) in the evolution of international climate change law. The UNFCCC allocates a leadership role to Annex I Parties, a group of 41 countries that were either members of the Organization for Economic Cooperation and Development (OECD) in the early 1990s or part of the former Soviet Bloc. All others fall into the category of non-Annex I countries, a diverse group that includes the Least Developed Countries (LDCs) and small island developing states (SIDS), OECD members Mexico and South Korea, emerging economies Brazil, China, India and

324  Research handbook on soft law South Africa, as well as oil-producing countries, such as Saudi Arabia, Iran, United Arab Emirates, Qatar and Nigeria. In the 1997 Kyoto Protocol, a legal ‘firewall’ was constructed between Annex I and non-Annex I countries’ mitigation commitments and the Protocol’s legally binding emission targets only applied to Annex I countries. This was politically highly controversial and the main reason why the US – the world’s largest emitter of greenhouse gases until overtaken by China in 2006 – decided not to ratify the Kyoto Protocol. This, combined with the rapid growth of greenhouse gas emissions in non-Annex I countries, seriously compromised the effectiveness of the Kyoto Protocol. Furthermore, the Kyoto Protocol only set emission targets for the period from 2008 to 2012, leaving open the question of post-2012 climate action. Negotiations therefore took place in 2005–2015 under the UNFCCC on long-term climate change cooperation. A key objective for Annex I countries was to dilute the Annex-based differentiation (Rajamani and Guérin 2017, p. 82). This was strongly opposed by powerful non-Annex I countries emphasizing developed countries’ historical responsibility for climate change. The 2009 UN Climate Change Conference in Copenhagen constituted the first attempt to adopt a new international climate treaty. The Conference was subject to unprecedented media attention and more than a hundred heads of State and government participated in it. However, given the slow progress ahead of Copenhagen, ‘expectations for the outcome were gradually scaled back’ (Brunnée and Toope 2010, p. 140). In the end, the Conference failed to fulfil even the more modest expectations, culminating in the Copenhagen Accord (UNFCCC 2010), a highly controversial document with ‘a questionable legal basis and fragile connection’ to the UNFCCC (Rajamani 2010, p. 834). The Accord resulted from informal negotiations by a small group of Heads of State and diplomats during the final hours of the deeply troubled conference (Meilstrup 2010, p. 131). To make matters worse, US President Barack Obama announced the Copenhagen Accord to international media before most parties even knew about its existence (Spence 2022). Protesting both the untransparent negotiating process and weak substance, five developing countries blocked the Accord’s formal adoption during a highly acrimonious COP closing plenary. Ultimately, the COP agreed to merely ‘take note’ of the Copenhagen Accord (UNFCCC 2010). The disappointing outcome from Copenhagen illustrated clearly that the world was not ready for a new climate treaty. As one of the negotiators describes, ‘the views between developed and developing countries of what needed to be delivered through a new climate agreement were completely at odds’ (Mpanu-Mpanu 2021, p. 182). In 2010, further negotiations took place leading to the adoption of a package of COP decisions known as the Cancun Agreements (UNFCCC 2011). Despite many advances, a number of core issues remained unsettled, including those related to climate change mitigation. COP 17 in 2011 thus focused on a new negotiating mandate. Politically, an important factor enabling a positive outcome was the agreement to continue the Kyoto Protocol for a second commitment period from 2013 to 2020 (Maljean Dubois and Wemaere 2012, p. 190). At the same time, a gradually expanding coalition that came to include the European Union (EU), small island developing states, LDCs, South Africa, Brazil and also the US, called for a mandate to negotiate a new post-2020 climate treaty (Bodansky 2016, p. 144). However, India in particular wanted to leave open the possibility of these negotiations resulting in a non-binding instrument (ibid.). The compromise was a mandate to adopt ‘a protocol, another legal instrument or an agreed outcome with legal force under the Convention applicable to all parties’ (UNFCCC 2011, para. 2). The first two options are clear, however, the formulation of the third alternative – an ‘agreed outcome with

Fighting a hard battle with a soft weapon  325 legal force’ – was intentionally obscure (Rajamani 2012, p. 501). It provided assurances that the outcome would have a legal force (Bodansky 2016, p. 144), but allowed India to continue to stress its opposition to an international climate treaty (Morcel 2012, p. 7). The outcome of these negotiations was the Paris Agreement the legal nature of which will be analyzed in detail in Section 3.2. below. 3.2

Negotiations on the Legal Design of the Future Climate Change Regime

During the negotiations on long-term international climate change cooperation, a range of views was put forward concerning the choice between hard and soft law instruments. Proposed legal outcomes included a new protocol, an amendment to the UNFCCC, COP decision(s), ministerial declaration and different combinations of these (Rajamani 2009, p. 805). Looking at country positions, most developed countries preferred a new protocol replacing the Kyoto Protocol and bringing all countries under the same treaty and mitigation regime (Kulovesi and Gutiérrez 2009, p. 242). The US, in turn, proposed a new implementing agreement under the UNFCCC (UNFCCC 2009). The terminology was influenced by domestic considerations, namely the desire to disconnect the outcome from the Kyoto Protocol and conclude an instrument that the President could ratify based on his executive powers. This is also reflected in the name and legal design of the Paris Agreement (Rajamani 2015, p. 826; Oberthür and Bodle 2016, p. 45). Climate vulnerable developing countries like Costa Rica and Tuvalu initially preferred a combination of amending the Kyoto Protocol and adopting a new protocol (Kulovesi and Gutiérrez 2009, p. 242). In contrast, key emerging economies, including China, Brazil and the Republic of Korea, preferred continuing under the Kyoto Protocol for Annex I countries while addressing developing country mitigation and Convention track outcomes through COP decisions (Morgenstern 2009, p. 242). As the negotiations progressed, it became increasingly evident that the vast majority of states, civil society organizations and experts converged on the importance of a new formal climate treaty, which would address climate change comprehensively. According to a leading think tank, ‘it is important that this final outcome be legally binding. Countries will deliver their strongest possible efforts only if they are confident that their major counterparts and competitors are as well’ (Pew Center on Global Climate Change 2009, p. 1). For Rajamani, ‘a legally binding treaty has considerable symbolic value. It would be in keeping with the gravity of the climate change problem and it would withstand the high level of public scrutiny that the process is subject to’ (Rajamani 2009, p. 805). Given that there was no consensus amongst key countries concerning the need for a new climate treaty, the implications of reflecting the outcome of the negotiations in soft law instruments, such as COP decisions, were also discussed (Morgenstern 2009, pp. 243–245). The mainstream view is that COP decisions are legally binding only where the relevant treaty has granted the COP the authority to adopt binding decisions concerning a specific question (ibid., p. 243). There are, however, different interpretations of which treaty provisions contain such an authorization (Brunnée 2002, p. 24). Views on the legal nature of other COP decisions remain more ambiguous with most scholars classifying them as non-binding. Brunnée argues that a significant ‘grey zone’ has developed with respect to the COP’s law-making powers (Brunnée 2002, p. 32). Yamin and Depledge argue that ‘many Parties would view the main role of the COP as a political one, with its decisions being primarily of a political rather than legal nature’ (Yamin and Depledge 2004, p. 405).

326  Research handbook on soft law Overall, the 2005–15 negotiations on long-term international climate change collaboration illustrate that for most countries, experts and other stakeholders, the choice between hard and soft law was of fundamental importance. As Bodansky has observed, ‘states clearly thought the issue of legal form mattered, and this belief itself became an important reality in the negotiations, which significantly shaped the ultimate result’ (Bodansky 2016, p. 142). A widely shared sentiment can be detected that soft law instruments were clearly less desirable than an international treaty. For example, an outcome based on COP decisions would involve ‘serious uncertainties regarding the legal nature and bindingness of COP decisions, their status in domestic law and the authority of the COP to establish central commitments by way of simple decisions’ (Morgenstern 2009, p. 247). It was also obvious that the two soft law instruments that emerged during the negotiations – the Copenhagen Accord and the Cancun Agreements – were not deemed as adequate. Instead, the negotiations continued until a new formal climate treaty was adopted. Thus, on 12 December 2015, all UNFCCC Parties agreed to adopt the Paris Agreement (UNFCCC 2016). The outcome was not explicitly identified as an international treaty or a new protocol mainly due to the US constitutional considerations discussed above. However, there were clear indications that the parties intended the outcome to be a formal international treaty within the meaning of the Vienna Convention on the Law of Treaties. The most important cues to this effect include Articles 16–27 of the Paris Agreement, covering issues, such as signature, consent to be bound, entry into force, depositary, reservations and withdrawal (Voigt 2017, p. 352). Such provisions are characteristic of international treaties but are never included in COP decisions. Nevertheless, the legal nature of the Paris Agreement has been subject to some scholarly discussion. Differences resembling those in the general discussion on the ‘softening’ of international law can be detected, including principled differences between some legal scholars and political scientists (Pauwelyn and Andanova 2015). Reacting to the outcome of COP 21, prominent international relations scholars emphasized what they described as the ‘voluntary’ elements of the Paris Agreement. For Slaughter, the Paris Agreement fell ‘woefully short’ by standards of traditional treaty and was ‘essentially a statement of good intentions’ (Slaughter 2015). However, the Agreement’s deficits were also ‘its greatest strengths’ (ibid.) In Slaughter’s view, the Agreement exemplified a new kind of global governance ‘that is far better suited to the kinds of global problems we face today’ (ibid.). Similarly, Falk argued that the Paris Agreement ‘went to great lengths to avoid obligating the parties, making compliance with pledged reductions in carbon emissions an unmistakably voluntary undertaking’ (Falk, 2016, emphasis added). According to Falkner, the Agreement brought to an end ‘a misguided approach of establishing mandatory emission reductions’ (Falkner 2016, p. 1107). The Paris Agreement therefore promised ‘a more realistic path’ than the Kyoto Protocol ‘mainly because it has managed to better align international climate policy with the realities of international climate politics’ (ibid., emphasis in the original). Assessments by several other scholars were markedly different, however. According to Rajamani to dismiss the Paris Agreement as voluntary ‘would do grave injustice to this carefully negotiated instrument’ (Rajamani 2016, p. 338). Bodansky regretted that ‘confusion about the legal character of the Paris Agreement is widespread’ (Bodansky 2016, p. 142). He stressed that the Paris Agreement is clearly a treaty within the meaning of international law and creates ‘legal obligations for its parties; and compliance with these obligations is not voluntary’ (ibid.) Bodle and Oberthür observe that it is necessary to separate questions, such as

Fighting a hard battle with a soft weapon  327 whether the Paris Agreement is a binding international treaty, from questions of how prescriptive and precise its specific provisions are, and what the legal nature of individual obligations or commitments is, for example, whether a given obligation is one of result or conduct (Bodle and Oberthür 2017, p. 91). Indeed, to better understand whether the Paris Agreement represents the ‘softening’ of international climate change law, it seems necessary to examine in more detail its provisions related to climate change mitigation and contrast them with the UNFCCC and the Kyoto Protocol. Scrutinizing this issue will constitute the second step in our analysis of whether international climate change law is softening. The conclusion from the first step is that the negotiations on the choice between hard and soft law instruments do not support the idea that international climate change law is softening. This is based on the predominant view among countries and experts concerning the choice of legal instrument, making a clear distinction between hard and soft law and expressing a strong preference for a hard law instrument given its international and domestic legal implications.

4.

THE ROLE OF INTERNATIONAL LAW IN REGULATING COUNTRIES’ GREENHOUSE GAS EMISSIONS?

4.1

Bottom-up and Top-down Approaches in the Evolution of International Climate Change Law

Is international climate change law softening in terms of the nature of legal obligations included in international climate treaties? This question forms the focus of this section, albeit with important limitations. International climate change law covers a broad range of substantive issues from adaptation, loss and damage, finance, technology and capacity building to mitigation. We limit our analysis to the general role of international law in regulating countries’ greenhouse gas emissions – a crucial but limited issue that was controversial already in the early 1990s when the UNFCCC was negotiated. According to Bodansky, international norms related to climate change mitigation have been shaped and re-shaped by conflicting demands for a strong international legal framework and binding emission targets (commonly known as the top-down approach) on one hand, and soft international norms and national autonomy (known as the bottom-up approach) on the other (Bodansky 2013, p. 35). While the top-down/ bottom-up characterization unquestionably captures some of the key divides concerning the role of international law in regulating climate change mitigation, recent scholarship has challenged it as too simplistic. According to van Asselt and Zelli, ‘the climate regime has always been a hybrid of top-down and bottom-up elements’ (van Asselt and Zelli 2018, p. 31). Depledge argues that the ‘top-down/bottom-up metaphor sets up a false dichotomy and confuses more than it clarifies’ (Depledge 2022). Our assessment below echoes these more critical views. We will start with a brief general overview of how international climate change law has evolved concerning mitigation obligations. We will then analyze in more detail the relevant provisions in the Paris Agreement. During negotiations for the UNFCCC, the target-based approach to emission reductions that had been successfully used in the 1987 Montreal Protocol on Substances that Deplete the Ozone Layer was advocated in particular by countries currently belonging to the EU; in contrast, the US, Japan and the Soviet Union were more sceptical of its suitability to the climate

328  Research handbook on soft law change context (Bodansky 2013, p. 5). The US favoured instead a bottom-up approach, arguing that countries should define and develop national climate change mitigation programs and policies (ibid., p. 6). As a compromise, the UNFCCC includes reflections of both proposals. It includes a general requirement for Annex I Parties to develop national mitigation policies and measures, while also setting a quantified target for Annex I countries to return emissions to their 1990 levels by 2000. It is, however, commonly understood that the latter does not represent a legally binding target to be achieved by each Annex I Party, but ‘a quantified “aim” that “Parties should strive to achieve”’ (Yamin and Depledge 2004, p. 110). The Kyoto Protocol was stronger and more specific in terms of setting mitigation targets and is often described as a top-down instrument (Depledge 2022). However, its targets were not based on science or otherwise imposed from top down, but reflected what countries were politically willing to accept at the time (van Asselt and Zelli 2018, p. 31). The main issues debated during the Kyoto negotiations were the stringency of the mitigation commitments and the flexibility that Annex I countries would have in implementing them (Bodansky, Brunnée and Rajamani 2017). The EU advocated strong commitments and limited flexibility, while the US and other Annex I countries preferred weaker commitments and greater flexibility (ibid.). Again, the outcome reflects both views, but this time with a stronger emphasis on the so-called top-down approach. The Kyoto Protocol established an economy-wide emission target for each Annex I Party for the period between 2008 and 2012, characterized as a ‘specific, time-bound obligations of result’ (Werksman and Herbertson 2010, p. 127). The Protocol also include a strong monitoring, reporting and verification system as well as a Compliance Committee with Facilitative and Enforcement Branches competent to impose sanctions on non-complying States. On the other hand, the Protocol allowed States flexibility in meeting their targets, starting from the ability to choose which economic sectors and greenhouse gases to focus on, and self-define the specific policies used (Bodansky, Brunnée and Rajamani 2017). The five-year commitment period provided some flexibility in terms of the timing of emission reductions and the Protocol also included three market-based flexibility mechanisms allowing Annex I Parties to fulfil part of their commitments by purchasing emission allowances from other countries (ibid.). It is commonly argued that international climate change law has subsequently shifted from the Kyoto Protocol’s top-down approach to a softer bottom-up approach (McGee and Stefferk 2016, p. 37). The shift started with the 2009 Copenhagen Accord, which allowed countries to self-define their national mitigation goals and actions without enshrining them into an international treaty. One of the many criticisms against the Copenhagen Accord, however, was that it was ‘too bottom up’ (Biniaz with Pershing 2021, p. 142). During negotiations for the Paris Agreement, a number of countries therefore pushed for stronger international norms related to climate change mitigation. Below, we will first describe the outcome of these negotiations and then analyze it in light of the argument that international climate change law is softening. 4.2

Mitigation under the Paris Agreement

Given how long and convoluted the negotiations leading to the Paris Agreement were, it is hardly surprising that also their outcome is complex, comprising collective goals, individual obligations that require nuanced differentiation between substantive and procedural obligations, as well as diverse elements designed to increase transparency and accountability.

Fighting a hard battle with a soft weapon  329 Article 2.1 of the Paris Agreement specifies a collective goal to limit the global average temperature increase to well below 2 degrees Celsius above pre-industrial levels and pursue efforts to limit it to 1.5 degrees. Article 4.1 specifies that the goal is to be achieved through the peaking of global greenhouse gas emissions as soon as possible, and balancing anthropogenic greenhouse gas emissions and removals by sinks in the second half of the century. Country-level action to reach these goals takes the form of Nationally Determined Contributions (NDCs). Under Article 4.2, each Party must prepare, communicate to the UNFCCC Secretariat and maintain a successive NDC ‘that it intends to achieve’. Each party must also pursue domestic mitigation measures with the aim of achieving the mitigation objective in its NDC. Under Article 4.9, parties must update their NDCs at five-year intervals. Article 4.3 creates a strong expectation that each NDC ‘will’ represent progression compared with the previous one and reflect a party’s ‘highest possible ambition’. Each party must also regularly report under Article 13.7 on progress made in implementing its NDC. Finally, a Global Stocktake under Article 14 will take place at five-year intervals to evaluate collective progress towards the Paris Agreement’s goals. Given the fundamental role of differentiation in shaping international climate change law, it is also worth noting that the Paris Agreement does not refer to the controversial categories of Annex I and non-Annex I countries but relies on a legal structure designed to promote self-differentiation (Rajamani and Guérin 2017, pp. 81–88; Voigt and Ferreira 2016, p. 285). The Agreement includes various references to the CBDRRC principle but includes language on national circumstances wherever the principle is mentioned so as to recognize a continuum of countries’ national circumstances (Biniaz with Pershing 2021) and add a dynamic element to the principle’s interpretation (Rajamani and Guérin 2017, p. 84). There have been conflicting assessments of the mitigation provisions in the Paris Agreement and, as seen above, scholars like Slaughter, Falk and Falkner have characterized them as ‘voluntary’. Subsequent literature has, however, painted a more nuanced image. Like in the UNFCCC and the Kyoto Protocol, provisions on climate change mitigation in the Paris Agreement embed careful compromises and political balancing acts. In the words of Rajamani, ‘each provision contains a carefully calibrated mix of the elements of legal character’ (Rajamani 2016, p. 353). The popular perception of the Paris Agreement as a soft or voluntary undertaking is largely based on the fact that countries’ mitigation goals are set nationally, not formally part of the treaty and mainly obligations of conduct to prepare, maintain and communicate an NDC and to pursue domestic mitigation measures rather than obligations of result to achieve the NDC target like in the Kyoto Protocol (Oberthür and Bodle 2016, pp. 51–54; Mayer 2018, pp. 259–260; Rajamani and Brunnée 2017, p. 542). Such a legal design was necessary for China, India and the US to join the Agreement (Winkler 2017, p. 147). However, as always, the outcome reflects a compromise between those advocating a strong role for international law in regulating greenhouse gas emissions and those preferring weaker legal controls; demands for strong international obligations were not completely dropped in favour of a voluntary system. This is clear when looking at the details, including that Article 4.2 of the Paris Agreement provides a standard for assessing domestic measures to implement the NDCs; such measures must be pursued ‘with the aim of achieving the NDCs’ objectives’ (Oberthür and Bodle 2016, p. 54). The Paris Agreement also imposes an obligation ‘to take adequate measures’ towards the realization of their NDCs (Mayer 2018, p. 262). Furthermore, there is a clear obligation to regularly report whether a party is on track to achieve its NDC (Winkler

330  Research handbook on soft law 2017, p. 147). In other words, while climate change mitigation under the Paris Agreement is not regulated through targets and timetables as under the Kyoto Protocol, the Agreement is by no means devoid of legal obligations. 4.3

Comparing Mitigation-related Provisions in the UNFCCC, Kyoto Protocol and the Paris Agreement – is Paris Agreement Softer?

In light of the above, to what extent is the Paris Agreement softer than the UNFCCC and the Kyoto Protocol and does it signify a major softening of international climate change law as is often claimed in literature? We complete our analysis by comparing key mitigation provisions in the UNFCCC, Kyoto Protocol and the Paris Agreement in Table 21.1 and the text below. The conclusions are not straightforward; especially the Kyoto Protocol and Paris Agreement seem to have different strengths and weaknesses. Indeed, aspects of the Paris Agreement appear even stronger than the Kyoto Protocol, especially when considering accumulating evidence of domestic legal effects. Comparing mitigation commitments, the UNFCCC included general commitments for all countries and a non-binding additional target for Annex I countries. The Kyoto Protocol’s strength was that it established binding mitigation targets and a clear legal obligation to achieve them. However, as noted above, the targets were not based on science or otherwise set top-down, but reflected the emission reductions that the affected countries were politically willing to accept at the time (van Asselt and Zelli 2018). Indeed, according to Depledge, ‘the pledging process for the Kyoto Protocol’s emission targets and the Paris Agreement’s NDCs were actually very similar’ (Depledge 2022, p. 16). The Kyoto targets only covered a five-year period and applied to a small group of Annex I countries. In comparison, the Paris Agreement’s provisions on mitigation cover all countries. The five-yearly mitigation cycles of the Paris Agreement can, in theory, continue without an end date making the Agreement look stronger than the Kyoto Protocol in this respect. Where the Paris Agreement is weaker than the Kyoto Protocol is that it lacks a clear legal obligation of result for individual parties to achieve their NDCs and ensure that their emissions do not exceed the mitigation target. However, as seen above, in the Agreement the negotiators attempted to balance this through various obligations of conduct. The comparison is therefore not one between strong legal obligations in the Kyoto Protocol and no legal obligations in the Paris Agreement, but between obligations of conduct and obligations of result, in other words, the question also boils down to the relative strength of procedural and substantive legal obligations. The example of Canada withdrawing from the Kyoto Protocol rather than taking measures to fulfil its mitigation commitments illustrates that also the Protocol entailed significant weaknesses. At the same time, the obligation of conduct to prepare and regularly update an NDC holds important potential to promote climate awareness and engage relevant domestic institutions and stakeholders in regular climate policy discussions in ways unseen when setting the Kyoto target. From a broader perspective, it is not uncommon for commitments in international environmental law to be expressed as either obligations of conduct or obligations of result (Werksman and Herbertson 2010, p. 122). Even in cases where the obligations are very vague and imprecise and therefore not easily unenforceable at the international level (ibid., p. 122):

Fighting a hard battle with a soft weapon  331 … the international legal character of an MEA [multilateral environmental agreement] will trigger domestic ratification procedures, and in some circumstances … the enactment of enabling legislation. The domestic legal effect of an international treaty can be more significant than what is reflected in the international instrument.

Table 21.1

Comparing mitigation-related provisions in the UNFCCC, Kyoto Protocol and Paris Agreement UNFCCC

Long-term mitigation goal(s)

Kyoto Protocol

Paris Agreement

Stabilization of GHG emissions No distinct goal

Limit the global average

in the atmosphere at a level

temperature increase to well

that would prevent dangerous

below 2 degrees Celsius and

anthropogenic interference with

pursue efforts to limit it to 1.5

the climate system. (Art. 2)

degrees (Art. 2.1.a) Balance between anthropogenic GHG emissions and removals by sinks in the second half of the century (Art. 4.1)

Short/mid-term mitigation goals Annex I Parties aim to return,

An overall reduction in GHG

Reach global peaking of GHG

individually or jointly, their

emissions by Annex I countries emissions as soon as possible,

GHG emissions to 1990 levels

of least 5% below 1990 levels

recognizing that this will take

by 2000 (Art. 4.2.b)

in 2008–12 (Art. 3.1)

longer for developing countries,

Each Annex I Party must make and undertake rapid reductions demonstrable progress by 2005 thereafter (Art. 4.1) Process for setting country-level No targets for individual Parties International negotiation

Nationally determined without

mitigation targets

building on what Annex

international negotiation

I Parties were willing to put

but following international

forward

guidance on substance and

Legal nature of country-level

process No targets for individual Parties Emission target for each Annex Obligation for all Parties to I Party and obligation to ensure prepare, communicate and

mitigation targets

that a Party’s greenhouse gas

maintain an NDC (Art. 4.2)

emissions do not exceed the

Obligation for all Parties to

target in 2008–12 (Art. 3.1)

pursue domestic mitigation

Individual emission targets

measures with the aim of

included in Annex B of the

achieving their NDCs’

Kyoto Protocol

objectives. (Art. 4.2) NDCs recorded in a public registry (Art. 4.12)

Updating of country-level mitigation targets

N/A

Treaty amendment to establish

Each party to communicate

the second commitment period

an NDC every five years (Art. 4.9) informed by the Global stocktake under Art. 14 Each successive NDC ‘will represent’ progression and reflect ‘highest possible ambition’ (Art. 4.3)

332  Research handbook on soft law Evidence is fast accumulating that the Paris Agreement’s legal effects at the domestic level have been ground-breaking. Unlike the UNFCCC and the Kyoto Protocol, the Agreement includes specific mitigation goals. While formulated as collective aims rather than binding obligations for individual states, they provide a long-term direction of travel for both international and national climate policy and act as yardsticks for assessing country-level mitigation action. Looking at the legal effects of these provisions, Article 4.1 of the Paris Agreement has been driving the adoption of carbon neutrality targets by more than a hundred countries (Höhne et al. 2021, p. 820) and thousands of non-state actors (UNFCCC website 2022). National courts in the Netherlands and Germany have referred to the Paris Agreement’s temperature goal in decisions requiring, respectively, a major oil company to reduce its greenhouse gas emissions from its entire value chain (Milieudefensie et al. v. Royal Dutch Shell plc.) and the government of Europe’s largest economy to be more specific about its climate change mitigation targets beyond 2030 (Neubauer et al.). The mitigation goals of the Paris Agreement have also been an important motivation for the adoption and updating of national climate change acts especially in Europe but also elsewhere (The World Bank 2021). Overall, the impact of the Paris Agreement on the evolution of domestic climate law and policy has been clearly much more profound than that of the UNFCCC and Kyoto Protocol. Thus, when looking at the ensemble of mitigation provisions in the UNFCCC, Kyoto Protocol and the Paris Agreement, the latest treaty has as its objective a fundamental economic and societal change never seen under the UNFCCC and the Kyoto Protocol. Evidence is accumulating that the climate sustainability transformation has, in many places, begun, even if still too slowly and the Paris Agreement operates as the engine driving the rapid evolution and expansion of global climate change law. Only from a specific perspective that focuses on the international legal placement of mitigation targets and obligations related to their achievement does the Paris Agreement point towards the softening of international climate change law when compared with the Kyoto Protocol.

5.

CONCLUSIONS: NOT THE POOR COUSIN OF THE KYOTO PROTOCOL

We started by asking whether international climate change law is softening. We did so against the backdrop of literature celebrating the Paris Agreement as a shift towards softer climate governance. With this background in mind, we proceeded to examine the respective roles of hard and soft law in the negotiations on long-term international climate change cooperation under the UNFCCC between 2005 and 2015. Here, we found that most states, stakeholders and experts held a strong preference for an international treaty over soft law instruments. Indeed, neither the 2009 Copenhagen Accord nor the 2010 Cancun Agreements were deemed sufficient, but the negotiations continued until a formal international treaty was adopted. In this sense, we found no evidence of the softening of international climate change law under the UNFCCC in terms of the choice of legal instruments. We then proceeded to analyze provisions related to climate change mitigation in the Paris Agreement, sometimes characterized as ‘voluntary’ and commonly portrayed as significantly softer than the Kyoto Protocol. Our analysis showed that in all three climate treaties – the UNFCCC, the Kyoto Protocol and the Paris Agreement – important compromises have been made concerning the role of international law in regulating countries’ greenhouse gas emis-

Fighting a hard battle with a soft weapon  333 sions, and all three treaties contain traces of the so-called top-down and bottom-up approaches. Therefore, the popular image painted in the literature of the top-down Kyoto Protocol and bottom-up Paris Agreement appeared as too simplistic. Looking at the mitigation-related provisions in the Paris Agreement, it became evident that important trade-offs were made during the negotiations concerning the Agreement’s coverage in terms of countries on the one hand, and the legal form of its mitigation commitments on the other. As a result, the Paris Agreement looks, at first glance, softer than the Kyoto Protocol. However, taking a more comprehensive and detailed view, the question is about the relative strength of procedural and substantive legal obligations. Furthermore, evidence keeps accumulating that some provisions in the Paris Agreement are having considerable legal effects domestically. Therefore, the Paris Agreement appears neither as the poor cousin of the Kyoto Protocol, nor a radical step towards the softening of international climate change law.

REFERENCES Abbott, K. W. and Snidal D., (2000). Hard and Soft Law in International Governance. International Organization. 54(3), 421–456. Affolder, N., (2021). Transnational Climate Law. In: P. Zumbansen, ed. The Oxford Handbook of Transnational Law. New York: Oxford University Press. Biniaz S. with Pershing J., (2021). The United States: Interesting Processes Lined the Road to Paris. In: H. Jepsen et al., eds. Negotiating the Paris Agreement: Insider Stories. Cambridge: Cambridge University Press. pp. 139–159. Bodansky, D., (2013). A Tale of Two Architectures: The Once and Future U.N. Climate Regime. In: H-J. Koch et al., eds. Climate Change and Environmental Hazards Related to Shipping: An International Legal Framework. Leiden: Brill. pp. 35–51. Bodansky, D., (2016). Legal Character of the Paris Agreement. Review of European, Comparative and International Environmental Law. 25(2), 142–150. Bodansky, D., Brunnée, J. and Rajamani, L., (2017). International Climate Change Law. Oxford University Press. Bodle, R. and Oberthür, S., (2017). Legal Form of the Paris Agreement and Nature of Its Obligations. In: D. Klein et al., eds. The Paris Agreement on Climate Change: Analysis and Commentary. New York: Cambridge University Press. pp. 92–106. Boyle, A., (2019). The Choice of a Treaty: Hard versus Soft Law. In: S. Chesterman et al., eds., The Oxford Handbook of United Nations Treaties. New York: Oxford University Press. Brunnée, J., (2002). COPing with Consent: Law-Making Under Multilateral Environmental Agreements Leiden Journal of International Law. 15(1), 1–52. Brunnée, J. and Toope, S. J., (2010). Legality and Legitimacy in International Law. Cambridge: Cambridge University Press. Bundesverfassungsgericht [BVerfG] Neubauer et al. v. Germany, [German Federal Constitutional Court], Mar. 24, 2021, Case No. BvR 2656/18/1, BvR 78/20/1, BvR 96/20/1, BvR 288/20, [Viewed 8 July 2023] Available from http://​www​.bverfg​.de/​e/​rs20210324​_1bvr265618en​.html. D’Aspremont, J., (2011). The Politics of Deformalisation in International Law. Goettingen Journal of International Law. 3(2), 503–550. D’Aspremont, J., (2019). Bindingness. In: J. d’Aspremont and S. Singh, eds., Concepts for International Law. Contributions to Disciplinary Thought. Cheltenham: Edward Elgar. pp. 67–82. Chinkin, C., (1983). The Challenge of Soft Law: Development and Change in International Law. International and Comparative Law Quarterly. 38(4), 850–866. Crawford, J., (2012). Brownlie’s Principles of Public International Law. 8th edition, Oxford University Press. Depledge, J., (2022). The ‘Top-Down’ Kyoto Protocol? Exploring Caricature and Misrepresentation in Literature on Global Climate Change Governance. International Environmental Agreements: Politics,

334  Research handbook on soft law Law and Economics [online]. 2022. [Viewed 12 July 2022]. Available from: https://​doi​.org/​10​.1007/​ s10784​-022​-09580​-9. Falk, R., (2016). Voluntary International Law and the Paris Agreement. Richardfalk.org [online]. 16 January 2016. [Viewed 12 July 2022]. Available from: https://​richardfalk​.org/​?s​=​Paris+​agreement​&​ submit​=​Search. Falkner, R., (2016). The Paris Agreement and the New Logic of International Climate Politics. International Affairs. 92(5), 1107–1125. Farber, D. A. and Peeters, M., (2016). The Emergence of Global Climate Change Law. In: D. A. Farber and M. Peeters, eds. Climate Change Law. Edward Elgar. pp. 687–702. Goldman, M., (2012). We Need to Cut Off the Head of the King: Past, Present and Future Approaches to International Soft Law. Leiden Journal of International Law. 25(2), 335–368. Higgins, R., (1994). Problems and Process: International Law and How We Use It. Oxford University Press. Höhne, N. et al., (2021). Wave of Net Zero Targets Opens Window to Meeting the Paris Agreement. Nature Climate Change. 11(10), 820–822. ICJ. Legal Consequences for States of the Continued Presence of South Africa in Namibia (South West Africa) Notwithstanding Security Council Resolution 276. Advisory Opinion of 21 June 1971. ICJ Reports 1971, p. 16. ICJ. Nuclear Tests. (Australia v. France). Judgment on Admissibility of 20 December 1974. ICJ Reports 1974, p. 253. ICJ. Pulp Mills on the River Uruguay. (Argentina v. Uruguay). Judgment, 20 April 2010. ICJ Reports 2010, p. 14. Jordan, A. et al., (2018). Governing Climate Change Polycentrically: Setting the Scene. In: A. Jordan et al., eds. Governing Climate Change: Polycentricity in Action. Cambridge: Cambridge University Press. Klabbers, J., (2018). On Responsible Global Governance. In: J. Klabbers et al., eds. Towards Responsible Global Governance. Helsinki: University of Helsinki Faculty of Law. pp 11–31. Klabbers, J., (2006). Reflections on Soft International Law in a Privatized World. Lakimies. 104(7–8), 1191–1205. Klabbers, J., (1996). The Redundancy of Soft Law. Nordic Journal of International Law. 65, 167–182. Koskenniemi, M., (2001). The Gentle Civilizer of Nations. The Rise and Fall of International Law 1870–1960. Cambridge: Cambridge University Press. Koskenniemi, M., (2007). Formalism, Fragmentation, Freedom: Speech on Kantian Themes in Today’s International Law. No foundations: Journal of Extreme Legal Positivism. (4), 7–28. Available from: https://​helda​.helsinki​.fi/​bitstream/​handle/​10138/​230989/​Koskenniemi​.pdf​?sequence​=​1​&​isAllowed​=​ y. Kulovesi, K. and Gutíerrez, M., (2009). Climate Change Negotiations Update: Process and Prospects for a Copenhagen Agreed Outcome in December 2009. Review of International, European and Comparative Environmental Law. 18(3), 229–243. Maljean Dubois, S. and Wemaere M., (2012). After Durban, What Legal Form for the Future Climate Regime. Carbon and Climate Law Review. 6(3), 187–196. Mayer, B., (2018). International Law Obligations Arising in relation to Nationally Determined Contributions. Transnational Environmental Law. 7(2), 251–275. McGee, J. and Steffek, J., (2016). The Copenhagen Turn in Global Climate Governance and the Contentious History of Differentiation in International Law. Journal of Environmental Law. 28(1), 37–63. Meilstrup, P., (2010). The Runaway Summit: The Background Story of the Danish Presidency of COP 1, the UN Climate Conference. Danish Foreign Policy Yearbook. 2010, 113–135. Milieudefensie et al. v. Royal Dutch Shell plc, District Court the Hague, Judgement of 26 May 2021. Available from: http://​climatecasechart​.com/​non​-us​-case/​milieudefensie​-et​-al​-v​-royal​-dutch​-shell​ -plc/​. Morcel, R., (2012). Unconstructive Ambiguity in the Durban Climate Deal of COP 17 / CMP 7. Sustainable Development Law & Policy. 12(2), 6–11. Morgenstern, L., (2009). One, Two or One and a Half Protocols? Assessment of Suggested Options for the Legal Form of the Post-2012 Climate Regime. Carbon and Climate Law Review. 3(3), 235–247.

Fighting a hard battle with a soft weapon  335 Mpanu-Mpanu, T., (2021). The Staircase of Paris. In: H. Jepsen et al., eds. Negotiating the Paris Agreement: Insider Stories. Cambridge: Cambridge University Press. pp. 182–196. Oberthür, S. and Bodle, R., (2016). The Legal Form and Nature of the Paris Agreement. Climate Law. 6(1-2), 40–57. Osofsky, H. M., (2012). Climate Change Law and Sub-State Actors. In: C. M. Baillet, ed. Non-State Actors, Soft Law and Protective Regimes. From the Margins. Cambridge: Cambridge University Press. pp. 179–199. Pauwelyn, J., (2012). Is It International Law or Not, and Does It Even Matter? In: J. Pauwelyn et al., eds. Informal International Lawmaking. Oxford: Oxford University Press. pp. 193–251. Pauwelyn, J. and Andanowa, L., (2015). ‘Legally Binding Treaty or Not? The Wrong Question for Paris Climate Summit’, EJILTalk, 4 Dec. 2015. [Viewed 8 July 2023] Available from https://​www​.ejiltalk​ .org/​a​-legally​-binding​-treaty​-or​-not​-the​-wrong​-question​-for​-paris​-climate​-summit/​. Pew Center on Global Climate Change, (2009). Post-2012 Climate Policy. A Copenhagen Agreement [online]. Arlington: Pew Center on Global Climate Change. Available from: https://​www​.c2es​.org/​wp​ -content/​uploads/​2009/​11/​copenhagen​-climate​-agreement​.pdf. Rajamani, L., (2009). Addressing the Post-Kyoto Stress Disorder: Reflections on the Emerging Legal Architecture of the Climate Regime. International and Comparative Law Quarterly. 58(4), 803–834. Rajamani, L., (2010). The Making and Unmaking of the Copenhagen Accord. International and Comparative Law Quarterly. 59(3), 824–843. Rajamani, L., (2012). The Durban Platform for Enhanced Action and the Future of the Climate Regime. International and Comparative Law Quarterly. 61(2), 501–518. Rajamani, L., (2015). The Devilish Details: Key Legal Issues in the 2015 Climate Negotiations. Modern Law Review. 78(5), 826–853. Rajamani, L., (2016). The 2015 Paris Agreement: Interplay between Hard, Soft and Non-Obligations. Journal of Environmental Law. 28(2), 337–358. Rajamani, L. and Brunnée, J., (2017). The Legality of Downgrading Nationally Determined Contribution under the Paris Agreement: Lessons from the US Disengagement. Journal of Environmental Law. 29(3), 537–551. Rajamani, L. and Guérin, E., (2017). Central Concepts in the Paris Agreement and How They Evolved. in: D. Klein et al., eds. The Paris Agreement on Climate Change: Analysis and Commentary. New York: Oxford University Press. pp. 74-90. Shaffer, G. C. and Pollack, M. A., (2009). Hard vs. Soft Law: Alternatives, Complements and Antagonists in International Governance. Minnesota Law Review. 94(3), 712–798. Slaughter, A.-M., (2015). The Paris Approach to Global Governance. [Viewed 8 July 2023]. Available from https://​www​.project​-syndicate​.org/​commentary/​paris​-agreement​-model​-for​-global​-governance​ -by​-anne​-marie​-slaughter​-2015​-12. Spence, C., (2022). Barack Obama. The Missing Hero. The Copenhagen Climate Summit. In: F. Dodds and C. Spence, eds. Heroes of Environmental Diplomacy. London: Routledge. UNFCCC., (1995). The Berlin Mandate: Review of the Adequacy of Article 4, Paragraph 2(a) and (b), of the Convention, Including Proposals Related to a Protocol and Decisions on Follow-up. Decision 1/CP.1. UN Doc FCCC/CP/1995/7/Add.1, 6 June 1995. UNFCCC., (2010). The Copenhagen Accord. Decision 2/CP.15. UN Doc. FCCC/CP/2009/11/Add.1, 30 March 2010. UNFCCC., (2011). The Cancun Agreements. Decision 1/CP.16. UN Doc. FCCC/CP/2010/7/Add.1, 15 March 2011. UNFCCC., (2016). Adoption of the Paris Agreement. Decision 1/CP.21. UN Doc. FCCC/CP/2015/10/ Add.1, 29 January 2016. van Asselt, H. and Zelli, F., (2018). International Governance: Polycentric Governance by and beyond the UNFCCC. In: A. Jordan et al., eds. Governing Climate Change: Polycentricity in Action? Cambridge: Cambridge University Press. pp. 29–46. van Asselt, H. et al., (2022). The Evolving Architecture of Global Climate Law. In: G. van Calster et al., eds. Research Handbook on Climate Change Mitigation Law. Cheltenham: Edward Elgar. Vihma, Antto, (2012). Soft Law, Hard Law and the North-South Politics of Climate Change. Unigrafia, Helsinki.

336  Research handbook on soft law Voigt, C., (2017). Institutional Arrangements and Final Clauses (Articles 16-29). In: D. Klein et al., eds. The Paris Agreement on Climate Change: Analysis and Commentary. New York: Oxford University Press. pp. 352–374. Voigt, C. and Ferreira, F., (2016). Dynamic Differentiation: The Principles of CBDR-RC, Progression and Highest Possible Ambition in the Paris Agreement. Transnational Environmental Law. 5(2), 285–303. Weil, P., (1983). Towards Relative Normativity in International Law? American Journal of International Law. 77(3), 413–442. Werksman, J. and Herbertson, K. (2010). The Aftermath of Copenhagen: Does International Law have a Role to Play in a Global Response to Climate Change? Maryland Journal of International Law. 25, 109, [Viewed 8 July 2023]. Available from https://​digitalcommons​.law​.umaryland​.edu/​mjil/​vol25/​ iss1/​8/​. Winkler, H., (2017). Mitigation (Article 4). In: D. Klein et al., eds. The Paris Agreement on Climate Change: Analysis and Commentary. New York: Oxford University Press. pp. 144–165. The World Bank., (2021). Reference Guide to Climate Change Framework Legislation [online]. Washington, DC: World Bank. [Viewed 12 July 2022]. Available from: https://​openknowledge​ .worldbank​.org/​bitstream/​handle/​10986/​34972/​World​-Bank​-Reference​-Guide​-to​-Climate​-Change​ -Framework​-Legislation​.pdf​?sequence​=​6. Yamin, F. and Depledge, J., (2004). The International Climate Change Regime. A Guide to Rules, Institutions and Processes. Cambridge: Cambridge University Press.

22. Soft law in city regulation and governance Astrid Voorwinden and Sofia Ranchordás

1. INTRODUCTION This chapter discusses the role of soft law in urban law and governance. Cities have used soft law instruments for two key reasons: First, to assert their global position and their need to address at the local level urgent problems with a global dimension (e.g., climate change); second, to regulate the ubiquitous use of digital technology in smart cities, defined here as urban spaces that harness technological innovation in pursuit of certain goals (e.g., liveability, sustainability). This chapter focuses on these two dimensions and explains how they contribute to the growing importance of soft law in the urban context. During the COVID-19 pandemic, a mix of hard and soft law measures was employed both at national and local levels, often in an uncoordinated attempt to safeguard public health under conditions of great uncertainty (Cheng 2021, p. 7; Utrilla Fernández-Bermejo 2021, p. 111). Cities, as centres of human agglomeration, became crucial actors in the international effort to combat the pandemic and they quickly resorted to non-binding instruments. In particular, many local governments deployed soft law measures to guide public administrations, address specific sectors, and appeal to citizens’ individual behaviour (Eliantonio et al. 2021, p. 1). This was the case, for example, in Spain where both state and local authorities employed soft law as a tool to interpret and complement hard law (Utrilla Fernández-Bermejo 2021, p. 111). The rapid dissemination of the disease had both local and global dimensions which needed to be addressed simultaneously at several levels. It is in this context that cities assumed a key role in containing the pandemic. Yet, the pandemic is not the only context in which cities employ soft law. Traditional hard law instruments have proven insufficient to address the global-local nature of urban problems, including rapid population growth, housing scarcity, pollution, and the consequences of climate change (Shaffer and Pollack 2010, p. 724). Cities constantly use their positions to address global issues, for example by advancing human rights (Oomen and Baumgärtel 2018, p. 607) and sustainability (Perrels and Nijkamp 1994). Therefore, at the resemblance of other fields of law discussed in this volume, urban law and governance have turned to soft law, that is, rules of conduct which, in principle, have no legally binding force but which still have practical effects (Snyder 1994, p. 198; Senden 2004, p. 111). Examples of soft rules used in the urban context are guidelines, declarations, covenants, ethical codes, and standards that seek to shape individual, business, and public behaviour. The growing use of soft law mirrors a new regulation and governance model for urban spaces which seeks to address long-term challenges and modes of thinking (e.g., sustainability), uncertainties and unpredictable futures (e.g., long-term impact of the pandemic), the involvement of private parties in urban management, and the acceleration of technological change (Tognetti 2021). An example of this growing use of soft law can be found in the development of ‘transnational city networks’ that develop, adopt, and apply soft law instruments (Acuto 2013b). Cities employ soft law primarily to assert their position on the international stage. Such networks 337

338  Research handbook on soft law bring local interests and expertise together in order to influence normative developments at international level (Frug and Barron 2006; Swiney 2020; Durmus and Oomen 2021). Together, cities formulate shared practices, create non-binding standards, work to influence multilateral agendas, and mirror state-based international law through their language, structures, and agreements (Acuto 2016). Soft law is not only employed to assert the international role of cities, but also to address the challenges of technological innovation. Indeed, scholars have argued that soft law provides more flexible, open, adaptable, and fast pathways to regulate new technologies (Hagemann et al. 2018; Marchant et al. 2020, p.7). Smart cities, in particular, utilize soft law instruments to regulate different facets of digitalization, from the regulation of digital platforms (e.g., Airbnb for home-sharing, Finck and Ranchordás 2016, p. 1299) to technical standards for privacy and security of smart technologies (e.g., privacy standards for RFID chips, Spiekermann 2012). Digitalization of urban spaces raises technical and ethical issues that require proactive regulatory responses. At the same time, the public-private nature of smart city projects means that technology companies have an important role in drafting the frameworks of urban governance. In this regard, soft law can provide a venue for multi-stakeholder engagement and participation to regulatory efforts (Hagemann et al. 2018). This chapter offers an urban law perspective of soft law, which is rooted in the global and technological challenges cities face today. This chapter’s contribution to the literature is twofold: First, it discusses how cities use soft law to take space on the global stage (section 2); second, it presents the development of soft law instruments in the context of smart cities (section 3). Section 3 considers soft law instruments in smart cities following three key themes: digital rights, sharing economy, and sustainability. Section 4 reflects upon the key characteristics of soft law in urban governance, namely its transnational character and its engagement with multiple stakeholders. Section 5 concludes.

2.

SOFT LAW FOR INTERNATIONAL AND MULTI-STAKEHOLDER ENGAGEMENT

This section delves into the complex nature of cities as global-local entities (Sassen 1991). It explains how their multifaceted nature affects their regulation and governance and how soft law entered the urban realm. In doing so, this section presents urban law and governance as distinct enterprises from public law, public policy, and other related fields of study (Davidson 2013). Many cities still struggle with the limited nature of their powers, despite the increasing amplitude and complexity of the problems they face. Traditional binding frameworks and instruments provided by national and international law are often focused on specific issues. This limited approach to law is insufficient to cities that struggle daily with fast-changing challenges that span across time, geography, and space. Soft law has offered a way to cities to extend their powers and engage with their growing problems and shrinking budgets. 2.1

Cities in Law and Governance

The regulation and governance of cities has been challenging for decades, starting with the definition and characterization of cities as such. Indeed, a number of paradoxes agitate the nature of the cities. First, cities are spaces of proximity and uncertainty. In urban spaces,

Soft law in city regulation and governance  339 people are constantly confronted with the unknown, which requires that they form strategies and relationships to address unpredictability (Keymolen and Voorwinden 2020). Urban heterogeneity forms a paradoxical but fundamental property of cities that drives urban settlements forward (Bairoch 1988). In the words of Edward Glaeser, cities can be defined by ‘the absence of physical space between people and companies. They are proximity, density, closeness’ (Glaeser 2011, p. 6). They bring together human, economic, and social capital, which then allows for the sharing of infrastructures, services, and interests. Yet, cities are spaces of distance as well, and to some extent, until the introduction of digital technologies, they were spaces of anonymity. Urban spaces are also spaces of transition, for their residents but also for themselves, as they exist in between different realities. Cities reveal themselves as ‘ecumenical phenomena that escape the dimensionality of geography’ (Philippopoulos-Mihalopoulos 2007, p. 1). They operate at local level, yet they are essential for the realization of national and international goals. Indeed, across countries, cities share enough common features and problems despite their differences that they are arguably closer to each other in their organization than states (Algaze 2018; Smith and Lobo 2019). Finally, although divided in most cases in municipal entities and governed by public local authorities, they are highly dependent on private actors for their economic, social, and cultural activities. Thus, cities are complex entities, with public-private elements, local, national and international dimensions. In part due to the paradoxical nature of the city, urban regulation and governance appear invisible to many fields of law (Nicola 2012, p. 1288). This was not always the case, since cities (or city-states) precede many nation-states and were historically well studied (Algaze 2008; Smith and Lobo 2019). Half a century ago, urban law and in particular urban planning were regarded as important fields of legal research, but the amount of scholarly legal work on cities has decreased rapidly in the last three decades (Davidson 2013). Constitutional and administrative law focused primarily to national, European, and international phenomena (Shaw and Štiks 2021). Nevertheless, other sciences such as urban studies, economics, and demography turned to the urban space as a source of inspiration for vibrant research. More recently, a small but growing number of legal scholars has been devoting more attention to urban law and to the (self)constitutionalization of cities as spaces of sovereignty and citizenship (Shaw and Štiks 2021). As cities and their metropolitan regions occupy the centre stage of the most crucial contemporary challenges (Schleicher 2010, 2013), their governance becomes an important topic for legal scholarship to analyze. 2.2

The Global Dimension of Urban Governance

The local character of cities is straightforward: Cities exist at local level, with place-bound infrastructure, services, and policies that interact directly with residents, visitors, and commuters. However, the phenomenon of the ‘global city’, coined by Saskia Sassen (1991), has gained attention in the last decades (Blank 2006, p. 266). By enabling and reshaping worldwide communications and trade, globalization has reshaped cities as ‘strategic territories’ of our economic system (Sassen 2005, p. 27). Their ability to attract people, capital, and goods, but also the trust and legitimacy conferred by their population (Swiney 2020, p. 269) has endowed them with important soft power. Although this soft power consists primarily of voluntary, self-enforced commitments, they have practical effect on a global scale. As Swiney observes, ‘[c]ities operate globally, signing and implementing documents that look and act like international law, forming powerful partnerships, and influencing the actions and policy positions

340  Research handbook on soft law of other international actors, including nation-states’ (Swiney 2020, p. 269). Their economic power has encouraged cities to participate in world affairs with ‘increased self-confidence’ (Oomen and Baumgärtel 2018, p. 608), even when their legal status within international law remains contested (Blank 2021). This section briefly explains cities’ main channels of (soft) influence at international level, without delving specifically into the role of soft law in international law (see d’Aspremont 2008, p. 1075). Legal scholars have identified different channels of influence, networks, and instruments embraced by local governments to act beyond the local scope (Frug and Barron 2006; Acuto 2013a). Many of these strategies can be qualified as soft law. Examples are inter-city alliances through transnational city networks (TCNs), discussed further in the following sections, and the creation of international charters for local self-government (Frug and Barron 2006). TCNs are particularly interesting as they precede the globalization movement, dating back to the 13th century in the High Middle Ages (Kern and Bulkeley 2009). The Hanseatic League is a well-known example of this type of networks. TCNs facilitate policy diffusion between cities (‘city-to-city diplomacy’) and catalyse policy influence of cities in international relations and policies (Acuto 2013a, p. 486). Furthermore, Acuto (2013a) identified five types of agency and participation for cities: (1) the design of policy frameworks (regime building); (2) partnering with transnational actors and the private sector (hybridization); (3) undertaking foreign policy, including through para-diplomatic initiatives (diplomatic entrepreneurship); (4) fostering international normative frameworks (normative mediation); and (5) the use of local, everyday management capacities to impact international affairs. An example of cities’ global influence is the so-called ‘Human Rights Cities’ (e.g., Vienna, Utrecht, Barcelona) and ‘sanctuary cities’ (e.g., Tampa Bay, Florida). These city networks and the soft law agreements upon which they are based help cities respond to global developments, including welcoming and integrating displaced refugees and immigrants (Oomen and Baumgärtel 2018). Human Rights Cities refers to ‘cities or communities where people of good will, in government, in organizations and in institutions, try and let a human rights framework guide the development of the life of the community’ (PDHRE 2007). ‘Sanctuary cities’ are cities or counties that refuse to cooperate with federal or national immigration authorities in order to protect low-priority immigrants from deportation. Sanctuary cities report immigrants that have allegedly committed serious crimes but offer a safe harbour to those who do not pose a risk to their communities. Barcelona, for example, has expressed since the 1990s its commitment to protect human rights as part of municipal policy (Grigolo 2019). The city has a Civil Rights Department, and it ratified the European Charter for the Safeguarding of Human Rights (Oomen and Baumgärtel 2018). Many of the cities that have joined the network of Human Rights Cities, particularly during the Syrian refugee crisis in 2015, have expressed hope that ‘cities [will] deliver where nation states have failed’ (Oomen 2016, p. 2). Human rights cities rely on these networks to position themselves as centres for human development, social justice, and policy innovation. In an unprecedented manner, cities have become important hubs for human rights activity and, through soft law agreements, drive the implementation and monitoring policies dealing with human rights and social justice (Reuter 2019, p. 382). The advancement of human rights in urban space has become crucial due not only to globalization but also due to the growing use of digital technology in so-called ‘smart cities’. As the next section will show, while smart cities and their technological capabilities have the

Soft law in city regulation and governance  341 potential to improve the quality of living of city residents, they also risk deepening existing socioeconomic vulnerabilities and inequalities for which rapid legal responses are needed.

3.

SOFT LAW IN THE SMART CITY

This section defines the smart city movement and three of its central themes: technology, sustainability, and economy. These three themes raise issues for human rights and for the public interest, which has led to the development of soft law instruments. Soft law has been argued to be a well-suited tool to guide technological innovation where more traditional forms of regulation show their limits (Hagemann et al. 2018). Following this reasoning, soft law instruments can react in an open and flexible way to a fast-paced change that blurs traditional lines and sectors, which fits the context of smart cities well. Therefore, this section examines how cities engage with drawing soft normative frameworks for digital rights, sustainability, and the sharing economy. 3.1

The Smart City: Definition, Goals and Policies

Over the past 15 years, local governments throughout the world have adopted smart city programs (Manville et al. 2014; Datta 2018). It should be noted that the term ‘smart city’ remains contested and multifaceted, counting a multitude of different definitions and dimensions (Nam and Pardo 2011; Chourabi et al. 2012; Cocchia 2014; Albino et al. 2015; Mora et al. 2017). Essentially, the smart city harnesses technological innovation to transform urban environments in pursuit of certain goals, linked to contemporary urban challenges. For instance, smart cities implement facial recognition cameras to increase police capacity and accuracy of law enforcement. Citizens participate in smart grids in order to share energy resources. Air pollution sensors could be placed throughout the city to control pollution. Thus, the smart city combines an infrastructural component with a set of desired outcomes, which are often sustainability and the improvement of the quality of life of residents (Ramaprasad et al. 2017). These goals are reflected in the ideological dimension of the notion of ‘smartness’, which is an elusive term that local governments adopt in their policies to convey different targets such as ‘sustainable development, economic growth, better quality of life for their citizens, and creating happiness’ (Albino et al. 2015, p. 5). In this section, we focus on three recurring themes in smart cities: technology, sustainability, and economic development. Indeed, the smart city relies on ICT and data to gather knowledge about urban environments and required interventions. This occurs on an infrastructural layer, with the implementation of networks of sensors, and on a data layer, with the use of big data analytics. These technologies can be employed in pursuit of sustainability through applications such as renewable energy sources, smart grids, positive energy districts, recycling and waste management techniques, and shared mobility. In doing so, the functional link between the smart city and sustainability sometimes appears in environmental policies at local level and in policy documents at European level (Ferrara 2015). This has resulted in the promotion of so-called ‘smart sustainable cities’ (Huovila et al. 2019). Finally, the smart city has an economic component. The label ‘smart’ serves as a marketing tool for cities to attract capital and foster the local economy in a context of competitive urban development (Giffinger et al. 2010; Shelton et al. 2015; Buck and While 2017). Digitalization of urban spaces has also involved

342  Research handbook on soft law new players in urban governance, such as technology companies that engage with the sharing economy. On all these three issues, as will be explored in this section, soft law instruments are being deployed. In pursuing its goals, the smart city raises a number of issues. The deployment of ubiquitous sensors and big data analytics in urban governance risks breaching people’s privacy, increasing discrimination, widening the digital gap, hampering personal autonomy, weakening democratic control, and outsourcing local public services. Urban areas are high stake stages for the future of digital rights, since they provide spaces and services people cannot opt out of. Citizens, unlike consumers, only have one provider of urban services (Ranchordás 2018). Additionally, technology can have a disproportionate effect on marginalized groups by reinforcing structural discrimination, systematic privacy violations, and power imbalances (Edwards 2016; Reuter 2019). The economic, social, and ecological costs of the uneven distribution of technological risks across the city, as socio-economic disparities, continue to worsen in the smart city. It is therefore not surprising that the past decade has seen a growing normative impulse in the smart city movement, facing the realization of the smart city’s risks (Bannerman and Orasch 2020, p. 17). Calls for resistance and for ‘citizen-centric’ smart cities that confront the legal and ethical risks posed by smart technologies have multiplied (Mann et al. 2020). The most well-known example is the city of Toronto, where the ambitious plans of Google’s parent company Alphabet were met by strong local resistance that led to its retraction in 2020 (Mann et al. 2020). A multitude of frameworks for the smart city have thus been emerging, many of which are composed of soft law tools. 3.2

Soft Law and Digital Rights

In April 2021, transnational city network Eurocities’ chair expressed support for a ban on biometric recognition systems (Dragonetti 2021). In June of the same year, the cities of Barcelona, London and Amsterdam launched the Global Observatory on AI to ‘help cities deploy artificial intelligence effectively and ethically’ (CC4DR 2021). The Observatory responded to the growing deployment of AI in urban spaces (e.g., facial recognition), although municipalities have scattered information on ethical frameworks and insufficient resources for in-depth studies. This initiative aims to share practices among cities, monitor AI policy, and formulate guidance standards. At the time of writing, the Observatory is working on a report on the ethical governance of algorithms in the cities as well as on an atlas of best practices in urban AI. This initiative exemplifies the rise of soft law instruments to tackle digital rights in the smart cities. Over the past years, cities have participated in the creation of multiple charters, manifestos, declarations, guidelines, coalitions, and other observatories on digital rights in urban spaces. This section considers two characteristics of soft law instruments in this context: they have a strongly normative character, and they claim a central role for cities in promulgating these norms based on democratic legitimation. Regarding their normative character, soft law instruments on digital rights tend to formulate ‘values’ and ‘principles’ that build on but move beyond legally binding norms. For instance, the Declaration of Cities Coalition for Digital Rights (CC4DR 2018, p. 1) recognizes privacy, data protection, and security as ‘human rights principles’ but it seeks to advance them by adding participatory democracy, digital literacy, and diversity. The Dutch Principles for the Digital City (VNG 2018) recognize the structural role of legislation on privacy and security, adding

Soft law in city regulation and governance  343 other notions such as accessibility, availability, and interoperability. The German Smart City Charter (BBSR 2017) edicts a number of ‘required’ values, such as co-creation, integration, and inclusion. The draft of the London Technology Charter summarizes its aim as ‘living the spirit, not just the letter of the GDPR’ (Mayor of London 2021). In doing so, these documents refer to both technical aspects of the smart city (e.g., open data, open standards, privacy by design) and ethical, social, political aspects (e.g., digital divide, right to communicate with public administration through non-digital ways). The coupling of ethics and technology can be found across all these documents, as with the CC4DR Coalition notion of ‘ethical digital services standards’ (CC4DR 2018, p. 2), or the London Charter’s reference to ‘human rights in design of technologies’ (Mayor of London 2021). The German Charter explicitly states the goal to ‘co-determine the normative image’ (BBSR 2017, p. 25) of the smart city and recognizes ‘the emergence of new power structures around data and automation that can threated democratic control’ (BBSR 2017, p. 28). These charters thus acknowledge the political nature of the smart city and formulate a ‘right to the city’ in the digital era, which the CC4DR Declaration formulates as the ‘ability collectively to engage with the city … to participate in shaping local digital infrastructure and services and, more generally, city policy-making for the common good’ (CC4DR 2018, p. 2). These few examples reflect an emerging reaction to the normative concerns raised by the smart city, in an effort to protect democratic governance and sovereignty (Goodman 2020). Furthermore, city charters place local governments in a central role to protect the rights of citizens in the context of digitalization of the city. The CC4DR was launched by 50 cities that aimed at ‘helping each other in the greenfield of digital-rights based policy-making’ in order to ‘work towards legal, ethical and operational frameworks to advance human rights in digital environments’ (CC4DR 2021). The coalition argues that cities, as the closest institutions to the citizenry, are potential ‘agile testbeds for new policies and rules’ which can use their legislative powers, governance competences, and purchasing power to actively promote digital rights. The coalition places its value in the ability to ‘look at digital rights from both the local and global perspectives’ (CC4DR 2021), uniting the dimensions discussed in section 2. This discourse can be found again in the German Charter, which was conceived to ensure that municipalities remain active actors in the digital transformation process. This active role requires cities to invest resources, develop local skills, establish local bylaws on freedom of information and transparency, and also engage with private firms which provide technological services. 3.3

Soft Law and the Sharing Economy

At first sight, one may wonder why the sharing economy (or collaborative economy) is mentioned in a book on soft law and, more specifically, in a chapter on urban law and governance and smart cities. First, the sharing-economy transactions are more common in cities than in rural areas (Finck and Ranchordás 2016) and since they typically occur beyond the law, they challenge locally regulated activities which are required to comply with extensive licensing and permits (Hatzopoulos 2018). This is the case of hotels, local transportation, and restaurants. The sharing economy, in its original sense of sharing underused space and goods with community members, took place mostly through informal relationships and remained thus largely unregulated until the last decade (Ranchordás 2015). Since 2008, the sharing economy is no longer limited to exchanges between acquaintances. Instead, it has become

344  Research handbook on soft law a multi-million-euro business model powered by digital platforms, some of them with hundreds of thousands if not millions of users worldwide (e.g., Airbnb). Second, the recent development of the sharing economy and its regulatory problems are intrinsically connected to the use of digital technology and its harnessing by smart cities (Ranchordás and Goanta 2020). The sharing economy allows smart cities to better use their infrastructures and the idle capacity of different goods and services (e.g., cars and housing). Third, soft law has been employed both at EU and local levels because it remains complicated (if not sometimes impossible) to enforce hard law due to the complex triangular relationship between platforms, platform users, and cities (Leshinsky and Schatz 2018). Sharing-economy platforms operated in the beginning of their existence either beyond the law, ignoring or setting aside hard law frameworks or celebrating soft law agreements with public authorities to avoid the enforcement of hard regulations. This happened for a variety of reasons. First, sharing-economy platforms, particularly in the home and ride-sharing sectors did not fit exactly in existing legal categories. Sporadic home-sharing on Airbnb was not entirely comparable to exploiting a hotel but it was not a ‘good Samaritan’ favour either. Traditional licensed sectors (e.g., hotels) claimed that cities should be strict and enforce existing legislation against platforms and their users to maintain the level-playing field between hotels and platforms and protect consumers. However, cities had limited ability to locate illegal accommodation and enforce the law. At the same time, platforms such as Airbnb offered to collaborate with municipalities and promote joint efforts to protect public values (Finck and Ranchordás 2016). This was the case of Amsterdam which in the early days of the sharing economy, employed soft law as its regulatory approach to this phenomenon. In 2014 and then later in 2016, Amsterdam signed two memoranda of understanding with Airbnb, allowing this platform to establish itself. The idea of the first memorandum of understanding (which was initially confidential) was to allow the platform to operate beyond existing national and local law. Even though Airbnb challenged housing laws that the municipality of Amsterdam had to enforce, the city would facilitate the growth of the platform by not implementing the laws under a number of conditions (e.g., home-sharing should be limited to a set number of days, the host should also be registered as resident). This soft law approach was not extended beyond the first two memoranda of understanding (Gemeente Amsterdam 2014, 2016) as it was insufficient to address the discontent of Amsterdam residents. In January 2020, the Dutch Council of State invalidated Amsterdam’s local policy not to fully enforce national rules on home-sharing (Raad van State 2020). Since July 2021, the new national law on tourist accommodation and home-sharing imposes a number of novel requirements on digital platforms (Staatsblad 2020). More recently, the city of Amsterdam has celebrated mostly covenants and other types of soft law agreements with parties that invest in the promotion of sharing-economy activities that advance the circular economy and social cohesion (Wennekes 2017). At EU level, the sharing economy remains indirectly regulated through different directives (e.g., Services Directive, E-Commerce) and one key soft law instrument. In 2016, the European Commission decided not to advance with any hard law instruments and adopted the Communication ‘A European Agenda for the collaborative economy’. The European Commission (2015, 2016) highlighted in this document that the sharing economy business model should be incentivized as it could contribute to the functioning of the single market and the Digital Single Market Strategy. The adoption of soft law can be explained by the need to confer flexibility to a complex, uncertain, and fast-changing field. The European

Soft law in city regulation and governance  345 Commission produced this Communication partially as a response to the many complaints of regulated sectors and the discontent regarding the regulatory disparities across Europe in the home-sharing sector. The existence of these soft law instruments has nonetheless not improved the position of cities that continue to grapple with the complexity of the sharing economy and its effect on housing prices and neighbourhood safety. Nowadays, the regulation of the sharing economy is populated by hard rules on zoning, including licences and permits, fire safety, and tourism regulation. These rules are important because digital platforms such as Airbnb and Blablacar, affect cities in different ways: On the one hand, the sharing economy has provided a boost to tourism during the financial crisis; on the other, it caused disruption due its effects on affordable housing, the large number of tourists, nuisance complaints, the safety of neighbourhoods, and the liveability of city centres. 3.4

Soft Law and Sustainability

The third area of soft law creation in smart cities is sustainability. Cities are the world’s largest consumers of energy and emitters of greenhouse gases, but they are also very vulnerable to environmental disasters, for which technological innovation is argued to provide mitigation (Bibri 2019). As such cities have been identified as a ‘privileged international locus’ for achieving sustainable development (Porras 2009, p. 538), leading to their increasing and explicit recognition in UN frameworks (Kosovac et al. 2020) and their inclusion in the Sustainable Development Goals (SDG 11). Within smart cities, sustainability is increasingly recognized as a central theme, although the inclusion of environmental aspects remains insufficiently evaluated and prioritized (Ahvenniemi et al. 2017; Fernández and Peek 2020). Two soft law tools mark this growing recognition: the creation of transnational smart city networks focused on climate change and the inclusion of sustainability in smart city standards. Indeed, environmental issues have led to setting up a multitude of transnational city networks over the past decades, such as the Covenant of Mayors for Climate, Energy and the Climate Leadership Group (C40), and Energy Cities. These networks ‘circumvent state-centric hierarchies’ by pooling resources, best practices, and planning instruments across a global network (Acuto 2013b, p. 96). They do not force any compliance onto their members, but instead focus on sharing possibilities of local action (Aust 2015) and advocating in international institutions (Stürner and Bendel 2019). In the field of smart cities, many networks have been set up over the past decade. A recent example is the Global Smart Cities Alliance on Technology Governance, launched in October 2019, with a goal to develop global policy standards, best practices and ‘new global governance guidelines for the responsible use of data and digital technologies in urban environments’ (World Economic Forum 2019). This Alliance regroups more than 200,000 cities and local governments, companies, start-ups, research institutions, and civil society communities, with the World Economic Forum serving as secretariat. The link between smart cities and sustainability begins to appear in networks that were initially focused on either climate change or on technology. Indeed, initiatives focused engaging mayors in climate action, such as the Forum of Mayors, have started referring to smart city technologies, which the Forum calls ‘an important step in the development of sustainable and resilient cities’ (Forum of Mayors 2020, p. 3). New international associations concentrate on the link between smart cities and sustainability, such as the World Smart Sustainable Cities Organizations (WeGO), founded by 50 cities in 2010. The UN and the International Telecommunication Union (ITU) started United

346  Research handbook on soft law for Smart Sustainable Cities (U4SSC), with a strong technical component. This latest initiative illustrates both the rise of the smart sustainable city at international level and the importance of another actor in soft law: standardization bodies. Standards and performance indicators have been developed as tools to define and monitor ‘smart sustainable cities’. Indeed, on top of hundreds of technical standards on interoperability, data protection, and cybersecurity (Lai et al. 2020), standardization bodies have recently published a multitude of indicators and standards that tie sustainability and smartness together. The ITU, the International Organization for Standardization (ISO), and the American National Standards Institute (ANSI) have each created dedicated study groups and committees on smart and sustainable cities (ANSI 2014; ISO 2018, 2019; ITU 2021a, 2021b). Among others, the aforementioned U4SSC has developed 91 Key Performance Indicators (KPIs) to measure smart city’s sustainability, based on the recommendations issued by the ITU (ITU 2016). The influence of these standards is reflected by the certifications issued by the World Council on City Data based on compliance with the indicators: in 2019, 52 cities were certified following the ISO standards (Huovila et al. 2019) and hundreds of cities have piloted the U4SSC KPIs (ITU 2021c). Here, soft law is employed as a tool to create a common language, a reliable evaluation framework, and measurable targets (Huovila et al. 2019). Although standardization does not have legally binding force, legitimacy arises from the expertise and consensus-based dynamics of the stakeholder community that creates them (Brunsson et al. 2012). The impact of the adoption of these standards on the ground, however, remains to be determined (Huovila et al. 2019).

4.

NEW STAKEHOLDERS IN URBAN SOFT LAW

Following the observations of the previous sections, soft law appears as a useful tool for urban governance in two ways: It fosters international collaboration, and it offers a channel of engagement with non-public stakeholders. Indeed, trans-municipal networks are crucial drivers of urban soft law creation. These networks participate to a repositioning of cities at international level. They serve as a platform for local leaders and policymakers to work together on shared challenges and to diffuse policy between local governments directly (Foster and Swiney 2021), as in the case with the CC4DR and the AI Observatory. By embracing the local and global dimension of the city, these networks exercise their norm-generating capacity, as described by Durmus and Oomen (2021) through different functions: influencing the global agenda (e.g., climate change), regulating the behaviour of members (e.g., adoption of standards), crystalizing norms between local governments (e.g., declaration of digital rights), and integrating separate hard law provisions (e.g., principled approach to data protection). As these networks produce different types of soft law instruments, an interrogation for the future remains on how binding and how impactful different forms of agreements are (Nijman 2011). Furthermore, engagement with different stakeholders is inherent to smart cities since the smart city movement depends on the active participation of private partners such as technology companies. This contributes to the hybridization of governance (Acuto 2013a), with local governments increasingly relying on public-private governance structures not only for the delivery of services, but also for para-diplomatic initiatives (e.g., industry presence in smart city networks). This is for instance visible within standardization efforts.

Soft law in city regulation and governance  347 Standardization enables a form of technical regulation (de Filippi 2016) since standards and norms incorporate legal and technical characteristics and principles in technologies. For instance, standards can incorporate privacy by design, which is a principle recognized within data protection legislation. The creation of standards is a highly political process. As identified by Merricks White (2018), standards are a political technology for the dissemination of ideas, steering outcomes, and assuring city leaders and decision-makers. In the context of smart cities, these standards establish management frameworks and guidance to leadership (Merricks White 2018). They are not only used on a technical level to enable interoperability of infrastructure, but also on a strategic level and a process level to set goals for smart city projects, measure their performance, and guide decision-making (BSI 2014). In doing so, smart city standards emerge from hybrid (or private) transnational actors whose decision-making processes, aims, and values differ from states and traditional public institutions. The dependence of states on private or hybrid standardization bodies to regulate new technologies is a well-known phenomenon, resulting in gaps of political and democratic legitimacy (Fosch Villaronga and Golia 2019). At the same time, as few hard law frameworks on the smart city have been developed, soft law can be a source of understanding and information for future regulatory efforts (Fosch Villaronga and Golia 2019).

5. CONCLUSION Cities are paradoxical entities, both the subject and object of multilevel regulation. They are local actors in the frontline of global problems, and they implement a layered field of hard law while hardly being recognized as lawmakers themselves. The strength of cities, as explored in this chapter, resides in their soft power and their ability to leverage it, especially in the context of new technologies deployed in smart cities. This chapter has shown that the development of soft law instruments in urban law and governance has two main features. First, it advances the transnational character of cities and expands their powers beyond the realm of hard law, by creating para-diplomatic networks that engage in policy exchange. Second, it allows cities to engage productively with multiple stakeholders about the digital technology and services that transform cities today. Using instruments such as TCNs, human rights cities, technical standards, digital rights charters, and covenants with private actors, city governments grapple with the challenges of urban innovation and transformation. Contrary to hard law, soft law permits a more proactive approach to rulemaking which is essential to address the problems generated by rapidly changing technologies. It helps local entities assert themselves internationally, consolidate their position as global-local actors, and regulate uncertain and complex technologies. However, the capacity of soft law to sufficiently address the impact of smart city technologies on citizens and their rights remains to be observed. The interaction between soft law instruments and hard law frameworks at local level will need to find a balance between expertise, flexibility, and democratic legitimacy. In navigating this tension, the governance of the smart city tentatively sets the stage for our urban future.

348  Research handbook on soft law

REFERENCES Acuto, M., (2013a). City leadership in global governance. Global Governance. 19(3), 481–498. Acuto, M., (2013b). Global cities, governance and diplomacy: the urban link. London: Routledge. Acuto, M., (2016). City diplomacy. In: C. M. Constantinou, P. Kerr and P. Sharp, eds. The SAGE handbook of diplomacy. London: SAGE Publications. pp. 510–520. Ahvenniemi, H., Huovila, A., Pinto-Seppä, I. and Airaksinen, M., (2017). What are the differences between sustainable and smart cities?. Cities. 60, 234–245. Albino, V., Berardi, U. and Dangelico, R. M., (2015). Smart cities: definitions, dimensions, performance, and initiatives. Journal of Urban Technology. 22(1), 3–21. Algaze, G., (2008). Ancient Mesopotamia at the dawn of civilization: the evolution of an urban landscape. Chicago: University of Chicago Press. Algaze, G., (2018). Entropic cities: the paradox of urbanism in ancient Mesopotamia. Current Anthropology. 59(1), 23–54. ANSI., (2014). ANSI establishes network on smart and sustainable cities [online]. The American National Standards Institute (ANSI). [Viewed 3 December 2021]. Available from: https://​www​.ansi​.org/​news/​ standards​-news/​all​-news/​2014/​05/​ansi​-establishes​-network​-on​-smart​-and​-sustainable​-cities​-27. D’Aspremont, J., (2008). Softness in international law: a self-serving quest for new legal materials. European Journal of International Law. 19(5), 1075–1093. Aust, H. P., (2015). Shining cities on the hill? The global city, climate change, and international law. European Journal of International Law. 26(1), 255–278. Bairoch, P., (1988). Cities and economic development: from the dawn of history to the present. Chicago: University of Chicago Press. Bannerman, S. and Orasch, A., (2020). Privacy and smart cities: a Canadian survey. Canadian Journal of Urban Research. 29(1), 17–38. BBSR., (2017). Smart City Charter: making digital transformation at the local level sustainable [online]. Bonn: Bundesinstitut für Bau-, Stadt- und Raumforschung (BBSR). [Viewed 31 July 2021]. Available from: https://​www​.bmi​.bund​.de/​SharedDocs/​downloads/​DE/​veroeffentlichungen/​themen/​bauen/​ wohnen/​smart​-city​-charta​-kurzfassung​-de​-und​-en​.pdf​?​_​_blob​=​publicationFile​&​v​=​4. Bibri, S. E., (2019). On the sustainability of smart and smarter cities in the era of big data: an interdisciplinary and transdisciplinary literature review. Journal of Big Data. 6(1), 1–64. Blank, Y., (2006). The city and the world. Columbia Journal of Transnational Law. 44(3), 875–939. Blank, Y., (2021). International legal personality/subjectivity of cities. In: H. P. Aust and J. E. Nijman, eds. Research handbook on international law and cities. Cheltenham, UK: Edward Elgar. pp. 103–120. Brunsson, N., Rasche, A. and Seidl, D., (2012). The dynamics of standardization: three perspectives on standards in organization studies. Organization Studies [online]. 33(5-6), 613–632. [Viewed 1 December 2021]. Available from: doi: 10.1177/0170840612450120. BSI., (2014). The role of standards in smart cities [online]. London: The British Standards Institution (BSI). [Viewed 6 December 2021]. Available from: https://​www​.bsigroup​.com/​LocalFiles/​en​-GB/​ smart​-cities/​resources/​The​-Role​-of​-Standards​-in​-Smart​-Cities​-Issue​-2​-August​-2014​.pdf. Buck, N. T. and While, A., (2017). Competitive urbanism and the limits to smart city innovation: The UK Future Cities initiative. Urban Studies. 54(2), 501–519. CC4DR., (2018). Declaration of Cities Coalition for Digital Rights [online]. Cities Coalition for Digital Rights (CC4DR). [Viewed 3 December 2021]. Available from: https://​ci​tiesfordig​italrights​.org/​assets/​ Declaration​_Cities​_for​_Digital​_Rights​.pdf. CC4DR., (2021). Presentation of the global observatory on urban AI [online]. Cities Coalition for Digital Rights (CC4DR). [Viewed 31 July 2021]. Available from: https://​ci​tiesfordig​italrights​.org/​ presentation​-global​-observatory​-urban​-ai. Cheng, X., (2021). Soft law in the prevention and control of the COVID-19 pandemic in China: between legality concerns and limited participatory possibilities. European Journal of Risk Regulation [online]. 12(1), 7–25. [Viewed 1 December 2021]. Available from: doi:10.1017/err.2020.111. Chourabi, H., Nam, T., Walker, S., Gil-Garcia, J. R., Mellouli, S., Nahon, K., Pardo, T. A. and Scholl, H. J., (2012). Understanding smart cities: an integrative framework. 45th Hawaii International Conference on System Sciences [online]. 2289–2297. [Viewed 1 December 2021]. Available from: doi: 10.1109/HICSS.2012.615.

Soft law in city regulation and governance  349 Cocchia, A., (2014). Smart and digital city: a systematic literature review. In: R. P. Dameri and C. Rosenthal-Sabroux, eds. Smart city: how to create public and economic value with high technology in urban space. Cham: Springer. pp. 13–43. Datta, A., (2018). The digital turn in postcolonial urbanism: smart citizenship in the making of India's 100 smart cities. Transactions of the Institute of British Geographers. 43(3), 405–419. Davidson, N. M., (2013). What is urban law today? An introductory essay in honor of the fortieth anniversary of the Fordham Urban Law Journal. Fordham Urban Law Journal. 40(5), 1579–1593. De Filippi, P., (2016). The interplay between decentralization and privacy: the case of blockchain technologies. Journal of Peer Production [online]. 7. [Viewed 6 December 2021]. Available from: hal-01382006f . Dragonetti, W., (2021). First AI regulation: cities left with mixed feelings [online]. Eurocities. [Viewed 31 July 2021]. Available from: https://​eurocities​.eu/​latest/​first​-ai​-regulation​-for​-cities​-its​-yes​-and​-no. Durmus, E. and Oomen, B., (2021). Transnational city networks and their contributions to norm-generation in international law: the case of migration. Local Government Studies [online]. 1–22. [Viewed 1 December 2021]. Available from: doi: 10.1080/03003930.2021.1932478. Edwards, L., (2016). Privacy, security and data protection in smart cities: a critical EU law perspective. European Data Protection Law Review [online]. 2(1), 28–58. [Viewed 3 December 2021]. Available from: doi: 10.21552/EDPL/2016/1/6. Eliantonio, M., Korkea-aho, E. and Vaughan, S., (2021). EJRR special issue editorial: COVID-19 and soft law: is soft law pandemic-proof?. European Journal of Risk Regulation [online]. 12(1), 1–6. [Viewed 8 November 2021]. Available from: doi: 10.1017/err.2021.1. European Commission., (2015). Commission staff working document, a single market strategy for Europe - analysis and evidence (SWD(2015) 202 final) [online]. Brussels: European Commission. [Viewed 1 December 2021]. Available from: https://​ec​.europa​.eu/​docsroom/​documents/​14012/​ attachments/​1/​translations/​en/​renditions/​native. European Commission., (2016). Communication, a European agenda for the collaborative economy (COM/2016/356 final) [online]. Brussels: European Commission. [Viewed 1 December 2021]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​ALL/​?uri​=​COM:​2016:​356:​FIN. Fernández, C. G. and Peek, D., (2020). Smart and sustainable? Positioning adaptation to climate change in the European smart city. Smart Cities [online]. 3(2), 511–526. [Viewed 1 December 2021]. Available from: doi: 10.3390/smartcities3020027. Ferrara, R., (2015). The smart city and the green economy in Europe: a critical approach. Energies. 8(6), 4724–4734. Finck, M. and Ranchordás, S., (2016). Sharing and the city. Vanderbilt Journal of Transnational Law. 49(5), 1299–1369. Forum of Mayors., (2020). Geneva Declaration of Mayors. Times of change: reshaping our urban realities [online]. Geneva: United Nations Economic Commission for Europe (UNECE). [Viewed 3 December 2021]. Available from: https://​unece​.org/​sites/​default/​files/​2021​-03/​Mayors​%20declaration​ %20booklet​%20​-​%20ver​.4​.pdf. Fosch Villaronga, E. and Golia, A. Jr., (2019). Robots, standards and the law: rivalries between private standards and public policymaking for robot governance. Computer Law & Security Review. 35(2), 129–144. Foster, S. R. and Swiney, C., (2021). City networks and the glocalization of urban governance. In: H. P. Aust and J. E. Nijman, eds. Research handbook on international law and cities. Cheltenham, UK: Edward Elgar. pp. 368–380. Frug, G. E. and Barron, D. J., (2006). International local government law. Urban Lawyer. 38(1), 1–62. Gemeente Amsterdam., (2014). Memorandum of understanding agreed upon by the municipality of Amsterdam and the owner/administrator of the platform Airbnb [online]. [Viewed 15 September 2021]. Binnenlands Bestuur. Available from: https://​www​.binnenlandsbestuur​.nl/​Uploads/​2016/​2/​ 2014​-12​-airbnb​-ireland​-amsterdam​-mou​.pdf. Gemeente Amsterdam., (2016). Agreement regarding the touristic rental of homes and houseboats between the City of Amsterdam and the operator of the Airbnb platform [online]. [Viewed 15 September 2021]. Gemeente Amsterdam. Available from: https://​assets​.amsterdam​.nl/​publish/​pages/​ 813509/​agreement​_amsterdam​_and​_airbnb​_mou​.pdf.

350  Research handbook on soft law Giffinger, R. Haindlmaier, G. and Kramar, H., (2010). The role of rankings in growing city competition. Urban Research & Practice. 3(3), 299–312. Glaeser, E. L., (2011). The triumph of cities: how our greatest invention makes us richer, smarter, greener, healthier, and happier. New York, N.Y.: Penguin. Goodman, E. P., (2020). Smart city ethics: how “smart” challenges democratic governance. In: M. D. Dubber, F. Pasquale and S. Das, eds. The Oxford handbook of ethics and AI. New York, N.Y.: Oxford University Press. pp. 823–840. Grigolo, M., (2019). The human rights city: New York, San Francisco, Barcelona. London: Routledge. Hagemann, R., Huddleston Skees, J. and Thierer, A., (2018). Soft law for hard problems: the governance of emerging technologies in an uncertain future. Colorado Technology Law Journal. 17(1), 37–130. Hatzopoulos, V., (2018). The collaborative economy and EU law. Oxford: Hart Publishing. Huovila, A., Bosch, P. and Airaksinen, M., (2019). Comparative analysis of standardized indicators for Smart sustainable cities: what indicators and standards to use and when?. Cities. 89, 141–153. ISO., (2018). ISO 37120:2018 Sustainable cities and communities — indicators for city services and quality of life. Geneva: International Organization for Standardization (ISO). ISO., (2019). ISO 37122:2019 Sustainable cities and communities – indicators for smart cities. Geneva: International Organization for Standardization (ISO). ITU., (2016). Recommendation ITU-T Y.4903/L.1603. Key performance indicators for smart sustainable cities to assess the achievement of sustainable development goals [online]. [Viewed 1 December 2021]. Available from: https://​www​.itu​.int/​rec/​dologin​_pub​.asp​?lang​=​e​&​id​=​T​-REC​-Y​.4903​-201610​ -I!!PDF​-E​&​type​=​items. ITU., (2021a). United 4 Smart Sustainable Cities [online]. International Telecommunication Union (ITU). [Viewed 31 July 2021]. Available from: https://​www​.itu​.int/​en/​ITU​-T/​ssc/​united/​Pages/​default​ .aspx. ITU., (2021b). ITU's implementation of the U4SSC KPIs on smart sustainable cities [online]. International Telecommunication Union (ITU). [Viewed 31 July 2021]. Available from: https://​www​ .itu​.int/​en/​ITU​-T/​ssc/​Pages/​KPIs​-on​-SSC​.aspx. ITU., (2021c). ITU’s implementation of the U4SSC KPIs [online]. International Telecommunication Union (ITU). [Viewed 3 December 2021]. Available from: https://​www​.itu​.int/​en/​ITU​-T/​ssc/​united/​ Pages/​publication​-U4SSC​-KPIs​.aspx. Kern, K. and Bulkeley, H., (2009). Cities, Europeanization and multi-level governance: governing climate change through transnational municipal networks. Journal of Common Market Studies [online]. 47(2), 309–332. [Viewed 3 December 2021]. Available from: doi: 10.1111/j.1468-5965.2 009.00806.x. Keymolen, E. and Voorwinden, A., (2020). Can we negotiate? Trust and the rule of law in the smart city paradigm. International Review of Law, Computers & Technology. 34(3), 233–253. Kosovac, A., Acuto, M. and Jones, T. L., (2020). Acknowledging urbanization: a survey of the role of cities in UN frameworks. Global Policy. 11(3), 293–304. Lai C. S., Jia, Y., Dong, Z., Wang, D., Tao, Y., Lai, Q. H., Wong, R. T. K., Zobaa, A. F., Wu, R. and Lai, L. L., (2020). A review of technical standards for smart cities. Clean Technologies. 2(3), 290–310. Leshinsky, R. and Schatz, L., (2018). “I don’t think my landlord will find out:” Airbnb and the challenges of enforcement. Urban Policy and Research [online]. 36(4), 417–428. Available from: doi: 10.1080/ 08111146.2018.1429260. Mann, M., Mitchell, P., Foth, M. and Anastasiu, I., (2020). #BlockSidewalk to Barcelona: technological sovereignty and the social license to operate smart cities. Journal of the Association for Information Science and Technology. 71(9), 1103–1115. Manville, C., Cochrane, G., Cave, J., Millard, J., Pederson, J. K., Thaarup, R. K., Liebe, A., Wissner, M., Massink, R. and Kotterink, B., (2014). Mapping smart cities in the EU [online]. Brussels: European Parliament. [Viewed 27 July 2021]. Available from: https://​www​.europarl​.europa​.eu/​RegData/​etudes/​ etudes/​join/​2014/​507480/​IPOL​-ITRE​_ET​%282014​%29507480​_EN​.pdf. Marchant, G., Tournas, L. and Gutierrez, C. I., (2020). Governing emerging technologies through soft law: lessons for artificial intelligence. Jurimetrics. 61(1), 1–18. Mayor of London., (2021). An emerging technology charter for London [online]. Mayor of London. [Viewed 31 July 2021]. Available from: https://​www​.london​.gov​.uk/​publications/​emerging​ -technology​-charter​-london.

Soft law in city regulation and governance  351 Merricks White, J., (2018). Politicising smart city standards. In: C. Coletta, L. Evans, L. Heaphy and R. Kitchin, eds. Creating smart cities. London: Routledge. pp. 33–48. Mora, L., Bolici, R. and Deakin, M., (2017). The first two decades of smart-city research: a bibliometric analysis. Journal of Urban Technology. 24(1), 3–27. Nam, T. and Pardo, T. A., (2011). Conceptualizing smart city with dimensions of technology, people, and institutions. In: Proceedings of the 12th Annual International Digital Government Research Conference: Digital Government Innovation in Challenging Times. New York, N.Y.: Association for Computing Machinery. pp. 282–291. Nicola, F., (2012). Invisible cities in Europe. Fordham International Law Journal. 35(5), 1282–1363. Nijman, J., (2011). The future of the city and the international law of the future. In: S. Muller, S. Zouridis, M. Frishman and L. Kistemaker, eds. The law of the future and the future of law. Oslo: Torkel Opsahl Academic EPublisher. pp. 213–229. Oomen, B., (2016). Introduction: the promise and challenges of human rights cities. In: B. Oomen, M. F. Davis and M. Grigolo, eds. Global urban justice: the rise of human rights cities. Cambridge: Cambridge University Press. pp. 1–22. Oomen, B. and Baumgärtel, M., (2018). Frontier cities: the rise of local authorities as an opportunity for international human rights law. European Journal of International Law. 29(2), 607–630. PDHRE., (2007). Human rights learning and human rights cities. achievements report [online]. Los Angeles, C.A.: The People’s Movement for Human Rights Learning (PDHRE). [Viewed 2 December 2021]. Available from: https://​www​.pdhre​.org/​achievements​-HR​-cities​-mar​-07​.pdf. Perrels, A. and Nijkamp, P., (1994). Sustainable cities in Europe: a comparative analysis of urban energy–environmental policies. 1st ed. London: Routledge. Philippopoulos-Mihalopoulos, A., (2007). Introduction: in the lawscape. In: A. Philippopoulos-Mihalopoulos, ed. Law and the city. London: Routledge-Cavendish. pp. 1–20. Porras, I. M., (2009). The city and international law: in pursuit of sustainable development. Fordham Urban Law Journal. 36(3), 537–601. Raad van State., (2020). Uitspraak 201901695/1/A3 van de Raad van State, 29 January 2020 ECLI:NL:RVS:2020:261. Ramaprasad, A., Sánchez-Ortiz, A. and Syn, T., (2017). A unified definition of a smart city. In: M. Janssen et al., eds. 16th International Conference on Electronic Government (EGOV). Cham: Springer. pp. 13–24. Ranchordás, S., (2015). Does sharing mean caring? Regulating innovation in the sharing economy. Minnesota Journal of Law, Science & Technology. 16(1), 413–475. Ranchordás, S., (2018). Cities as corporations? The privatization of the cities and the automation of local law [online]. Oxford Business Law Blog. [Viewed 3 December 2021]. Available from: https://​www​.law​.ox​.ac​.uk/​business​-law​-blog/​ blog/2018/04/law-and-autonomous-systems-serie s-cities-corporations-privatization. Ranchordás, S. and Goanta, C., (2020). The new city regulators: platform and public values in smart and sharing cities. Computer Law and Security Review. 36, 1–15. Reuter, T. K., (2019). Human rights and the city: including marginalized communities in urban development and smart cities. Journal of Human Rights. 18(4), 382–402. Sassen, S. (1991). The global city: New York, London, Tokyo. Princeton, N.J.: Princeton University Press. Sassen, S., (2005). The global city: introducing a concept. The Brown Journal of World Affairs. 11(2), 27–43. Schleicher, D., (2010). The city as a law and economic subject. University of Illinois Law Review. 2010(5), 1507–1564. Schleicher, D., (2013). City unplanning. Yale Law Journal. 122(7), 1670–1737. Senden, L., (2004). Soft law in European Community law. Oxford: Hart Publishing. Shaffer, G. C. and Pollack, M. A., (2010). Hard vs. soft law: alternatives, complements and antagonists in international governance. Minnesota Law Review. 94(3), 706–799. Shaw, J. and Štiks, I., (2021). The constitutionalisation of cities and the future of global society. Identities [online]. 1–17. [Viewed 2 December 2021]. Available from: doi: 10.1080/1070289X.2021.1970981. Shelton, T., Zook, M. and Wiig, A., (2015). The ‘actually existing smart city’. Cambridge Journal of Regions, Economy and Society. 8(1), 13–25.

352  Research handbook on soft law Smith, M. E. and Lobo, J., (2019). Cities through the ages: one thing or many?. Frontiers in Digital Humanities [online]. 6(12). [Viewed 2 December 2021]. Available from: doi: 10.3389/ fdigh.2019.00012. Snyder, F., (1994). Soft law and institutional practice in the European Community. In: S. Martin, ed. The construction of Europe: essays in honour of Emile Noël. Dordrecht: Kluwer Academic Publishers. pp. 197–225. Spiekermann, S., (2012). The RFID PIA – developed by industry, endorsed by regulators. In: D. Wright and P. de Hert, eds. Privacy impact assessment. London: Springer. pp. 323–346. Staatsblad., (2020). Wet toeristische verhuur van woonruimte [online]. Staatsblad van het Koninkrijk der Nederlanden 2020, 460. [Viewed 15 September 2021]. Available from: https://​zoek​.off​icielebeke​ ndmakingen​.nl/​stb​-2020​-460​.html. Stürner, J. and Bendel, P., (2019). The two-way ‘glocalisation’ of human rights or: how cities become international agents in migration governance. Peace Human Rights Governance. 3(2), 215–240. Swiney, C., (2020). The urbanization of international law and international relations: the rising soft power of cities in global governance. Michigan Journal of International Law. 41(2), 227–278. Tognetti, R., (2021). For a soft law of contemporary project: food for thought and a manifesto. In: L. Crespi, ed. Design of the unfinished: a new way of designing leftovers regeneration. Cham: Springer. pp. 229–236. Utrilla Fernández-Bermejo, D., (2021). Soft law governance in times of coronavirus in Spain. European Journal of Risk Regulation. 12(1), 111–126. VNG., (2018). Principles for the digital city [online]. The Hague: Association of Netherlands Municipalities (Vereniging van Nederlandse Gemeenten, VNG). [Viewed 3 December 2021]. Available from: https://​vng​.nl/​files/​vng/​nieuws​_attachments/​2018/​principles​-for​-the​-digital​ -city​_20181025​.pdf. Wennekes, L. (2017). Steden zetten vol in op circulaire economie [online]. Gemeente.nu. [Viewed 18 September 2021]. Available from: https://​www​.gemeente​.nu/​ruimte​-milieu/​duurzaamheid/​gemeenten​ -zetten​-op​-circulaire​-steden. World Economic Forum., (2019). World Economic Forum to lead G20 Smart Cities Alliance on Technology Governance [online]. The World Economic Forum Blog. [Viewed 3 December 2021]. Available from: https://​www​.weforum​.org/​press/​2019/​06/​world​-economic​-forum​-to​-lead​-g20​-smart​ -cities​-alliance​-on​-technology​-governance.

23. Soft law and citizenship regimes Timothy Jacob-Owens and Jo Shaw1

1. INTRODUCTION In the modern state paradigm, the attribution of citizenship in its membership sense is regarded as a matter for national sovereignty. Although it is unlikely that the paradigm of national sovereignty as reference point for citizenship ever truly reflected the complex tapestry of human life in relation to migration and mobility, not to mention the complexity of the ‘state’ form of the polity, the basic structure has stayed in place such that states remain largely free to attribute citizenship, or not, as they see fit, according to their domestic laws. This comes across particularly clearly when one looks at international law, which refers to the concept of ‘nationality’ as the legal baseline of the recognized connection between states and individuals.2 There are only limited norms within international law which could restrict that sovereignty, or which are aimed to make the recognition of each state’s nationality by other states work smoothly. Traditional regulatory techniques of command and control operate within states to determine the details of the process of acquisition and loss of citizenship and to settle the range of rights which are attributed to citizens. Citizenship, like the process of bordering, when articulated via the medium of law, is a hard-edged instrument lending a certain type of order to human life. Laws determine citizenship status and rights and in many cases these provisions are subject to authoritative interpretation by courts. In many states, the details of acquisition and loss of citizenship are based on constitutional norms which determine the fundamental parameters of membership and belonging, bringing into play principles such as equality and other human rights guarantees (as well as, in some cases, religious and ethnic preferences). Linkages between constitutional norms and citizenship laws are sometimes direct and in other cases merely implicit. Some of these constitutional norms may also reflect international legal norms in areas such as statelessness. Focusing our attention on national citizenship regimes within the framing of state sovereignty would not seem to offer promising terrain for the study of the implications of ‘soft law’ for citizenship regimes, except perhaps in the sphere of ‘integration’ of non-citizens into the polity. We argue that the implications of a soft law optic for citizenship law are more complex than this quick pen portrait allows. We seek to show how, in several respects, citizenship law leaves space for ‘soft law’ ideas and principles, encompassing a range of ‘soft’ norms and institutions which operate both within and especially ‘beyond’ the state. In this chapter, we describe the lay of the land with respect to the implications of ‘soft law’ for the domain of citizenship across three dimensions: as a counter-paradigm to ‘hard law’ (the law-making or normative dimension); as a tool of governance, especially in the context of multi-level governance (the governance dimension); and as a gateway idea opening up scholarship on citizenship to be more receptive to a range of comparative approaches (the methodological dimension). We structure our analysis of these three dimensions around three sites of convergence between soft law and citizenship regimes, wherein the former increasingly impacts upon the latter, determining the edges and sometimes even the core meaning of citizenship. The 353

354  Research handbook on soft law first of these is the right to a nationality, widely characterized as ‘the right to have rights’ following Hannah Arendt’s (1968, p. 177) famous phrase. The second concerns the regulation of the modes of citizenship acquisition – that is, the conditions under which citizenship of a given state may be acquired. The third relates to multicultural citizenship, understood as a set of differentiated, group-specific cultural rights intended to secure the equal membership of minorities and indigenous peoples (Kymlicka 1995). The latter departs from traditional accounts of citizenship as a uniform bundle of civil, political, and social rights (e.g. Marshall 1950), offering an illuminating case through which to examine the expanding ‘content’ of citizenship as it relates to soft law. These examples are by no means exhaustive; our aim is simply to provide a series of ‘capsule’ illustrations of the manifold intersections between soft law and citizenship regimes. Throughout the chapter, we give especially close attention to the transnational legal dimensions of citizenship regimes, although the argument we develop here could also apply within the state, in so far as regional or local dimensions of citizenship may hollow out national citizenship regimes just as much as transnational norms may do. Drawing on this material, the analysis suggests that ‘soft law’ is already a significant factor in relation to the progressive evolution of citizenship regimes, especially taking into consideration the regional and international law dimensions, and thus it may come, in the future, to play an even more substantial role both in relation to legal development and also in relation to knowledge about that legal development, which takes into consideration a global approach which goes beyond the standard tropes of knowledge about citizenship regimes in the Global North. We also present in the conclusion a brief discussion of whether and how such ‘soft law’ interventions could lead us closer to the ideas and practices of an ‘international citizenship law’, with significant impacts upon national citizenship regimes, with all the democracy, legitimacy, and accountability issues that such a new framing for citizenship would entail.

2.

THE LAW-MAKING DIMENSION

From the perspective of international law-making, the term ‘soft law’ can be understood to refer to a range of ‘non-legally binding but normatively worded instruments used in contemporary international relations by states and international organizations’ (Boyle 2018, p. 121).3 On this view, soft law instruments, such as inter-state declarations or guidelines issued by international organizations, are formally distinct from binding (‘hard’) sources of international law, such as treaties. The former can nonetheless play a number of important roles in the development of the latter: soft law instruments can, for example, provide a blueprint for subsequent treaty-making, offer interpretative authority concerning the substantive content of existing treaty norms, constitute evidence of state practice and opinio juris for the purposes of establishing customary rules of international law, and/or be declarative of general principles of law within the meaning of Article 38(1)(a) of the Statute of the International Court of Justice (1946; Boyle 2018, pp. 125–127). Notwithstanding its non-bindingness, soft law can therefore have a variety of significant legal effects, from clarifying and consolidating existing ‘hard’ norms to anticipating and influencing their future creation. Soft law is perhaps especially important in the establishment and development of new(er) domains of law-making; Catherine Redgwell (2017, p. 956) has noted, for example, that international environmental law has been ‘a particularly fertile area for soft law’ (on international climate change soft law, see Kulovesi and Recio’s chapter in this volume).

Soft law and citizenship regimes  355 To what extent has soft law played an equivalent role vis-à-vis the legal regulation of citizenship regimes, in particular ‘beyond the state’? Nationality law is decidedly not a new area of international law-making, but the focus of law-making has changed in decisive ways since the first international instruments were adopted more than 100 years ago. In its 1923 Nationality Decrees Issued in Tunis and Morocco Advisory Opinion, the Permanent Court of International Justice affirmed the orthodox view that ‘nationality is not, in principle, regulated by international law’, unless one state has voluntarily undertaken specific treaty obligations in respect of another. At that time, international treaty-making in the domain of citizenship focused principally on the maintenance of ‘state order’, as illustrated by the 1930 Convention on Certain Questions Relating to the Conflict of Nationality Laws, which reiterated that ‘it is for each State to determine under its own law who are its nationals’ (Article 1) and addressed such ‘problems’ as dual citizenship (Articles 3–6) and the nationality of married women (Chapter III). Since the latter half of the 20th century, however, domestic citizenship regimes have increasingly been subject to a broader range of constraints ‘beyond’ the state, notably in the form of emerging international human rights norms, which centre on the interests of individuals rather than those of states (Spiro 2011, p. 710; Shaw 2020, p. 224). An obvious example is the right to a nationality, found in Article 15 of perhaps the most famous soft law instrument in the world, the 1948 Universal Declaration of Human Rights (UNGA 1948; UDHR). The law-making effects of this provision have been somewhat mixed in practice. Unlike much of the rest of the Declaration, Article 15 of the UDHR was not subsequently transformed into a binding norm under one of the international human rights covenants. Instead, Article 24(3) of the International Covenant on Civil and Political Rights (1966; ICCPR) merely provides that ‘[e]very child has the right to acquire a nationality’, as reiterated in Article 7 of the Convention on the Rights of the Child (1989). According to the subsequent ‘soft’ jurisprudence of the United Nations Human Rights Committee (1989, para 8), states are therefore required ‘to adopt every appropriate measure, both internally and in cooperation with other States, to ensure that every child has a nationality when [they are] born’. Relevant treaty norms were also introduced in the 1954 Convention Relating to the Status of Stateless Persons and the 1961 Convention on the Reduction of Statelessness, which create obligations to grant citizenship to individuals who would otherwise be left stateless but stop short of creating a general right to a nationality. The practical impact of all of these norms is further limited by gaps in ratification by states, the absence of ‘hard’ international enforcement mechanisms, and the lack of domestic legal effects in dualist states (Shaw 2020, pp. 245–251).4 The normative ambitions of Article 15 of the UDHR have, however, been more fully realized in the context of regional human rights regimes. Within the Inter-American system, Article 20 of the American Convention on Human Rights (1969; ACHR) provides that ‘[e]very person has the right to a nationality’, Article 15 of the UDHR having been expressly cited in the travaux préparatoires (Organization of American States 1969, p. 232). The Inter-American Court of Human Rights has found violations of this provision in cases concerning the denial of access to citizenship (Case of the Yean and Bosico Girls v Dominican Republic, judgment of 8 September 2005) and arbitrary deprivation of citizenship (Case of Ivcher-Bronstein v. Peru, judgement of 6 February 2001). The African Court on Human and Peoples’ Rights (ACtHPR) has likewise found violations of the right to a nationality, notwithstanding its formal textual absence from the African Charter on Human and Peoples’ Rights (1981). In Anudo v United Republic of Tanzania (Application No 012/2015, judgment of 22 March 2018), the Court held

356  Research handbook on soft law that Article 15 of the UDHR constituted a customary norm of international law, finding the respondent state to have breached its corresponding obligation not to arbitrarily deprive the applicant of his Tanzanian citizenship.5 A similar ‘hardening’ of the soft right to a nationality has also occurred at the domestic level. For example, in Hussein v Attorney General (Civil Suit No. 437 of 2019, judgment of 18 March 2022), the High Court of Uganda found the government’s practice of denying Ugandan citizenship to Ugandan-born persons of Somali origin to be unlawful, being both unconstitutional and incompatible with its international obligations under Article 15 of the UDHR, as well as the 1961 Statelessness Convention. Beyond a general right to a nationality, international norms have also begun to encroach upon the specific modes of citizenship acquisition applicable within domestic regimes, notably in the context of non-discrimination. Discrimination in access to citizenship is prohibited, for instance, under Article 5(d)(iii) of the 1965 International Convention on the Elimination of All Forms of Racial Discrimination (CERD) and under Article 9 of the 1979 Convention on the Elimination of All Forms of Discrimination against Women (CEDAW). Both of these ‘hard’ treaty provisions were anticipated by earlier soft law, namely Article 3 of the 1963 Declaration on the Elimination of All Forms of Racial Discrimination (UNGA 1963) and Article 5 of the 1967 Declaration on the Elimination of Discrimination against Women (UNGA 1967). Their substantive content has later been clarified and developed by the soft jurisprudence of the relevant treaty monitoring bodies. The UN Committee on the Elimination of Racial Discrimination (2004, para 13) has expressed the view, for instance, that state parties are required to ‘[e]nsure that particular groups of non-citizens are not discriminated against with regard to access to citizenship or naturalization, and to pay due attention to possible barriers to naturalization that may exist for long-term or permanent residents’. With respect to Article 9 of CEDAW, the UN Committee on the Elimination of Discrimination against Women (2011, para 37) considers that state parties are required, inter alia, to ensure that women have equal rights to transmit their nationality to their children and to foreign spouses. Both treaty provisions have also inspired the development of additional soft law instruments. For example, citing Article 9 of CEDAW, the 2016 Human Rights Council Resolution on The Right to a Nationality: Women’s Equal Nationality Rights in Law and in Practice ‘urges states to take immediate steps to reform nationality laws that discriminate against women by granting equal rights to men and women to confer nationality on their children and spouses and regarding the acquisition, change or retention of their nationality’ (Human Rights Council 2016, para. 5). In a similar vein, the High Commissioner on National Minorities (HCNM) of the Organization for Security and Co-operation in Europe (2008, p. 19; OSCE) has cited Article 5 of CERD in support of its recommendation that facilitated naturalization policies for external ‘kin-state’ minorities (Pogonyi 2022) should not discriminate on the grounds of national or ethnic origin, nor give rise to discrimination against dual citizens. Finally, in addition to norms constraining the modes of citizenship acquisition, recent decades have also witnessed an ‘internationalisation’ of multicultural citizenship, principally in the form of the codification and diffusion of minority rights norms through treaties and related international standards and guidelines (Kymlicka 2007). These include Article 27 of the ICCPR, which provides that ‘ethnic, religious or linguistic minorities […] shall not be denied the right […] to enjoy their own culture, to profess and practise their own religion, or to use their own language’. Subsequent soft law developments have expanded on the apparently limited substantive content of this provision, whose cautious formulation suggests that it entails no more than negative obligations of non-interference on the part of state parties. The

Soft law and citizenship regimes  357 1992 UN Declaration on the Rights of Persons Belonging to National or Ethnic, Religious and Linguistic Minorities (UNGA 1992), whose preamble notes that it was ‘[i]nspired by the provisions of article 27’ of the ICCPR, suggests instead that positive obligations are required: Article 4 provides, for example, that states ‘should take appropriate measures so that, wherever possible, persons belonging to minorities may have adequate opportunities to learn their mother tongue or to have instruction in their mother tongue’. In a similar vein, according to the soft jurisprudence of the Human Rights Committee (1994, para 6.2), under Article 27 of the ICCPR, ‘positive measures by States may […] be necessary to protect the identity of a minority and the rights of its members to enjoy and develop their culture and language and to practice their religion’. The Committee (1994, paras 5.1–5.2) further considers that states may not restrict the application of Article 27 to their own citizens; rather, permanent residents, migrant workers ‘or even visitors’ may constitute a ‘minority’ for these purposes. This constitutes a significant departure, by means of soft law, from the original intentions of the drafters: according to the travaux préparatoires, ‘[i]t was agreed that the article should cover only separate or distinct groups, well-defined and long-established on the territory of a State’ and ‘may not be interpreted as entitling any group settled in the territory of a State, particularly under the terms of its immigration laws, to form within that State separate communities which might impair its national unity or security’ (UNGA 1955, paras 184–186). Related developments have also occurred at the regional level, notably in respect of the 1995 Council of Europe Framework Convention for the Protection of National Minorities (FCNM), whose provisions drew inspiration from the non-binding ‘Copenhagen Document’ of the Conference on Security and Co-operation in Europe, as well as proposals from the Venice Commission (1991) and Parliamentary Assembly of the Council of Europe (PACE; 1993). While primarily applied to so-called ‘old’ or ‘historical’ minorities, there has recently been a tentative move to also include immigrant-origin groups within the treaty’s scope, partly driven by the soft jurisprudence of the Advisory Committee on the Framework Convention, which has argued that the treaty as a whole should not be restricted to the citizens of the state, nor to those with ‘longstanding, firm and lasting ties’ (Jacob-Owens 2022). At both the global and regional levels, soft law developments have thus contributed to expanding the substantive content and personal scope of application of multicultural citizenship norms. In turn, these norms can produce ‘hard’ legal effects at the domestic level, as the invocation of the FCNM as a source of normative standards and interpretative authority in minority rights adjudication in Central, Eastern, and South-Eastern European states attests (Palermo 2008). In sum, as in other areas of (international) law-making, soft law has contributed in a variety of ways to the emergence of citizenship norms beyond the state. First, soft law instruments have paved the way for subsequent ‘hard’ treaty-making, as in the examples of the ACHR, CERD, CEDAW, and FCNM. Second, the substantive content of relevant treaty provisions has been clarified and developed by means of soft law, such as the UN Minorities Declaration and the soft jurisprudence of the relevant treaty monitoring bodies. Third, some of these treaty provisions have given rise to further soft law-making as a means of articulating new citizenship norms. Examples here include the recommendations of the OSCE HCNM concerning kin-state citizenship policies and non-discrimination. Finally, the case law of the ACtHPR demonstrates that a soft citizenship norm, namely the right to a nationality under Article 15 of the UDHR, can have hard legal effects via its recognition as international custom. These observations suggest that soft law has played a central role in the ongoing development of international citizenship norms.

358  Research handbook on soft law

3.

THE GOVERNANCE DIMENSION

Beyond its role in conventional (international) law-making, soft law can be understood as a key tool of governance, including – and arguably especially – beyond the state.6 This is perhaps most readily apparent in the multi-level governance system of the European Union (EU), a paradigmatic case being its ‘open method of coordination’, which relied on ‘soft’ guidelines, reporting, and monitoring, rather than ‘hard’ legislation and judicial enforcement (Maher 2021, p. 28; see also van Gerven and Stiller’s chapter in this volume). We can discern certain similarities in the EU institutions’ response to ‘investor citizenship’ schemes introduced by a number of member states, notably Bulgaria, Cyprus, and Malta (Džankić 2019, pp. 189–199). Such schemes allow wealthy ‘third-country nationals’ to acquire national citizenship in return for a substantial financial investment. In turn, owing to its derivative character under Article 20 of the Treaty on the Functioning of the European Union, such individuals can thereby also acquire supranational citizenship of the EU, with concomitant free movement and other rights. Following a debate on the issue in January 2014, the European Parliament adopted Resolution on the Sale of EU Citizenship, which, while acknowledging that ‘matters of residency and citizenship lie within the exclusive competence of the Member States’, declared that ‘access to funds should not be the main criterion in conferring EU citizenship on third-country nationals’ and ‘call[ed] on Malta to bring its current citizenship scheme into line with the EU’s values’ (European Parliament 2014, recital M. and para. 9). No immediate change to the scheme occurred, however, prompting the European Commission to threaten infringement proceedings. Commission officials then met with the Maltese authorities to agree on a suitable amendment to its investor citizenship scheme, namely the introduction of a residence requirement (European Commission 2014). The Commission (2019, p. 24) has subsequently issued a general report on investor citizenship schemes, signalling its intention to ‘monitor wider issues of compliance with EU law raised by the schemes’ and to ‘take necessary action, as appropriate’. Notwithstanding, and perhaps because of, the established position in EU law that the regulation of citizenship acquisition remains within the exclusive competence of the member states, the EU institutions have thus turned to a softer governance approach – albeit coupled with the potential threat of infringement proceedings – in seeking to limit member state discretion concerning the acquisition of national and EU citizenship by investment. Both in the EU context and in more diffuse international ‘regime complexes’, multi-level governance is now no longer limited to ‘top-down’ regulation, but rather can increasingly be understood in ‘experimentalist’ terms as a dynamic interplay between local and supranational actors, through which shared problems, objectives, and norms are articulated, addressed, and contested (De Búrca and Scott 2006; De Búrca et al. 2014). Within regional and international human rights system, for example, the interaction between social movements, international institutions, and national governments has given rise to more participatory forms of standard-setting (De Búrca 2021). One such example is the United Nations Declaration on the Rights of Indigenous Peoples (UNDRIP), adopted by the General Assembly in 2007 (UNGA 2007), which affirms a wide range of group-specific cultural rights and hence contributes to the international diffusion of multicultural citizenship norms. In contrast to the development of earlier international instruments in this field, from which indigenous peoples themselves were excluded, the UNDRIP was a direct product of mobilization by indigenous activists, who sought to improve ‘the recognition and implementation of indigenous rights at both

Soft law and citizenship regimes  359 domestic and international level’ (Gómez Isa 2016, pp. 185–186). To this end, these activists deliberately opted for a soft declaration rather than a binding treaty, partly in order to create greater scope for input from non-state actors, a matter of particular significance for indigenous and other non-dominant groups (Gómez Isa 2016, p. 194). This example thus illustrates the potential role of soft law as a more participatory means of articulating international (multicultural) citizenship norms in the context of experimentalist multilevel governance systems. Once articulated, these norms can in turn give rise to further bottom-up mobilization. For example, Council of Europe standard-setting in the field of minority rights has had a ‘constitutive effect’ on activism surrounding questions of citizenship and integration in Germany, prompting and informing the efforts of church groups and Turkish community associations, among others (Checkel 1999, p. 105). Finally, soft law may also provide the basis for another ‘technology’ of governance: indicators. Following Davis et al. (2012, pp. 73–78), indicators can be understood as being characterized by: a name; rank-ordered structure; simplified (often numerical) representation of raw data; and potential use for evaluation against some (implicit) standard(s). Indicators are well-established in fields such as democracy and the rule of law and often rely, at least in part, on soft law instruments as sources for the normative standards on which they are based. For example, the United Nations Rule of Law Indicators were ‘directly inspired’ by ‘human rights principles and criminal justice standards’ found not only in international treaties but also in non-binding guidelines and declarations issued by various UN bodies (United Nations 2011, p. 67). In the domain of citizenship, soft law-based indicators have begun to be used in relation to birth registration and the provision of legal identity, measures that are currently being promoted as part of international efforts to end statelessness worldwide (e.g., UNHCR 2014). For example, target 16.9 of the UN Sustainable Development Goals (SDGs) is to ‘provide legal identity for all, including birth registration’ by 2030 (Manby 2020; Sperfeldt 2021). This target is linked to a corresponding indicator, namely the ‘[p]roportion of children under 5 years of age whose births have been registered with a civil authority, by age’. The inclusion of legal identity and birth registration in the SDGs has spawned further international initiatives, such as the World Bank’s West Africa Unique Identification for Regional Integration and Inclusion (WURI) project and its Identification for Development Initiative (ID4D), which involve the elaboration of principles and technical standards, funding for domestic implementation, and monitoring and reporting procedures. At the domestic level, these initiatives have in turn led to new forms of legal identity provision. One such example is ‘e-ID Togo’, a new digitized, biometric identification system for residents of Togo introduced in 2020, which was conceived and funded in cooperation with WURI and ID4D (Offermann 2022, pp. 62–63). In this instance, we thus see the emergence of an indicators-based ‘soft’ governance approach to the provision of legal identity and birth registration, linked to the wider international norms of statelessness avoidance and the right to a nationality. Overall, these examples suggest that soft law has been central to emerging forms of citizenship governance beyond the state. While offering a more flexible, dynamic alternative to top-down, command-and-control regulation, the use of soft law in such contexts is not without its drawbacks. As the examples of the European Commission and the World Bank illustrate, indicators and other governance standards are frequently developed by technical ‘experts’ employed by international organizations, rather than elected representatives working on behalf of national governments. This has the effect of side-lining traditional legislative bodies and endowing such organizations with significant ‘indirect power’ to influence public

360  Research handbook on soft law policy-making (Broome et al 2018). In this sense, indicators and other soft standards can be seen to contribute to a technocratic rather than democratic form of (multi-level) governance, raising issues of accountability and democratic legitimacy (Mörth 2009; see also Mörth’s chapter in this volume). Moreover, in the context of the SDGs, it has been argued that the imposition of generic standards for birth registration and legal identity provision without proper regard to domestic context – in particular, the local ‘politics of documentation and evidence’ – will have the opposite effects to those intended, further entrenching marginalisation and statelessness (Manby 2020; Jain 2022, pp. 283–286). On the other hand, the example of the UNDRIP demonstrates how an experimentalist approach to multilevel governance can foster, rather than stifle, bottom-up participation.

4.

THE METHODOLOGICAL DIMENSION

Building further on its role in the formulation of indicators, soft law can also take on methodological significance when we go beyond the doctrinal analysis of legal sources and develop politically- and sociologically-informed comparative approaches to the study of citizenship regimes. Given their typically rank-ordered and numerical character (see above), indicators readily lend themselves to quantitative social scientific inquiry seeking to describe and explain global variation in relation to a given social phenomenon. The field of citizenship studies is replete with indicators-based scholarship, primarily focused on measuring the ‘inclusiveness’ or ‘exclusiveness’ of the rules regulating citizenship acquisition. Early examples include Harald Waldrauch’s (2006) ‘modes of acquisition of nationality’ and Marc Morjé Howard’s (2009) Citizenship Policy Index, which sought to capture the conditions for access to citizenship across the EU member states, and Thomas Janoski’s (2010) Barriers to Naturalization Index, which focused on the countries of the Organization for Economic Co-operation and Development (OECD). More recent efforts have begun to expand the geographic scope of such research. The latest version of the widely used ‘access to nationality’ strand of the Migrant Integration Policy Index (MIPEX) (Solano and Huddleston 2020), for instance, covers 13 countries beyond the EU and OECD, including a handful from the so-called Global South. The most ambitious attempt to expand the scope of comparative citizenship research is the Global Citizenship Observatory’s (GLOBALCIT) Citizenship Law Dataset (Vink et al. 2021), which captures the modes of citizenship acquisition and loss reflected in domestic legislation in 190 states across the world. Building on related data, the GLOBALCIT Global Birthright Indicators (Honohan et al. 2020) capture the restrictiveness of the conditions for acquiring citizenship at birth across 179 different countries. In the domain of multicultural citizenship, the Multiculturalism Policy Index (www​.queensu​.ca/​mcp) records the presence or absence of group-specific measures for ‘immigrant minorities’, ‘historical national minorities’, and ‘indigenous peoples’ across 21 ‘Western democracies’. While clearly offering a meaningful basis for comparison, existing studies of this sort share a number of common limitations. First, as several of the previous examples illustrate, there has been a pronounced geographical bias in favour of studying (Western) Europe and North America, and only slowly is the data being collected to allow studies to be extended to the Global South or globally across the world. Second, as Maarten Vink (2017, p. 241) observes, the underlying political agenda driving comparative citizenship studies has generally focused on immigrant integration and security in ‘Western’ states and on democratization and devel-

Soft law and citizenship regimes  361 opment in ‘non-Western’ contexts, to the (limited) extent that the latter have been included; ‘[o]ther political agendas, such as an international one of avoiding statelessness and refugee protection, have so far hardly inspired comparative studies on citizenship regimes’. Third, citizenship indicators are typically based exclusively on doctrinal sources, usually just domestic legislation, and hence capture only the ‘law in the books’ rather than the ‘law in action’ (or indeed ‘soft law’ sources). Soft law might plausibly play a role in addressing these limitations. The alternative political agendas proposed by Vink could be served by an approach to indicators-based comparative research structured by reference to international (soft) standards concerning the avoidance of statelessness or the rights of refugees. Given the (purportedly) global applicability of such standards, an approach of this sort might also assist in expanding the geographic scope of such studies. An approach of this kind is illustrated by a recent joint report from the Institute on Statelessness and Inclusion (ISI) and GLOBALCIT, which uses an index of the grounds for citizenship loss under domestic legislation to compare deprivation powers across 190 countries worldwide and assesses these in light of the Principles on Deprivation of Nationality as a National Security Measure, a set of (soft) international norms and standards endorsed by the UN Special Rapporteur on the Protection and Promotion of Human Rights and Fundamental Freedoms while Countering Terrorism (Van der Baaren et al. 2022). Future comparative research in this vein might also expand beyond conventional doctrinal sources to examine the implementation of citizenship law and policy in practice, for instance through analysis of domestic soft law. Existing work in this direction has used indicators-based comparative research to explore how administrative procedures create barriers to naturalization (Huddleston 2013; Huddleston and Falcke 2020) and analysed domestic policy guidance and interviews with public officials to illuminate the role of individual discretion in access to citizenship (Sredanovic 2022). In this context, international norms and standards concerning administrative practice in areas such as birth registration might form a suitable baseline for comparison. A soft law-based approach to comparative citizenship studies is not without its own limitations, however. Depending on their source, the use of soft law instruments in this context may simply reinforce the ‘conceptual imposition’ common to existing attempts to expand indicators-based comparative citizenship research to ‘non-Western’ settings, wherein researchers assume the generalizability of the Eurocentric norms or concepts on which such indicators are based (Džankić 2022). Moreover, especially given the blurred line between explanatory and evaluative indicators-based comparisons (Nelken 2016, p. 401), there is a clear risk that imposing Eurocentric normative standards as a baseline for global comparison will inevitably reinforce the supposed ‘superiority’ of states in the Global North, as a consequence of them being already more closely aligned with those standards. On the other hand, the stronger protection of the right to a nationality within the regional human rights systems of Africa and the Americas demonstrates that Global North states are not always ‘ahead’, even by these ‘conventional’ standards. Furthermore, alternative normative standards might also be articulated via soft law instruments, where these have been developed by or with non-Western and/or traditionally marginalized groups, as the example of the UNDRIP illustrates. The use of such standards as a baseline for the development of indicators might thus help to foster a more balanced approach to comparative citizenship research, producing knowledge that better represents perspectives from both Global South and North (Pedroza and Palop-García 2022). In

362  Research handbook on soft law this sense, soft law might have a role to play in improving indicators-based global comparisons by making the outcomes more legitimate and acceptable.

5. CONCLUSION This chapter has presented a new way of mapping the field of citizenship law and citizenship studies, using the ideas that stem from thinking about ‘soft law’. Table 23.1 summarizes some of the capsule illustrations that we have used in this chapter, highlighting how we can map the ‘dimensions’ and ‘sites’ of soft law. Table 23.1

Dimensions and sites of soft law Sites

Right to a nationality

Regulating modes of

Multicultural citizenship

acquisition of citizenship Dimensions Development of linguistic rights

Law-making

‘Hardening’ of the ‘norm’ contained Gradual development of

dimension

in Article 15 UDHR by regional

what are now CEDAW

impacting on enjoyment of

human rights courts

norms as basis for

citizenship

greater gender equality in nationality acquisition / transfer Governance

SDG 16.9 on birth registration and

EU approach to investor

Impact of UNDRIP on citizenship

dimension

legal identity

citizenship regimes in

of Indigenous peoples

Member States Methodological

ISI/GLOBALCIT grounds for

Citizenship Regime

dimension

citizenship loss index

Inclusiveness Index

Multiculturalism Policy Index

It is clear that there exists a wide range of productive intersections between soft law and citizenship regimes. The list could go well beyond the illustrations we have used. In section 2, we saw how soft law has contributed to the emergence of international norms shaping and constraining domestic citizenship practices in relation to the right to a nationality, the modes of citizenship acquisition, and multicultural citizenship. Over a decade ago, Peter Spiro (2011, p. 718) observed that such ‘developments are [thus far] only suggestive of the ultimate crystallization of an international regime regarding citizenship practices’ and that ‘[f]or now, the prospective norms are provisional, fragile, and unstable’. This remains an important caveat. Nonetheless, to the extent that such ‘crystallization’ will occur, it seems likely that soft law will have a central role to play. One plausible next step, for example, could be the formulation of a non-binding ‘Global Citizenship Compact’, along the lines of the existing Global Compacts on Refugees and on Migration, which could serve to establish international standards concerning such topical issues as investor citizenship schemes, the toleration of dual citizenship, and external voting rights for non-resident citizens (Hase and Mantha-Hollands 2019, p. 44). The sense that soft law has an important role to play with respect to the development of an international citizenship regime is reinforced by the growth of multi-level citizenship governance, exemplified in section 3 by the EU institutions’ approach to investor citizenship schemes, the articulation of multicultural citizenship norms through experimentalist human rights systems, and the indicators-based global governance of legal identity and birth registra-

Soft law and citizenship regimes  363 tion. Untethered from the constraints of conventional international law-making, supranational and global governance actors are relatively free to (attempt to) create new, potentially more far-reaching constraints on domestic citizenship regimes. However, if limited to top-down, technocratic standard-setting, that freedom comes at the price of accountability and legitimacy. The future development of global citizenship governance will thus need to pay due attention to the bottom-up participation of relevant constituencies. To the extent that states across the world remain subject to a growing array of common norms and standards, this creates ever increasing scope for meaningful global comparison in the field of citizenship studies, as discussed in section 4. A soft law-based approach might help researchers to move beyond the political agendas and conventional doctrinal sources that have so far characterized scholarship in this field. As observed in relation to the governance dimension, future studies of this sort will need to be sensitive to the underlying sources of their normative standards and assumptions, with a view to improving not only their geographic scope but also their representativeness.

NOTES 1. This chapter draws in part on the Leverhulme Major Research Fellowship MRF-2016-190 (2018-2022) held by Jo Shaw. We are very grateful to Rainer Bauböck, Jelena Džankić, and Maarten Vink for their comments on an earlier draft. The usual disclaimer applies. 2. For the purposes of this chapter, we use the terms ‘citizenship’ and ‘nationality’ interchangeably. 3. This is just one of a number of ways in which the term is understood by international lawyers: see, for instance, Chinkin 1989. 4. Dualist states are those in which domestic and international law are treated as two distinct legal orders, such that the latter does not automatically have effect in the former, absent incorporation, for instance by means of domestic legislation. 5. Article 15 of the UDHR provides both that ‘[e]veryone has the right to a nationality’ and that ‘[n]o one shall be arbitrarily deprived of [their] nationality nor denied the right to change [their] nationality’. 6. This is not to suggest that states are not also important governance actors: see Bell and Hindmoor 2009.

REFERENCES African Charter on Human and Peoples' Rights., (1981). Nairobi, signed 27 June 1981, in force 21 October 1986, 1520 UNTS 217. American Convention on Human Rights (ACHR)., (1969). San José, signed 22 November 1969, in force 18 July 1978, 1144 UNTS 123. Arendt, H., [1951] (1968). The origins of totalitarianism. New York: Harcourt, Brace and Jovanovich. Bell, S. and Hindmoor, A., (2009). Rethinking governance: the centrality of the state in modern society. Cambridge: Cambridge University Press. Boyle, A., (2018). Soft law in international law-making. In: M. D. Evans, ed. International law. 5th ed. Oxford: Oxford University Press. Broome, A., Homolar, A. and Kranke, M., (2018). Bad science: international organizations and the indirect power of global benchmarking. European Journal of International Relations. 24(3), 514-539. Checkel, J. T., (1999). Norms, institutions, and national identity in contemporary Europe. International Studies Quarterly. 43(1), 83-114. Chinkin, C., (1989). The challenge of soft law: development and change in international law. International and Comparative Law Quarterly. 38(4), 850-866.

364  Research handbook on soft law Convention on Certain Questions Relating to the Conflict of Nationality Laws., (1930). The Hague, signed 13 April 1930, in force 1 July 1937, 179 League of Nations Treaty Series 89. Convention on the Elimination of All Forms of Discrimination against Women (CEDAW)., (1979). New York, signed 18 December 1979, in force 3 September 1981, 1249 UNTS 13. Convention on the Reduction of Statelessness., (1961). New York, signed 30 August 1961, in force 13 December 1975, 989 UNTS 175. Convention on the Rights of the Child., (1989). New York, signed 20 November 1989, in force 2 September 1990, 1577 UNTS 3. Convention Relating to the Status of Stateless Persons., (1954). New York, signed 28 September 1954, in force 6 June 1960, 360 UNTS 117. Council of Europe Framework Convention for the Protection of National Minorities (FCNM)., (1995). Strasbourg, signed 1 February 1995, in force 1 February 1998, European Treaty Series No. 157. Davis, K. E., Kingsbury, B. and Merry, S. E., (2012). Indicators as a technology of global governance. Law & Society Review. 46(1), 71–104. de Búrca, G., (2021). Reframing human rights in a turbulent era. Oxford: Oxford University Press. de Búrca, G. and Scott, J., (2006). Introduction: new governance, law and constitutionalism. In: G. De Búrca and J. Scott, eds. Law and new governance in the EU and the US. Oxford: Hart Publishing. de Búrca, G., Keohane, R. and Sabel, C., (2014). Global experimentalist governance. British Journal of Political Science. 44(3), 477–486. Džankić, J., (2019). The global market for investor citizenship. London: Palgrave Macmillan. Džankić, J., (2022). Avoiding measure for a measure’s sake: limits and value added of citizenship law indicators. In: É. Fargues, ed. Going global: opportunities and challenges for the development of a comparative research agenda on citizenship policies at the global level. San Domenico di Fiesole: European University Institute. European Commission., (2014). Joint press statement by the European Commission and the Maltese authorities on Malta's Individual Investor Programme (IIP). Memo 14/70. European Commission., (2019). Investor citizenship and residence schemes in the European Union. COM(2019) 12 final. European Commission for Democracy through Law (Venice Commission)., (1991). Proposal for a European Convention for the Protection of Minorities. CDL(91)007. European Parliament., (2014). European Parliament resolution of 16 January 2014 on EU citizenship for sale. Resolution 2013/2995(RSP). Gómez Isa, F., (2016). The role of soft law in the progressive development of indigenous peoples’ rights. In: S. Lagoutte, T. Gammeltoft-Hansen and J. Cerone, eds. Tracing the roles of soft law in human rights. Oxford: Oxford University Press. Hase, J. and Mantha-Hollands, A., (2019). Toward a global compact on citizenship? WZB Mitteilungen. 163, 43–44. High Commissioner on National Minorities., (2008). The Bolzano/Bozen recommendations on national minorities in inter-state relations. The Hague: Organization for Security and Co-operation in Europe. Honohan, I., Rougier, R., Bauböck, R. and Erdilmen, M., (2020). Global birthright indicators. San Domenico di Fiesole: European University Institute. Howard, M. M., (2009). The politics of citizenship in Europe. Cambridge: Cambridge University Press. Huddleston, T., (2013). The naturalisation procedure: measuring the ordinary obstacles and opportunities for immigrants to become citizens. RSCAS Policy Paper 2013/16. San Domenico di Fiesole: European University Institute. Huddleston, T. and Falcke, S., (2020). Nationality policies in the books and in practice: comparing immigrant naturalisation across Europe. International Migration. 58(2), 255–271. International Convention on the Elimination of All Forms of Racial Discrimination (CERD)., (1965). New York, signed 7 March 1966, in force 4 January 1969, 660 UNTS 195. International Covenant on Civil and Political Rights (ICCPR)., (1966). New York, signed 16 December 1966, in force 28 March 1979, 999 UNTS 171. Jacob-Owens, T., (2022). Immigration and multicultural citizenship in Europe: insights from the Framework Convention for the Protection of National Minorities. International Journal on Minority and Group Rights. 29(1), 167–197. Jain, N., (2022). Manufacturing statelessness. American Journal of International Law. 116(2), 237–288.

Soft law and citizenship regimes  365 Janoski, T., (2010). The ironies of citizenship: naturalization and integration in industrialized countries. Cambridge: Cambridge University Press. Kymlicka, W., (1995). Multicultural citizenship: a liberal theory of minority rights. Oxford: Clarendon Press. Kymlicka, W., (2007). Multicultural odysseys: navigating the new international politics of diversity. Oxford: Oxford University Press. Maher, I., (2021). Revisiting soft law: governance, regulation and networks. In: M. Eliantonio, E. Korkea-aho and O. Ştefan, eds. EU soft law in the member states: theoretical findings and empirical evidence. Oxford: Hart Publishing. Manby, B., (2020). ‘Legal identity for all’ and statelessness: opportunity and threat at the junction of public and private international law. Statelessness & Citizenship Review. 2(2), 248–271. Marshall, T. H., (1950). Citizenship and social class and other essays. Cambridge: Cambridge University Press. Mörth, U., (2009). Soft regulation and global democracy. In: M.-L. Djelic and K. Sahlin-Andersson, eds. Transnational governance: institutional dynamics of regulation. Cambridge: Cambridge University Press. Nelken, D., (2016). From pains-taking to pains-giving comparisons. International Journal of Law in Context. 12(4), 390–403. Offermann, M., (2022). Report on citizenship law: Togo. RSC/GLOBALCIT-CR 2022/06. San Domenico di Fiesole: European University Institute. Organization of American States., (1969). Documents of the 1969 Inter-American Conference on Human Rights. OEA/Ser.K/XVI/1.2. Washington, DC: General Secretariat of the Organization of American States. Palermo, F., (2008). Domestic enforcement and direct effect of the Framework Convention for the Protection of National Minorities: on the judicial implementation of the (soft?) law of integration. In: A. Verstichel, A. Alen, B. de Witte and P. Lemmens, eds. The Framework Convention for the Protection of National Minorities: a useful pan-European instrument?. Antwerp: Intersentia. Parliamentary Assembly of the Council of Europe., (1993). Additional Protocol on the Rights of National Minorities to the European Convention on Human Rights. Recommendation 1201 (1993). Strasbourg: Council of Europe. Pedroza, L. and Palop-García, P., (2022). Without the South, it’s not global: creating knowledge about naturalization jointly. In: É. Fargues, ed. Going global: opportunities and challenges for the development of a comparative research agenda on citizenship policies at the global level. San Domenico di Fiesole: European University Institute. Pogonyi, S., (2022). The right of blood: ‘ethnically’ selective citizenship policies in Europe. National Identities [online]. [Viewed 13 September 2022]. Available from: DOI:10.1080/14608944.2021.201 3185. Redgwell, C., (2017). Sources of international environmental law: Formality and informality in the dynamic evolution of international environmental law norms. In: S. Besson and J. d’Aspremont, eds. The Oxford handbook of the sources of international law. Oxford: Oxford University Press. Shaw, J., (2020). The people in question: citizens and constitutions in uncertain times. Bristol: Bristol University Press. Solano, G. and Huddleston, T., (2020). Access to nationality [online]. Migrant Integration Policy Index 2020. [Viewed 20 July 2022]. Available from: www​.mipex​.eu/​access​-nationality. Sperfeldt, C., (2021). Legal identity in the sustainable development agenda: actors, perspectives and trends in an emerging field of research. International Journal of Human Rights. 26(2), 217–238. Spiro, P., (2011). A new international law of citizenship. American Journal of International Law. 105(4), 694–746. Sredanovic, D., (2022). Implementing citizenship, nationality and integration policies: the UK and Belgium in comparative perspective. Bristol: Bristol University Press. Statute of the International Court of Justice., (1946). Annexed to the Charter of the United Nations, San Francisco, signed 26 June 1945, in force 24 October 1945, 33 United States Treaty Series 993. United Nations., (2011). The United Nations Rule of Law Indicators: Implementation Guide and Project Tools. New York: United Nations.

366  Research handbook on soft law United Nations Committee on the Elimination of Discrimination against Women., (2011). Concluding observations: Kuwait. 15th Session, UN Document CEDAW/C/KWT/CO/3-4. United Nations Committee on the Elimination of Racial Discrimination., (2004). General recommendation No. 30 on discrimination against non-citizens. 64th Session, UN Document CERD/C/64/Misc.11/ rev.3. United Nations General Assembly., (1948). Universal Declaration of Human Rights. UNGA Resolution 217 A(III), Official Records of the UN General Assembly, 3rd Session, Supplement No. 13, UN Document A/810 (1948) 71. United Nations General Assembly., (1955). Draft international covenants on human rights: annotation prepared by the Secretary-General. 10th Session, UN Document A/2929. United Nations General Assembly., (1963). Declaration on the Elimination of All Forms of Racial Discrimination. UNGA Resolution 1904 (XVIII), Official Records of the UN General Assembly, 18th Session, Supplement No. 15, UN Document A/5515 (1963) 35. United Nations General Assembly., (1967). Declaration on the Elimination of Discrimination against Women. UNGA Resolution 2263 (XXII), Official Records of the UN General Assembly, 22nd Session, Supplement No. 15, UN Document A/RES/22/2263 (1967) 35. United Nations General Assembly., (1992). UN Declaration on the Rights of Persons Belonging to National or Ethnic, Religious and Linguistic Minorities. UNGA Resolution 47/135, Official Records of the UN General Assembly, 47th Session, Supplement No. 49, UN Document A/RES/47/135 (1992) 210. United Nations General Assembly., (2007). United Nations Declaration on the Rights of Indigenous Peoples (UNDRIP). UNGA Resolution 61/295, Official Records of the UN General Assembly, 61st Session, Supplement No. 49, UN Document A/RES/61/295 (2007) 16. United Nations High Commissioner for Refugees (UNHCR)., (2014). Global action plan to end statelessness 2014–2024 [online]. Geneva: UNHCR. [Viewed 16 September 2022]. Available from: https://​www​.unhcr​.org/​ibelong/​global​-action​-plan​-2014​-2024/​. United Nations Human Rights Committee., (1989). General comment no. 17: Article 24 (the rights of the child). New York: United Nations. United Nations Human Rights Committee., (1994). General comment no. 23(50): Article 27 (the rights of minorities). 50th Session, UN Document CCPR/C/21/Rev.1/Add.5. United Nations Human Rights Council., (2016). Human Rights Council Resolution on The Right to a Nationality: Women’s Equal Nationality Rights in Law and in Practice. UN Human Rights Council Resolution 32/7, 32nd Session, UN Document A/HRC/RES/32/7. Van der Baaren, L., Vink, M., Van Waas, L., Brekoo, A., Moss, D. and Vlieks C., (2022). Instrumentalising citizenship in the fight against terrorism: a global comparative analysis of legislation on deprivation of nationality as a security measure. Institute on Statelessness and Inclusion/ Global Citizenship Observatory. Vink, M., (2017). Comparing citizenship regimes. In: A. Shachar, R. Bauböck, I. Bloemraad and M. Vink, eds. The Oxford handbook of citizenship. Oxford: Oxford University Press. Vink, M., Van der Baaren, L. Bauböck, R., Honohan, I. and Manby, B. (2021). GLOBALCIT citizenship law dataset [online]. San Domenico di Fiesole: European University Institute. [Viewed 16 September 2022]. Available from: https://​hdl​.handle​.net/​1814/​73190. Waldrauch, H., (2006). Acquisition of nationality. In: R. Bauböck, E. Ersbøll, K. Groenendijk and H. Waldrauch, eds. Acquisition and loss of nationality: policies and trends in 15 European states. Volume 1: comparative analyses. Amsterdam: Amsterdam University Press.

Cases Nationality Decrees Issued in Tunis and Morocco. PCIJ, Advisory Opinion No. 4, 7 February 1923, PCIJ Series B No. 4, p. 24. Case of Ivcher-Bronstein v. Peru. IACHR, judgement of 6 February 2001. [Viewed 16 September 2022]. Available from: https://​www​.corteidh​.or​.cr/​docs/​casos/​articulos/​seriec​_74​_ing​.pdf. Case of the Yean and Bosico Girls v Dominican Republic. IACHR, Judgement of 8 September 2005. [Viewed 16 September 2022]. Available from: https://​www​.corteidh​.or​.cr/​docs/​casos/​articulos/​seriec​ _130​_​%20ing​.pdf.

Soft law and citizenship regimes  367 Anudo v United Republic of Tanzania. AfCTHPR, Application No 012/2015, judgement of 22 March 2018. [Viewed 16 September 2022]. Available from: https://​www​.african​-court​.org/​en/​images/​Cases/​ Judgment/​-012​%20​-​%202015​%20​-​%20Anudo​%20Vs​.​%20Tanzania​%20​-​%20Judgment​%2022​ %20March​%202018​%20​-​%20Optimized​.pdf. Hussein v Attorney General. High Court of Uganda, Civil Suit No. 437 of 2019, judgement of 18 March 2022.

24. Soft law: booster or brake for the promotion of gender equality in the EU? Birte Böök and Linda Senden

1. INTRODUCTION At EU and national level, many different – hard and soft – laws and policies have been established in the past decades to enhance gender equality and women’s inclusion and empowerment, both in the area of employment and in the market sphere. Yet, statistics across the globe reveal that this has not sufficed to bring about the required change in practice. Even worse, the latest report of the World Economic Forum shows that the COVID-19 pandemic has had a devastating effect, as the global gender gap has increased by a generation from 99.5 years in 2020 to 132 years in 2022 (World Economic Forum 2022, p. 5). The gender gap concerns girls and women’s economic participation and opportunity, education, health and survival, and political empowerment. The Gender Equality Index score for the EU is still only 77.6 points out of 100 (EIGE 2021). So, despite the progressive development of human rights protection and many specific gender laws and policies over the past decades, women’s under-representation, exclusion, and subordination in social, economic, cultural, and political life not only persists, but is actually increasing. While women’s levels of education have been steadily rising, this has not translated into significantly greater empowerment of women in the labour market and political domain. The persistent gender segregation in higher education and in employment are root causes of that, combined with the (unpaid) care gap that has been exacerbated by the pandemic (OECD 2021). Women remain over-represented in low-paid sectors such as care and education and under-represented in the ‘jobs of tomorrow’ (i.e., the digital and technological domains), which marginalizes them in shaping these increasingly important sectors of society. Women’s under-representation contributes to the spread of gender biases and stereotypes that exist in the offline world to the online world, via the use of algorithms and machine learning (Xenidis and Senden 2020; Gerards and Xenidis 2021; Böök and Senden, forthcoming). The gender gap also remains a key and even growing problem in the political sphere. In the 2022 Global Gender Gap Report, the global score for the political gender gap stood at 22 per cent. It is now expected to take 155 years to close, compared to 95 years in 2020 (World Economic Forum 2022, p. 13). For Europe the overall score for political empowerment is at 39.8 per cent. Women’s severe under-representation in political decision-making negatively impacts on crisis responses, which are persistently male-dominated and not gender-sensitive (UNDP 2021). Women also remain severely under-represented in leadership positions in the private sector, which impacts upon markets and society. This under-representation, alongside the gender segregation in work, education, and decision-making, and women’s higher share in unpaid care, negatively impact women’s earnings. This adds to the gender pay gap as well as a pension gap. As a result, women face a higher risk of poverty, exacerbated by global crises (UN Women 2022). 368

Soft law  369 Given the current state of gender equality in Europe (and beyond), our starting point is that gender equality projects, laws, and policies appear to be ‘deactivated’ in practice by structural elements and that there are serious obstructions to securing the effectiveness of hard equality laws in daily practice. Thus, one must not overestimate the role the law can play in and of itself without, at the same time, tackling other fundamental societal and institutional problems that hamper potential for change, such as gender stereotyping and biases and the lack of institutional commitment. Having made that caveat, in this chapter, we seek to extract the position and potential contribution of soft law in the development and promotion of gender equality (law) in the EU. In order to compare the effectiveness of soft and hard law in achieving gender equality goals, it is first necessary to provide a comparative framework in which to assess the potential and usefulness of soft law as a tool. Section 2 of this chapter will set out the conceptual and methodological framework necessary for making such an assessment, by first outlining our understanding of the concept of soft law and the different functions it fulfils (Senden 2004, pp. 118–121). We then focus on the meaning of effectiveness and the identification of benchmarks that can be considered relevant for carrying out an assessment of the potential effectiveness of soft law acts.1 We also explain how we will assess these acts and outline the limitations of such an exercise. In Section 3, we set out the body of EU gender equality law, including a general mapping of the use of soft law instruments and their functions in this area. So as to flesh out more concretely the potential contribution of such instruments to furthering gender equality in Europe, the chapter will zoom in on three topical domains in which soft law has played a particular role and which are thus instrumental in shedding further light on whether the use of soft law can be seen to boost change or to act as a brake on progress. Section 4 will thus proceed to a more detailed scrutiny of such instruments in the fields of equal pay for women and men, gender-balanced company boards, and gender-based violence and assess them against the identified benchmarks. Section 5 will summarize and compare the findings of these case studies and present some conclusions.

2.

EFFECTIVENESS ASSESSMENT FRAMEWORK

To be clear, we are not aiming to give a comprehensive analysis of the effectiveness of soft law adopted in the field of gender equality law in this chapter. It would be impossible to draw causal conclusions with regard to outcomes and impacts within the framework of such a limited – legal – study. Rather, the aim is to develop here below an assessment framework that allows for an initial mapping of the possible effects of soft law measures in this field and that could provide a basis for a more in-depth analysis of this issue in the future. 2.1

The Concept of Soft Law and its Functions

Any evaluation of the contribution of soft law to achieving gender equality in Europe requires the establishment of an assessment framework and benchmarks for testing its effectiveness. A first step in the development thereof is to clarify the concept of ‘soft law’ and distinguish the different purposes that soft law measures seek to achieve. In an earlier work, soft law has been defined as ‘rules of conduct that have not been attributed legally binding force as such, but which nevertheless may have certain (indirect) legal effects, and that are aimed at and may

370  Research handbook on soft law have practical effects’ (Senden 2004, p. 112). Importantly, this definition emphasizes that soft law acts seek to provide rules of behaviour and as such excludes acts or documents that are merely informational. Acts that contain such rules of conduct and that seek to influence behaviour can be categorized into three main groups on the basis of their function, as can be distilled from the framing of their goals, wording, substance, and addressees: – Pre-law, those acts that are adopted as a preparation for future legislation or that can be seen as an incentive for this; – Post-law, those acts that complement existing hard law/legislation and are geared towards facilitating or enhancing its implementation; – Para-law, those acts that are adopted as an alternative to legislation. The first category includes certain communications of the European Commission and its action plans and programmes,2 resolutions of the European Parliament and (European) Council conclusions, provided they are not of a purely informational nature. The acts fulfilling a post-law function are most often adopted by the Commission and can be subdivided into three further groups: interpretative acts, decisional acts, and guidance documents. Interpretative acts seek to clarify how the Commission understands particular provisions of EU primary and/or secondary law and how they should be applied in its view. Decisional acts are adopted in respect of areas or topics with respect to which the Commission holds a discretionary, implementing power; these acts then set out how the Commission intends to apply this power in particular, individual cases. More recently another category of post-legislative acts has developed, which concerns guidance documents that seek to steer the exercise of implementing powers of the Member States and national public authorities, rather than those of the European Commission. They do, however, often connect to Commission interpretative acts to the extent that they also contain interpretative elements, yet go a step further than that (Senden 2013, pp. 60–61; Van Dam 2020, pp. 25–26). Soft law acts that can be said to fulfil a para-law function and are intended to provide an alternative to legislation are the most far-reaching forms of soft law. Key to such acts is that they aim to lay down new rules of conduct, which are not necessarily linked to existing EU legislation or Treaty provisions and cannot be considered inherent to the existing legal framework, or at least are not limited to it. As such, they are intended not only to have practical effects, but also to create new legal obligations. The European Commission as well as the Council may adopt these. Steering acts falling within this category may be intended as temporary measures to be replaced by a hard legislative measure, or they may turn out to be only temporary in the event that they lack sufficient effect and are followed up by a hard legislative measure. Yet, such steering acts may also be intended as rather permanent alternatives to legislation and not necessarily to have the same effects as legislation; they may then be geared more towards establishing cooperation or coordinated action between the Member States than to the harmonization of national rules for instance. Recommendations mostly fit into the category of steering para-law acts, with Commission recommendations often being geared more towards harmonizing effects than Council ones. However, depending on the contents, codes of conduct and certain resolutions may also qualify as falling within this category. It should be noted that such acts may be adopted with a view to exerting influence through demonstrating potential avenues for regulation and providing a less restrictive path for States to take with respect to achieving certain policy goals. For the EU institutions, adopting soft law measures for which implementation is optional can

Soft law  371 be a means to circumvent Member State opposition to a legal measure or to bypass lack of EU competence. Such measures may have the effect of influencing more hesitant Member States to reconsider and eventually adopt policy measures or strategies in the policy area in question. In Sections 3 and 4, we will classify the different soft law acts under discussion from the perspective of the pre-, post-, and para-law functions. For those measures that have a post-law function, we will also clarify whether the type of soft law at play is interpretative, decisional or constitutes guidance. 2.2

Effectiveness and its Benchmarks

To assess the effectiveness of soft law, we need to be clear what we mean by ‘effectiveness’ and against what benchmarks or standards it can be evaluated. In its simplest form, effectiveness is ‘the quality of being able to bring about an effect’ (Savaşan 2019, p. 18). But any effectiveness assessment will be geared towards establishing a qualified or particular, intended effect. By establishing rules of conduct, soft law acts are geared towards producing practical effects and therewith some kind of change on the ground. By their very nature the incentive for such change must be triggered by some factor other than that of being legally binding, even if soft law acts are not necessarily devoid of legal effects. How then to qualify these practical effects further and what kind of change can or should be involved, taking into account the different functions that soft law acts may fulfil? Some approaches towards the assessment of effectiveness of EU law have centred on compliance, to the extent that four waves of compliance research can be distinguished.3 However, Snyder (1993, p. 19) goes beyond compliance, by holding that effectiveness is taken to mean the fact that ‘law matters: it has effect on political, economic and social life outside the law – that is, apart from simply the elaboration of legal doctrine’. This emphasis on the social context can be linked with the need to secure substantive effectiveness of EU law and policymaking or its output legitimacy (Scharpf 1999; Schmidt 2013). That is, it should ensure policy effectiveness and meet the needs of those concerned, in this context, citizens who have suffered from gender discrimination and inequality. As such, it is not so much about the effectiveness of the law in books but about the effect of law in practice.4 His focus thus goes beyond ‘classic’ implementation, which concentrates on the process of transposing EU law into national law, executing it within the domestic legal orders and securing compliance and enforcement by certain rules, duties, procedures, and mechanisms as well as by monitoring bodies and courts.5 But ‘implementation’ can also be understood in a broader sense, referring to ‘the art of deliberately achieving social change through law’ as argued by Snyder (1993, p. 26). This then encompasses not only actions of policymakers and legislators but also the various other institutional perspectives within the EU system that contribute to social change (Snyder 1993, p. 26). His conception of ‘compliance’ is also broader than commonly accepted: it is geared not only towards acting in accordance with certain accepted standards (Savaşan 2019, p. 18), but beyond that, towards considering the actions of national administrations and the ‘bureaucratic politics’ that comes with this relating to ‘ongoing negotiations, political and legal processes and organisational change’ (Snyder 1993, p. 26). Snyder’s approach thus puts emphasis on social, organizational, and behavioural change. Snyder’s approach can be connected to an approach more recently developed by Mitchell, which relies on the distinction of three indicators for effectiveness: outputs, outcomes, and impacts (Mitchell 2008). Outputs relate to laws and regulations that States adopt to implement

372  Research handbook on soft law and transform national law. Outcomes are geared towards bringing about behavioural change and impacts relate to policy change, such as securing more environmental quality or equality. Mitchell underlines that outputs in themselves are not enough to induce outcomes or impacts, which will be produced only when the measure is able to bring about both the desired behavioural as well as policy change (Savaşan 2019, p. 20). Combining the approaches of Snyder (1993) and Mitchell (2008) offers potential for building an assessment framework for the purposes of our inquiry. By its nature, soft law concerns rules of conduct that aim to produce ‘practical effects’, and its use seeks to orient behaviour towards bringing about a certain change, which may include or relate to impacts in the sense of policy change6 or changing policy principles.7 Also Snyder’s broader conceptualization of both the notions of implementation and compliance enables study of the practical effects of soft law rather than one focused merely on ‘classic’ legal implementation, compliance, and enforcement, the latter also involving an element of compulsion (Savaşan 2019, p. 18; cf. Snyder 1993, p. 27). In assessing the potential effectiveness of soft law in the field of gender equality, we will consider not only the classic legal implementation dimension (Senden 2004, pp. 321–447; Ştefan 2013, p. 367; Georgieva 2016, pp. 1–33; Láncos 2018, pp. 755–784),8 but also the potential effects resulting in behavioural, social, or organizational change as well as policy change. The distinction between different categories of soft law is relevant in this regard, as effects differ depending on the role played by soft law acts: those fulfilling a post-law function are in and of themselves geared towards supporting better legal implementation, compliance, and enforcement of other EU legal provisions, while those fulfilling a para-law function are standalone and geared towards bringing about certain outputs, outcomes, and impacts. It is already broadly recognized that, in bringing about such practical effects in terms of behavioural, organizational, social, and policy change but also certain legal effects, soft law relies more on persuasion, argumentation, and guidance than on enforcement.9 Beveridge (2012, p. 31) has asked what form successful persuasion might take: rational argument, peer pressure, measurable targets and benchmarks, or incentives. Some authors have drawn attention to the institutional dimension, stating that recommendations can consist of ‘a normative framework for future negotiations …, concretizing the duty of institutional cooperation …, producing a stand-still effect on the non-conforming conduct of a State or institutions’ (Ştefan 2013, p. 16; Andone and Greco 2018, p. 82). Others have emphasized more negatively that there is a likelihood of administrators ignoring steering instruments (such as recommendations) without accompanying hard law and that actors may not be aware of their existence (Hartlapp and Hofmann 2021, p. 137). These observations underscore the relevance of the institutional dimension in securing effectiveness. Drawing on these approaches, we propose a productive assessment framework for effectiveness based on four key questions: 1. Nature – What type(s) of soft law is/are at issue and what function does it fulfil? 2. Output – What is the ‘classic’ legal implementation, compliance, and enforcement effect? 3. Outcome – Has there been any behavioural, organizational, or social change on the part of the institutions/actors involved, e.g., awareness or publicity of any kind, stand-still effect on non-conforming behaviour, information campaigns, development of new rules or tools by other actors such as social partners, NGOs, employers, etc.?

Soft law  373 4. Impact – Has there been any identifiable effect in terms of realizing policy change and progress in the field that can be traced back to an EU soft law act?

3.

THE DEVELOPMENT OF EU GENDER EQUALITY LAW AND THE CONCOMITANT USE OF SOFT LAW

EU gender equality law is amongst the oldest areas of EU law, beginning with a single provision in the Treaty establishing the European Economic Community (EEC); its Article 119 (now Article 157 of the Treaty on the Functioning of the European Union (TFEU)) established the principle of equal pay between men and women for equal work and work of equal value, thus imposing on Member States the duty to ensure that this principle is applied. While this article originally aimed to address purely economic concerns, early case law of the Court of Justice of the EU (CJEU) recognized that it pursues also a more important, social aim, as well as expressing a fundamental right (C-43/75 Defrenne II, paras 8–12; C-149/77 Defrenne III, paras 25–27). Since these early beginnings, the EU gender equality acquis has impressively expanded. 3.1

EU Gender Equality ‘Hard’ Law

The Treaty on European Union (TEU), the TFEU, as well as the Charter of Fundamental Rights of the EU (CFEU), all contain provisions relevant for gender equality. Article 2 TEU states that equality between women and men is one of the fundamental values on which the Union is based, while Article 3(3) TEU proclaims the promotion of such equality as an essential task of the EU. The foundation for the EU taking action in the area of sex discrimination is Article 19(1) TFEU. Articles 8 and 10 TFEU contain important mainstreaming obligations for the EU with respect to gender equality (Article 8) and other grounds of discrimination (Article 10). Article 21 CFEU prohibits ‘any discrimination based on any ground’, including sex. Article 23 CFEU contains a specific obligation to ensure equality between women and men in all areas, going beyond equal pay and employment, whilst simultaneously providing for the possibility of positive action measures favouring the under-represented sex. Article 33 CFEU provides for the legal, social, and economic protection of the family and introduces rights with respect to maternity, as well as maternity and parental leave, thus addressing the reconciliation of work and family life. Beyond these treaty provisions, there is a multitude of directives regulating the prohibition of sex discrimination. These address issues such as equal pay and employment (Directive 2006/54/EC), social security (Directive 79/7/EEC), self-employment (Directive 2010/41/ EU), burden of proof in sex discrimination cases (Directive 97/80/EC), pregnant workers (Directive 92/85/EEC), goods and services (Directive 2004/113/EC), and parental leave (Directive 2010/18/EU). The latter has been replaced by the Work-Life Balance Directive (2019). The Horizontal Directive, initially proposed in 2008 (European Commission 2008), is now to be picked up again. This directive is intended to address the fragmented and hierarchical nature of EU equality law, where different legal instruments protect various grounds, with not all grounds being protected to the same extent due to important differences in scope. Recently, political agreement has been reached on the gender-balanced boards directive, proposed already by the Commission in 2012 (European Commission 2022b). Furthermore, the

374  Research handbook on soft law Commission has tabled two directive proposals in the last few years – on equal pay transparency in March 2021 (European Commission 2021), and on gender-based violence on 8 March 2022 (European Commission 2022a).10 There is also a significant body of CJEU case law, which has played an important role in developing EU law in the area of equal treatment between women and men and in interpreting EU equality law. For example, the Court’s case law played an important part in developing key concepts of gender equality law, including direct (e.g., C-177/88 Dekker; C-132/92 Birds Eye Walls) and indirect discrimination (e.g., C-170/84 Bilka; C-167/97 Seymour) as well as effective judicial protection (e.g., C-14/83 Von Colson; C-222/84 Johnston; C-271/91 Marshall II). 3.2

EU Gender Equality ‘Soft’ Law

The legal advances thus made should not be viewed in isolation from developments taking place on the policy level. There is a body of ‘tertiary law’ in these domains, the EU institutions having adopted a wide range of soft law acts as well, often either in combination with or in advance of hard law. EU soft law acts need to be distinguished by their source or authors and the objective or function they serve. Most importantly, these include (European) Council conclusions, Council of Ministers recommendations, Commission communications, (action) plans, guidelines, recommendations, and codes of conduct, and European Parliament resolutions and guidelines. Relevant Commission instruments in this category include: the Guidelines on the Test-Achats case (European Commission 2012a), the Communication on work-life balance (European Commission 2017a), the Communications on the Gender Equality Strategy (GES), most recently that of 2020–2025 (European Commission 2020a), and the Recommendation on Equal Pay Transparency (European Commission 2014). Council conclusions can also be seen as a form of soft law and some of these have specifically concerned gender equality, such as its Conclusions on violence against women and girls (Council 2014) and on gender-equal economies in the EU (Council 2019). Other conclusions have also concerned institutional mechanisms highlighting the mainstreaming principle within all of its activities as a specific goal (Council 2013) and most recently on mainstreaming gender equality in the EU budget (European Council 2020; Court of Auditors 2021). Some examples from the European Parliament include guidelines on gender-neutral language in the European Parliament (European Parliament 2018a), and resolutions on gender equality in EU trade agreements (European Parliament 2018b) and gender mainstreaming in the work of the European Parliament (European Parliament 2016). Sometimes soft law acts are adopted jointly by EU institutions, such as the EU’s new Action Plan on Gender Equality and Women’s Empowerment in External Action 2021–2025 (GAP III), which was launched by the European Commission and the High Representative for Foreign Affairs and Security Policy (European Commission 2020b). We can categorize these soft law acts according to whether they serve as preparatory or incentivizing measures for future legislation (pre-law), are aimed at facilitating or enhancing implementation of an existing hard law (post-law) or are adopted in place of legislation (para-law). For example, the Test-Achats Guidelines (2012) fall into the post-law category of guidance documents that have been adopted with the aim of facilitating Member State compliance at the national level with respect to the application of the Goods and Services Directive (2004). The Commission’s Communication on work-life balance (2017a) falls into the pre-law category. Indeed, the Commission mentions its intention of adopting a legislative proposal for

Soft law  375 a directive on work-life balance in the communication (European Commission 2017a). Half a decade later, this directive has now been adopted. Similarly, the Commission’s (2020a) newest Gender Equality Strategy for 2020–2025 also falls within this category and has already acted as a herald of another legislative proposal – that of a directive on combating violence against women and domestic violence (see European Commission 2022a). Council conclusions are often also of a pre-law nature but contain para-law elements too. For example, the Council conclusions on gender-equal economies (2019) appeal to the Commission and the Member States to enhance the promotion of gender equality in relation to the economy with specific suggestions for how to achieve this. Thus, they can be seen as incentives for future legislation. The same can be said of the European Parliament’s resolutions on gender equality in EU trade agreements (2018b) and on gender mainstreaming in the work of the European Parliament (2016). The guidelines on gender-neutral language in the European Parliament (2018a) are of an internal nature only and thus are only geared towards bringing about change within the European Parliament. The use of soft law in its three forms is therefore very much part and parcel of the EU’s approach towards enhancing gender equality in the Member States.

4.

CASE STUDIES

Having identified our assessment framework and provided a general introduction into the use of soft law in EU gender equality law, we will test our framework by examining the effectiveness of soft law in three specific areas of gender equality: equal pay, gender-balanced decision-making, and gender-based violence. In doing so, we will focus on some core soft law acts adopted in these three key areas taking a historical-legal and empirical perspective, drawing not only on academic literature, European Institute for Gender Equality (EIGE) and Commission reports, but also on country reports on the implementation of the EU gender equality acquis in the EU Member States, which have been produced under the auspices of the European Equality Law Network,11 as well as its annual comparative analyses and relevant thematic reports. 4.1

Equal Pay for Men and Women

As seen, the right to equal pay between women and men for equal work and work of equal value was enshrined already in Article 119 EEC. Nevertheless, securing this right has remained a key problem, which is why in 2014 the Commission adopted the Pay Transparency Recommendation. Yet, in March 2021, the European Commission (2021) put forward a proposal for legislation on pay transparency and enforcement mechanisms, which can be taken as a sign that the Recommendation has not delivered sufficient results. The interpretation of Article 119 EEC led to the famous Defrenne-saga before the European Court of Justice in the 1970s (cases C-80/70 Defrenne I, C-43/75 Defrenne II, and C-149/77 Defrenne III). These cases concerned the scope of the equal pay principle, what constitutes equal work, and the horizontal effect of Article 119. This incited the adoption of a Directive on the implementation of the equal pay principle in 1975, later integrated into the recast directive in 2006 (Directive 2006/54/EC). Despite this long-standing legal framework, complemented by many other equal pay cases decided by the CJEU,12 the gender pay gap remained a huge

376  Research handbook on soft law concern, as recognized by all EU decision-making institutions. In 2007, the Commission thus concluded that while direct pay discrimination for the exact same work had become rare, the existing legal framework had been less effective in ensuring implementation of the principle of equal pay for work of equal value, indicating obscure pay structures and lack of information and awareness as key obstacles to achieving this (European Commission 2007). In 2008 and 2012, the European Parliament adopted resolutions on equal pay between men and women with recommendations on how to better implement the principle of equal pay, which also included the introduction of wage transparency measures and gender-neutral job evaluation and classification systems (see European Parliament 2008, 2012). In 2010, the Council called upon the Member States to put in place measures that promote pay transparency and gender-neutral job evaluation and classification (Council 2010a). Subsequent Commission documents referred to the need to close the gender pay gap (European Commission 2013a), emphasizing again the need to overcome the lack of transparency in pay systems, the lack of legal certainty on the concept of work of equal value and procedural obstacles (European Commission 2013b). In 2014, these institutional concerns culminated in the adoption of the Commission’s Recommendation on Equal Pay Transparency, based on Article 292 TFEU (European Commission 2014). If we consider only the stated goal of the Recommendation, we would classify it as a post-law act and one that seeks to enhance ‘classic’ legal implementation of the long-existing hard law EU equal pay principle. Yet, when considering the substance of the Recommendation, it is clear that it also formulates new, specific rules of conduct that are not already contained in the underlying hard law. It does so by presenting a toolbox of measures that can assist Member States in taking a tailor-made approach, and via them also assist employees, employers, and social partners (European Commission 2014, recitals 14–15). To this end, it sets out four specific tools/measures, of which Member States are encouraged to implement at least one so as to enhance wage transparency: entitlement to request pay information, company reporting, pay audits, and equal pay collective bargaining. As such, one can say that the Recommendation also fulfils a para-law function, even though Member States would be able to choose the one(s) that would best fit their national, specific circumstances. Clearly this approach implies a focus on bringing about in fact some kind of behavioural, organizational, social, and policy change (outcomes and impacts), beyond simply improved legal implementation of the equal pay principle (outputs). When turning to the question of its actual effectiveness in these terms, Benedi Lahuerta and Zbyszewska (2018, p. 172) have observed that the Recommendation constitutes ‘a soft law instrument [that] leaves MS plenty of flexibility to “cherry pick” measures, which may vary widely in terms of effectiveness and enforceability’. The Commission evaluated the Recommendation twice, in 2017 and in 2020. In its first report, it concluded that the Recommendation has not achieved the objective of more effective implementation because of its non-binding nature (European Commission 2017b, p. 3). Only 11 Member States had some legislation on pay transparency in place by then, and only six of them had either adopted new measures or improved existing measures pursuant to the Recommendation.13 While most of them included one of the four proposed measures, only five had included more than one and only Sweden had all four measures enshrined in its national legal framework (European Commission 2017b, p. 3). In a third of the Member States such measures were entirely absent. The Commission’s overall conclusion was that only a small group of Member States had thus used the opportunity of the Recommendation to review their national legal systems for

Soft law  377 strengthening the principle of equal pay by introducing pay transparency measures, either as suggested in the Recommendation or otherwise. This contributed to the Commission’s Action Plan for 2017–2019, which stated that the Commission would assess the opportunity for amendments to the recast Directive with the aim, in particular, of improving pay transparency (European Commission 2017c). In 2020, the Commission published the results of the REFIT14 evaluation of the Recast Directive as well as – again – of the Recommendation and concluded that although seven Member States maintained measures already in place before 2014 and that six others had amended or adopted transparency measures after the adoption of the Recommendation, very few new measures had actually been taken after 2017 (European Commission 2020c, pp. 14–15).15 In terms of outputs, we can thus conclude that the effectiveness of the Recommendation has been very limited. Importantly, however, the 2017 report also addressed explicitly the ‘practical application and impact’ of pay transparency measures at the national level (European Commission 2017b, p. 10). Potentially, when such measures are implemented in the Member States, there is some evidence that they can lead to beneficial outcomes and have impacts. The intervention logic the Commission has applied in this regard is also based on an analysis of outputs, outcomes, and impacts (European Commission 2017b, p. 88). Overall, the Commission conclusion on the basis of its own evaluation report as well as the one produced by the European Equality Law Network (Foubert 2017), was that the right to equal pay is neither adequately applied nor enforced in practice and that in many Member States pay transparency is still lacking. In view of this, it is no surprise that the Commission came up with a proposal for a directive to enhance pay transparency and enforcement measures in 2021 (European Commission 2021). The Commission’s proposal is much more detailed and forceful on the measures to be adopted by the Member States than those contained in the Recommendation, including information rights, a pay evaluation duty, and a reporting duty (Senden and Hesdahl 2022). It also entails more duties for a variety of actors, such as employers, social partners, works councils, labour inspectorates, equality bodies, and monitoring bodies. It clearly seeks to enhance institutional commitment to and ownership of the problem of equal pay and to rely not only on an individual rights-based approach but to tackle it more collectively. The use of soft law by the European Parliament, the Commission, and the Council has contributed to a dialogue on the problem of equal pay for men and women. This process can be seen to have worked as a booster for bringing about organizational change and laying a foundation for proposing more stringent pay transparency and enforcement measures. However, this evaluation can be doubted in respect of the Recommendation: substance-wise it provided only little guidance for a better ‘classic’ legal implementation of the EU equal pay rules, as its provisions were not very specific, and the toolbox it offered was taken up by only a few Member States. Given the long-standing position of the equal pay principle in EU law, and the lack of any political controversy on its very legal recognition, one can thus say that adopting a recommendation rather than a directive as now proposed delayed the principle’s effective implementation. 4.2

Balanced Participation of Women and Men in Decision-making16

The second case study concerns the EU’s approach to enhancing balanced participation of women and men in decision-making processes and bodies. There is a persistent and quantifiable gender gap in this regard. Yet, quite differently from the case of equal pay, it has

378  Research handbook on soft law taken a while for the EU to take any action. Council recommendation 84/635/EEC (1984), recommending that Member States take steps to ensure that positive action includes, as far as possible, measures having a bearing on active participation by women in decision-making bodies (Council 1984, pp. 34–35), was a first step. But it was not until the 1990s that the European Parliament called on the Member States to take specific action in this area (European Parliament 1994). In 1996 the Council adopted a Recommendation on the balanced participation of women and men in the decision-making process (Council 1996). It sought to incite the Member States to develop a hard and/or soft law strategy to bring about more gender equality in decision-making, other provisions of the Recommendation making clear that this recommendation was geared towards the public as well as the private sector, and including associations and organizations at all levels of society. It focused on how to change attitudes and behaviour and mobilize all the actors in economic and social life to achieve equal opportunities, how to improve knowledge, and how to facilitate women’s participation in politics and in the public and private sectors (see European Commission 2000, pp. 4, 9). To this end, the Recommendation contained provisions for better data collection, on combating stereotypes in education and training, on public information campaigns and more awareness-raising generally, the promotion of good practice, equality plans, and positive action measures. In 2000, the Commission evaluated the 1996 Council Recommendation. The Commission concluded that ‘[e]fforts and results vary considerably in the Member States of the European Union and the European institutions’ (European Commission 2000, p. 3) and that the results regarding comprehensive, integrated strategies strived after in the Member States did not correspond to the Council’s expectations, even if all Member States had adopted enforcement structures (European Commission 2000, pp. 10–11). While the Recommendation thus had some impact on Member States policies, it failed to achieve gender balance in decision-making positions (European Commission 2000, p. 21). In subsequent years, the Commission identified equality in decision-making as one of the priority areas for action (European Commission 2010a). In 2015, the Council adopted specific Conclusions on ‘[e]quality between women and men in the field of decision-making’, calling upon the Commission and the Member States, to ‘consider a broad range of different measures, legislative and/or non-legislative, voluntary or binding, as well as the exchange of good practice, with a view to improving the gender balance in decision-making bodies in all areas’ (Council 2015, point 24). So, the Council advocated both hard and soft law outputs to be brought about in the Member States, outcomes in terms of behavioural, organizational, and social change of a variety of actors involved, as well as impacts in terms of policy change/ effectiveness. However, while the Conclusions clearly emphasize the responsibility of the private sector and companies to bring about the required change, they do not make any mention of the proposal for a directive that the Commission had already put forward in 2012 on the promotion of gender-balanced company boards, which was blocked in the Council until mid-2022, when political agreement was finally reached on its adoption (European Commission 2022b). So, for a decade, the Council itself has remained very reluctant to pass any legislation that would force the Member States to take any more stringent actions to promote gender-balanced company boards.17 The 1996 Council Recommendation can be typified as a soft law act fulfilling a para-law function and geared towards producing outputs (national comprehensive and integrated strategies to promote women’s position in decision-making), outcomes (behavioural, organiza-

Soft law  379 tional, and social change), as well as impacts (policy change/effectiveness in terms of a higher number of women in decision-making bodies). While it made some contribution in all these respects, it has clearly not been sufficient. Having said that, its adoption has contributed to a certain institutional dynamic and interaction on tackling this problem at various levels, and in both the public and private sectors, and it can also be seen as a stimulus for keeping the issue on the political and institutional agenda in later years. However, the Council’s approach over the past decade appears paradoxical or at least incoherent; while the 2015 Council Conclusions (Council 2015) can be seen as fulfilling a para-law function, seeking to steer Member State as well as EU and private actors to bring about certain outputs, outcomes, and impacts, the Council refrained from linking this to the 2012 proposal for a directive with similar goals in relation to the private sphere of decision-making (see European Commission 2012b). In that sense, the Council conclusions (2015) have not been used expressly as a booster for the adoption of the measures proposed by the Commission in this field or for pushing the adoption of the proposed directive. Nonetheless, the proposal has been put on the political agenda by different Council presidencies, keeping the debate on it alive and leading to the recent agreement. From the perspective of outcomes and impacts, there is clearly still a need for hard law action in quite a number of Member States, as there is evidence that hard law approaches have been most effective in increasing the number of women on corporate boards – as has been shown to be the case for several countries that moved from soft law and self-regulatory approaches to hard law ones (Senden 2014; Senden and Kruisinga 2018). It is not clear to what extent the 2015 Council Conclusions or other actions on the part of the Commission have boosted such shifts in approach, as comparative research has revealed that the regulatory choices made in this field – and the decision to adopt any rules or policies at all – can be traced back to a number of other key factors, including political factors and context, the protection of property rights, the understanding and scope of the concept of equality, market features, cultural factors, the role of women’s organizations and female leaders/ politicians and the role of the media (Senden 2014; Senden and Kruisinga 2018, pp. 87–94). Because of these manifold reasons for lacking sufficient political commitment to bring about change by way of hard legal action, it also cannot be concluded that soft law in this area has functioned as a brake. Rather one could say that the use of soft law has been part and parcel of a long-standing institutional dynamic in this field, which also contributed to the Commission’s Directive proposal and the subsequent debate as to the appropriateness of soft/self-regulatory approaches vis-à-vis hard law ones. The Member States’ approaches range from very soft regulations to very hard quotas, but in recent decades an increasing number of States have been moving towards more hard law approaches. It has also become clear that it is only in very advanced equality States, such as Sweden and Finland, that the softer approaches can lead to satisfactory results in terms of realizing the intended numerical impact. These shifting national approaches are the basis of the recent agreement on the Directive proposal, thereby implying an acknowledgement that soft law approaches in themselves fail to do the trick when it comes to the desired policy change. 4.3

Gender-based Violence

The third case study concerns gender-based violence, in respect of which a first EU legislative instrument was proposed in March 2022 (European Commission 2022a). Until recently, only some specific aspects of gender-based violence had been addressed legally at EU level (e.g.,

380  Research handbook on soft law through directives on harassment and trafficking). One of the reasons for this was the EU’s concentration on employment and the labour market. Sexual harassment in the workplace was one of the issues that received a lot of attention and EU action early on. Both the Commission and the Council adopted resolutions, and the Commission also issued a recommendation on the protection of the dignity of women and men at work (European Commission 1991) and a code of practice on measures to combat sexual harassment, which called on Member States to adopt laws in respect of employment rights (Roggeband 2021, p. 354). The recommendation recommends that ‘Member States take action to promote awareness that conduct of a sexual nature, or other conduct based on sex affecting the dignity of women and men at work … is unacceptable’ (European Commission 1991, Article 1) and to ‘implement the Commission’s code of practice on the protection of the dignity of women and men at work’ (European Commission 1991, Article 2) in the public sector whilst encouraging its use in the private sector. At the time, there was still a lack of political will for EU legislation in this area, so the Commission’s recommendation was mainly fulfilling a para-law function, especially in respect to the code of practice that it encouraged the Member States to implement. Certainly, the recommendation and code of practice aimed to bring about behavioural and organizational changes in relation to the workplace and leading to social and policy changes (impact/ outcome). These measures, while considered to have been weak in several substantive respects, did bring about the adoption of national legislative measures on sexual harassment during the late 1990s in several Member States (Zippel 2006, p. 103). However, in its evaluation report on the recommendation in 1996, the Commission concluded that neither the recommendation nor the code of practice had ‘led to the adoption of sufficient measures … to ensure a working environment where sexual harassment can be effectively prevented and combated’ (European Commission 1996, p. 4). Thus, the Commission determined that the issue needed to be addressed at the European level through a binding measure (European Commission 1996). In 2002, Directive 76/207/EEC (1976) was amended to include the first legal provision on sexual harassment in an employment context. This move towards hard law to address sexual harassment could be taken as a sign that at the very least the policy measures adopted in this area by the Council and the Commission in the 1990s ended up bringing about enough awareness and political will for this move towards binding legislation to be possible. In that sense, beyond its initial para-law function, the soft law measures adopted in this field can be said to have also acted as pre-law. At the same time, the early to mid-1990s also saw developments concerning sexual harassment at work at the international level (United Nations 1995). Thus, the influence of the relevant EU soft law must not be taken to have been the only factor at play on the road to hard law. Gender-based violence as a broader issue only gained more specific attention from the EU institutions in the 2010s. The adoption of the Women’s Charter saw the Commission committing itself to adopting ‘a comprehensive and effective policy framework to combat gender-based violence’ (European Commission 2010b, p. 4). Moreover, the Council has during its various presidencies issued recommendations and other policy documents on the subject.18 In 2010 and 2011, the Council Conclusions on the Eradication of Violence Against Women in the European Union and the Council Conclusions on the European Pact for Gender Equality for the period 2011–2020 provided further guidance on measures to effectively address violence against women and reaffirmed the EU’s commitment to eradicating all types of gender-based violence (Council 2010b, 2011). Moreover, the Commission made combating

Soft law  381 gender-based violence one of its key priorities in its Strategic Engagement for Gender Equality of 2016–2019 (European Commission 2016, p. 9) and again in its Gender Equality Strategy 2020-2025 (European Commission 2020a). Despite repeated calls from the European Parliament in particular, there was no comprehensive legislation on gender-based violence for some time, resulting in different levels of protection and prevention across the Member States. Of great importance in this respect is the adoption of the Council of Europe’s Istanbul Convention (Council of Europe 2011), which since coming into force in 2011 constitutes the only legally binding instrument on gender-based violence in Europe and contains specific obligations for Member States with respect to tackling it. While many of the EU Member States have ratified the Convention and the EU has signed it, accession of the EU to the Istanbul Convention has been hindered by questions concerning the legal requirements relating to the EU’s conclusion of mixed international agreements.19 There are also six European countries that have signed the Convention but not yet ratified it.20 The GES 2020–2025 retains accession to the Istanbul Convention as a goal, as well as the Commission’s ambition to propose comprehensive legislation on gender-based violence to complement it. The main reason for the lack of EU-level legislation on gender-based violence has been the absence of EU competences. The adoption of the Lisbon Treaty widened the scope for action in that the Declaration on Article 8 of the TFEU (which lays down the gender mainstreaming obligation of the Union) specifically refers to combating domestic violence as being part of the efforts to be taken to enhance equality between women and men.21 In December 2019, the Presidency of von der Leyen provided new fuel for this, including a comprehensive directive on gender-based violence. In March 2022, the proposal for such a directive was published (European Commission 2022a). Given that at the time legislative measures were being considered to fall outside the EU’s competence, the adopted soft law acts must be understood as being adopted as para-law measures. As for their effectiveness, the Council conclusions referred to above (Council 2010b, 2011) called specifically for (soft and hard) legal measures in the form of national strategies and laws on gender-based violence (outputs), as well as for organizational and social and policy changes that would make prevention and protection of women from violence more effective (outcomes/impacts). Similarly, the European Parliament’s resolutions on the elimination of violence against women (European Parliament 2009) and on combating cyber violence (European Parliament 2021) can be said to act as pre-law measures, calling repeatedly for the Commission to adopt comprehensive legislation on gender-based violence and making specific suggestions for both the EU and the Member States to tackle the different dimensions of violence against women in all its forms. Here, again, both hard and soft legislative measures or outputs, as well as social, policy, and behavioural changes are ultimately advocated for (impacts and outcomes), with the European Parliament for example also calling for awareness raising of women’s rights and the need for respect of these (European Parliament 2014, point 5). It can be said that at both the EU and Member State levels, measures have been taken over the years to address violence against women and domestic violence, signalling that the above-discussed soft law measures did contribute to some extent to achieving progress towards better protection and prevention of gender-based violence. However, there are still serious gaps in the legal and policy framework on combating gender-based violence on both levels. The Impact Assessment Report of the Commission highlights that national jurisdictions

382  Research handbook on soft law have gaps in relation to prevention, protection, access to justice, victim support, and policy coordination, whilst at the EU level there remain significant gaps and shortcomings in this respect (European Commission 2022c, pp. 13, 20). Ultimately, the developments at EU and national levels were not sufficient to bring about the necessary outcomes, impacts or outputs, which is why the Commission followed the European Parliament’s recommendation of addressing the issue comprehensively at EU level in order to provide minimum standards of protection and rights among all Member States. It seems fair to say that soft law measures in this field were initially adopted as para-law measures that eventually turned into pre-law, in the sense that they have provided the stepping stones for the adoption of hard law (provided that the directive proposal is successful). Soft law thus acted as a tool to broaden policy making into a new domain, paving the way for hard legal measures to follow (Agustín 2013, pp. 89–115). However, many other factors also influenced this development, including the adoption of the Istanbul Convention and other international developments. Moreover, the impact of EU soft law on gender-based violence is perhaps most visible at the EU level rather than at the Member State level. Although not a brake, the role of EU soft law as a booster is therefore rather limited. At least at the EU level, however, it can be said to have had a role to play in the path towards comprehensive EU legislation ensuring a minimum level of protection for women and girls within the EU. This then also shows a certain level of effectiveness of soft law measures to the extent that they enable or encourage EU institutions to move forward in legislating on such matters.

5. CONCLUSION We have sought to develop a framework as a first step for testing the potential effectiveness of soft law and exploring its application in three important areas of gender equality. When comparing and contrasting the areas studied more closely, there are three findings that it is important to highlight. First, all areas reveal the use of a huge variety of soft law instruments over a long period of time, adopted by all EU decision-making institutions. As such, soft law has served as a connector or a communication device in the dialogue on the issues involved. Its use has become a consistent part of the exercise of the powers of the EU institutions, a kind of constitutional convention, without this being established as such in the EU treaties and enforceable in any way (Gardiner 2022). A second observation is that its effectiveness is generally rather limited. In all of the areas analysed, some effects could be identified at the levels of outputs, outcomes and impacts, but all evaluation studies available also reveal that these were not sufficient to bring about the desired ‘classic’ legal implementation, nor the behavioural, organizational, social, and necessary policy changes. In all areas, there has thus been a move towards the adoption or proposal of EU hard law. Thirdly, an interesting difference between the first case study and the other two has emerged. While equal pay for women and men has been a long-standing principle and even fundamental right under EU law since the 1960s, the equal representation of women and men in decision-making and the general prohibition of gender-based violence lack a similarly firm status and anchoring in primary EU law. As a result, the competence issue has stood in the way of legislating in these areas and the use of soft law has consequently been that of a stepping

Soft law  383 stone. As such, in these areas soft law has served as a booster, together with developments at the international level, for employing a more purposeful interpretation of the legal basis provisions that can potentially be used for adopting legislation in these fields. Yet, in the area of equal pay one can say that more stringent equal pay transparency and enforcement measures were already long overdue, and that the Pay Transparency Recommendation (2014) can be considered to have actually further delayed the realization of equal pay for women and men.

NOTES 1. Note that when using the term ‘act’ here, we are not referring to legal acts in the sense of hard law. 2. The distinction between proposed rules of conduct and provisions of an informational nature is not always easy to make. 3. First wave, e.g., Siedentopf and Ziller (1988, p. 3); second wave, e.g., Mastenbroek (2005); third wave, Steunenberg (2010) and Toshkov (2011); fourth wave, Falkner et al. (2007). 4. There is however an increasing line of empirical research into how procedures work in practice and may obstruct effective judicial protection. 5. See e.g., Ballesteros et al. (2013). See also Treib (2014) for a critical assessment of existing doctrinal approaches to the topic of implementation. 6. See, in a similar sense, Andone and Greco (2018). 7. As referenced by Jacobsson (2004); Andone and Greco (2018). 8. Cf. for such approaches, Senden (2004); Ştefan (2013); Georgieva (2016); Láncos (2018). 9. See Andone and Greco (2018, p. 83), under reference to Ştefan (2013, p. 2). See also di Robilant (2006); Terpan (2015, pp. 88–89). 10. This proposal draws from a thematic report prepared for the Commission under the auspices of the European Equality Law Network, which compared the way the EU and EEA countries criminalize different types of gender-based violence in their national laws. For the report, see De Vido and Sosa (2021). 11. See https://​www​.equalitylaw​.eu. 12. See, for an overview, European Commission (2019). 13. These Member States were Austria, Belgium, Denmark, Finland, France, Germany, Ireland, Italy, Lithuania, and Sweden, as well as the United Kingdom (still an EU Member State at that time). 14. The Regulatory Fitness and Performance Programme, or REFIT, is the Commission’s programme for ensuring that EU legislation remains fit for purpose and delivers the results intended by EU lawmakers. 15. Some of these concerned States that already had measures in place before, such as France and Portugal. 16. This section draws partly on Böök et al. (2020). 17. For detailed accounts of what the Directive proposal entails and the variety of approaches to the issue in the Member States, see Senden and Kruisinga (2018). 18. See, e.g., Council (2010c). 19. These questions have recently been answered by the CJEU (Grand Chamber) in Opinion 1/19 of the Court (C-1/19 Convention d’Istanbul). 20. See Council of Europe (2022) for a chart of signatures and ratifications of the Istanbul Convention. 21. See the declarations annexed to the final act of the intergovernmental conference which adopted the Treaty of Lisbon (European Union 2007).

REFERENCES Agustín, L. R., (2013). Gender-based violence and the framing of equality policies. In: Gender equality, intersectionality, and diversity in Europe. New York, N.Y.: Palgrave Macmillan. pp. 89–115.

384  Research handbook on soft law Andone, C. and Greco, S., (2018). Evading the burden of proof in European Union soft law instruments: the case of Commission recommendations. International Journal for the Semiotics of Law. 31(1), 79–99. Ballesteros, M., Mehdi, R., Eliantonio, M. and Petrovic, D., (2013). Tools for ensuring implementation and application of EU law and evaluation of their effectiveness [online]. Brussels: Policy Department for Citizens’ Rights and Constitutional Affairs, European Parliament. [Viewed 26 October 2022]. Available from: https://​www​.europarl​.europa​.eu/​RegData/​etudes/​etudes/​join/​2013/​493014/​IPOL​ -JURI​_ET(2013)493014​_EN​.pdf. Benedi Lahuerta, S. and Zbyszewska, A., (2018). EU equality law after a decade of austerity: on the Social Pillar and its transformative potential. International Journal of Discrimination and the Law. 18(2-3), 163–192. Beveridge, F., (2012). ‘Going soft’? Analysing the contribution of soft and hard measures in EU gender law and policy. In: E. Lombardo and M. Forest, eds. The Europeanization of gender equality policies: a discursive-sociological approach. London: Palgrave Macmillan. pp. 28–48. Böök, B. and Senden, L., (Forthcoming). The puzzle of algorithmic discrimination in European equality law: putting together some practical, legal and conceptual pieces. In: Research handbook on European antidiscrimination law. Cheltenham, UK: Edward Elgar Publishing. Böök, B., Hoof, F. van, Senden, L. and Timmer, A., (2020). Gendering the COVID-19 crisis: a mapping of its impact and call for action in light of EU gender equality law and policy. European Equality Law Review. 2, 22–44. Case C-80/70 Gabrielle Defrenne v Belgian State ECLI:EU:C:1971:55. (Defrenne I) Case C-43/75 Gabrielle Defrenne v Société anonyme belge de navigation aérienne Sabena ECLI:EU:C:1976:56. (Defrenne II) Case C-149/77 Gabrielle Defrenne v Société anonyme belge de navigation aérienne Sabena ECLI:EU:C:1978:130. (Defrenne III) Case C-14/83 Sabine von Colson and Elisabeth Kamann v Land Nordrhein-Westfalen ECLI:EU:C:1984:153. (Von Colson) Case C-170/84 Bilka–Kaufhaus GmbH v Karin Weber von Hartz ECLI:EU:C:1986:204. (Bilka) Case C-222/84 Marguerite Johnston v Chief Constable of the Royal Ulster Constabulary ECLI:EU:C:1986:206. (Johnston) Case C-177/88 Elisabeth Johanna Pacifica Dekker v Stichting Vormingscentrum voor Jong Volwassenen (VJV-Centrum) Plus ECLI:EU:C:1990:383. (Dekker) Case C-271/91 M. Helen Marshall v Southampton and South-West Hampshire Area Health Authority ECLI:EU:C:1993:335. (Marshall II) Case C-132/92 Birds Eye Walls Ltd. v Friedel M. Roberts ECLI:EU:C:1993:868. Case C-167/97 Regina v Secretary of State for Employment, ex parte Nicole Seymour-Smith and Laura Perez ECLI:EU:C:1999:60. (Seymour) Case Opinion C-1/19 Opinion of the Court (Grand Chamber) of 6 October 2021 on Istanbul Convention ECLI:EU:C:2021:198. Charter of fundamental rights of the European Union (2016). Official Journal C202, 7 June, pp. 389–405. (CFEU) Consolidated version of the Treaty on European Union (2008). Official Journal C115, 9 May, pp. 13–45. (TEU) Consolidated version of the Treaty on the Functioning of the European Union (2012). Official Journal C326, 26 October, pp. 47–390. (TFEU) Council., (1984). Council recommendation 84/635/EEC on the promotion of positive action for women. Official Journal L331, p. 34. Council., (1996). Council recommendation 96/694/EC on the balanced participation of women and men in the decision-making process. Official Journal L319, p. 11. Council., (2010a). Council conclusions on strengthening the commitment and stepping up action to close the gender pay gap, and on the review of the implementation of the Beijing Platform for Action. Official Journal C345, p. 1. Council., (2010b). Council conclusions on the eradication of violence against women in the European Union. Employment and Social Policy Council meeting, Brussels, 8 March 2020 [online]. Brussels:

Soft law  385 Council of the European Union. [Viewed 28 October 2022]. Available from: https://​www​.consilium​ .europa​.eu/​uedocs/​cms​_data/​docs/​pressdata/​en/​lsa/​113226​.pdf. Council., (2010c). The Stockholm Programme — an open and secure Europe serving and protecting citizens. Official Journal C115, p. 1. Council., (2011). Council conclusions on the European Pact for gender equality for the period 2011–2020. Employment, Social Policy, Health and Consumer Affairs Council meeting, Brussels, 7 March 2011 [online]. Brussels: Council of the European Union. [Viewed 28 October 2022]. Available from: https://​www​.consilium​.europa​.eu/​uedocs/​cms​_data/​docs/​pressdata/​en/​lsa/​119628​.pdf. Council., (2013). Council conclusions on the effectiveness of institutional mechanisms for the advancement of women and gender equality. Employment, Social Policy, Health and Consumer Affairs Council meeting, Brussels, 9 and 10 December 2013 [online]. Brussels: Council of the European Union. [Viewed 28 October 2022]. Available from: https://​www​.consilium​.europa​.eu/​uedocs/​cms​ _data/​docs/​pressdata/​en/​lsa/​139978​.pdf. Council., (2014). Council conclusions – “preventing and combating all forms of violence against women and girls, including female genital mutilation”. Justice and Home Affairs Council meeting, Luxembourg, 5 and 6 June 2014 [online]. Brussels: Council of the European Union. [Viewed 28 October 2022]. Available from: https://​www​.consilium​.europa​.eu/​media/​28077/​143103​.pdf. Council., (2015). Equality between women and men in the field of decision-making - draft Council conclusions. Permanent Representatives Committee, Brussels, 2 December 2015 [online]. Brussels: Council of the European Union. [Viewed 28 October 2022]. Available from: https://​data​.consilium​ .europa​.eu/​doc/​document/​ST​-14325​-2015​-INIT/​en/​pdf. Council., (2019). Gender-equal economies in the EU: the way forward - draft Council conclusions. Presidency, Brussels, 22 November 2019 [online]. Brussels: Council of the European Union. [Viewed 28 October 2022]. Available from: https://​data​.consilium​.europa​.eu/​doc/​document/​ST​-14254​-2019​ -INIT/​en/​pdf. Council directive 76/207/EEC on the implementation of the principle of equal treatment for men and women as regards access to employment, vocational training and promotion, and working conditions (1976) Official Journal L039, p. 40. Council directive 79/7/EEC on the progressive implementation of the principle of equal treatment for men and women in matters of social security (1978) Official Journal L6, p. 24. Council directive 92/85/EEC on the introduction of measures to encourage improvements in the safety and health at work of pregnant workers and workers who have recently given birth or are breastfeeding (tenth individual directive within the meaning of Article 16 (1) of directive 89/391/EEC) (1992) Official Journal L348, p. 1. Council directive 97/80/EC on the burden of proof in cases of discrimination based on sex (1998) Official Journal L14, p. 6. Council directive 2004/113/EC implementing the principle of equal treatment between men and women in the access to and supply of goods and services (2004) Official Journal L373, p. 37. (Goods and Services Directive) Council directive 2010/18/EU implementing the revised framework agreement on parental leave concluded by BUSINESSEUROPE, UEAPME, CEEP and ETUC and repealing directive 96/34/EC (text with EEA relevance) (2010) Official Journal L68, p. 13. Council of Europe., (2011). Convention on preventing and combating violence against women and domestic violence (CETS No. 210) [online]. [Viewed 27 October 2022]. Available from https://​rm​.coe​ .int/​168008482e (Istanbul Convention) Council of Europe., (2022). Chart of signatures and ratifications of Treaty 210. Convention on preventing and combating violence against women and domestic violence [online]. CoE Treaty Office. [Viewed 31 October 2022]. Available from: https://​www​.coe​.int/​en/​web/​conventions/​full​-list​?module​ =​signatures​-by​-treaty​&​treatynum​=​210. Court of Auditors., (2021). Gender mainstreaming in the EU budget: time to turn words into action. Special Report No 10/2021 [online]. Luxembourg: European Court of Auditors. [Viewed 29 October 2022]. Available from: https://​www​.eca​.europa​.eu/​en/​Pages/​DocItem​.aspx​?did​=​58678. Dam, J. C. A. van, (2020). Guidance documents of the European Commission in the Dutch legal order. PhD Thesis. Leiden University, Leiden Law School.

386  Research handbook on soft law De Vido, S. and Sosa, L., (2021). Criminalisation of gender-based violence against women in European states, including ICT-facilitated violence [online]. Luxembourg: Publications Office of the European Union. [Viewed 29 October 2022]. Available from: doi: 10.2838/960650. Di Robilant, A., (2006). Genealogies of soft law. The American Journal of Comparative Law. 54(3), 499–554. EIGE., (2021). Gender equality index [online]. EIGE. [Viewed 30 September 2022]. Available from: https://​eige​.europa​.eu/​gender​-equality​-index/​2021. European Commission., (1991). Commission recommendation 92/131/EEC on the protection of the dignity of women and men at work. Official Journal L49, p. 1. European Commission., (1996). Consultation of management and labour on the prevention of sexual harassment at work (COM/96/373 final) [online]. Brussels: European Commission. [Viewed 29 October 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​PDF/​?uri​=​CELEX:​ 51996DC0373. European Commission., (2000). Report on the implementation of Council recommendation 96/694 on the balanced participation of women and men in the decision-making process (COM/2000/120 final) [online]. Brussels: European Commission. [Viewed 29 October 2022]. Available from: https://​eur​-lex​ .europa​.eu/​LexUriServ/​LexUriServ​.do​?uri​=​COM:​2000:​0120:​FIN:​EN:​PDF. European Commission., (2007). Tackling the pay gap between women and men (COM/2007/424 final) [online]. Brussels: European Commission. [Viewed 29 October 2022]. Available from: https://​eur​-lex​ .europa​.eu/​legal​-content/​EN/​TXT/​PDF/​?uri​=​CELEX:​52007DC0424​&​from​=​EN. European Commission., (2008). Proposal for a Council directive on implementing the principle of equal treatment between persons irrespective of religion or belief, disability, age or sexual orientation (COM/2008/426 final) [online]. Brussels: European Commission. [Viewed 29 October 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​PDF/​?uri​=​CELEX:​52008PC0426​&​ from​=​en. European Commission., (2010a). Strategy for equality between women and men 2010-2015 (COM/2010/491 final) [online]. Brussels: European Commission. [Viewed 29 October 2022]. Available from: https://​eur​-lex​.europa​.eu/​LexUriServ/​LexUriServ​.do​?uri​=​COM:​2010:​0491:​FIN:​EN:​ PDF. European Commission., (2010b). A strengthened commitment to equality between women and men. a women's charter (COM/2010/78 final) [online]. Brussels: European Commission. [Viewed 29 October 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​PDF/​?uri​=​CELEX:​ 52010DC0078​&​from​=​en. European Commission., (2012a). Guidelines on the application of Council directive 2004/113/EC to insurance, in the light of the judgment of the Court of Justice of the European Union in case C-236/09 (Test-Achats). Official Journal C11, p. 1. (Test-Achats Guidelines) European Commission., (2012b). Proposal for a directive on improving the gender balance among non-executive directors of companies listed on stock exchanges and related measures (COM/2012/614 final) [online]. Brussels: European Commission. [Viewed 31 October 2022]. Available from: https://​ eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​PDF/​?uri​=​CELEX:​52012PC0614​&​from​=​EN. European Commission., (2013a). Towards Social Investment for Growth and Cohesion – including implementing the European Social Fund 2014-2020 (COM/2013/83 final) [online]. Brussels: European Commission. [Viewed 29 October 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​ -content/​EN/​TXT/​PDF/​?uri​=​CELEX:​52013DC0083​&​from​=​EN. European Commission., (2013b). Report on the application of directive 2006/54/EC on the implementation of the principle of equal opportunities and equal treatment of men and women in matters of employment and occupation (recast) (COM/2013/861 final) [online]. Brussels: European Commission. [Viewed 29 October 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​ TXT/​PDF/​?uri​=​CELEX:​52013DC0861​&​from​=​EN. European Commission., (2014). Commission recommendation 2014/124/EU on strengthening the principle of equal pay between men and women through transparency. Official Journal L69, p. 112. (Pay Transparency Recommendation) European Commission., (2016). Strategic engagement for gender equality 2016-2019 [online]. Luxembourg: Publications Office of the European Union. [Viewed 29 October 2022]. Available from: doi: 10.2838/454429.

Soft law  387 European Commission., (2017a). An initiative to support work-life balance for working parents and carers (COM/2017/252 final) [online]. Brussels: European Commission. [Viewed 29 October 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​PDF/​?uri​=​CELEX:​52017DC0252​&​ from​=​EN. European Commission., (2017b). Report on the implementation of Commission recommendation on strengthening the principle of equal pay between men and women through transparency (COM/2017/671 final) [online]. Brussels: European Commission. [Viewed 29 October 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​PDF/​?uri​=​CELEX:​52017DC0671​&​ from​=​EN. European Commission., (2017c). EU Action Plan 2017-2019. Tackling the gender pay gap (COM/2017/678 final) [online]. Brussels: European Commission. [Viewed 29 October 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​?uri​=​CELEX:​52017DC0678. European Commission., (2019). Equal pay: overview of landmark case-law of the Court of Justice of the European Union [online]. Luxembourg: Publications Office of the European Union. [Viewed 29 October 2022]. Available from: https://​op​.europa​.eu/​en/​publication​-detail/​-/​publication/​8ecbe429​ -fb90​-11e9​-8c1f​-01aa75ed71a1. European Commission., (2020a). A union of equality: gender equality strategy 2020-2025 (COM/2020/152 final) [online]. Brussels: European Commission. [Viewed 29 October 2022]. Available from: https://​ eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​PDF/​?uri​=​CELEX:​52020DC0152​&​from​=​EN. European Commission., (2020b). EU Gender Action Plan (GAP) III – an ambitious agenda for gender equality and women’s empowerment in EU external action (JOIN/2020/17 final) [online]. Brussels: European Commission. [Viewed 29 October 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​ -content/​EN/​TXT/​PDF/​?uri​=​CELEX:​52020JC0017​&​from​=​EN. European Commission., (2020c). Evaluation of the relevant provisions in the directive 2006/54/EC implementing the treaty principle on ‘equal pay for equal work or work of equal value’ (SWD/2020/50 final) [online]. Brussels: European Commission. [Viewed 29 October 2022]. Available from: https://​ ec​.europa​.eu/​info/​law/​better​-regulation/​have​-your​-say/​initiatives/​1839​-Evaluation​-of​-the​-provisions​ -in​-the​-Directive​-2006​-54​-EC​-implementing​-the​-Treaty​-principle​-on​-equal​-pay​_en. European Commission., (2021). Proposal for a directive to strengthen the application of the principle of equal pay for equal work or work of equal value between men and women through pay transparency and enforcement mechanisms (COM/2021/93 final) [online]. Brussels: European Commission. [Viewed 29 October 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​PDF/​ ?uri​=​CELEX:​52021PC0093​&​from​=​EN. European Commission., (2022a). Proposal for a directive on combating violence against women and domestic violence (COM/2022/105 final) [online]. Strasbourg: European Commission. [Viewed 29 October 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​PDF/​?uri​=​CELEX:​ 52022PC0105​&​from​=​EN. European Commission., (2022b). Commission welcomes political agreement on Gender Balance on Corporate Boards. Press Release [online]. European Commission. [Viewed 29 October 2022]. Available from: https://​ec​.europa​.eu/​commission/​presscorner/​detail/​en/​ip​_22​_3478. European Commission., (2022c). Impact assessment report. Accompanying the document proposal for a directive on combating violence against women and domestic violence (SWD/2022/62 final). Strasbourg: European Commission. [Viewed 29 October 2022]. Available from: https://​ eur​ -lex​ .europa​.eu/​legal​-content/​EN/​TXT/​PDF/​?uri​=​CELEX:​52022SC0062​&​from​=​EN. European Council., (2020). Special meeting of the European Council (17, 18, 19, 20 and 21 July 2020) – conclusions (EUCO 10/20) [online]. Brussels: European Council. [Viewed 29 October 2022]. Available from: https://​www​.consilium​.europa​.eu/​media/​45109/​210720​-euco​-final​-conclusions​-en​ .pdf. European Parliament., (1994). Resolution of 28 February 1994 on women in decision-making bodies. Official Journal C61, p. 248. European Parliament., (2008). Resolution of 18 November 2008 with recommendations to the Commission on the application of the principle of equal pay for men and women (2008/2012(INI)). Official Journal C16E, p. 21. European Parliament., (2009). Resolution of 26 November 2009 on the elimination of violence against women. Official Journal C285E, p. 53.

388  Research handbook on soft law European Parliament., (2012). Resolution of 24 May 2012 with recommendations to the Commission on application of the principle of equal pay for male and female workers for equal work or work of equal value (2011/2285(INI)). Official Journal C264E, p. 75. European Parliament., (2014). Resolution of 25 February 2014 with recommendations to the Commission on combating violence against women (2013/2004(INL)). Official Journal C285, p. 2. European Parliament., (2016). Report on gender mainstreaming in the work of the European Parliament (2015/2230(INI)) [online]. Committee on Women’s Rights and Gender Equality. [Viewed 29 October 2022]. Available from: https://​www​.europarl​.europa​.eu/​doceo/​document/​A​-8​-2016​-0034​_EN​.html. European Parliament., (2018a). Gender-neutral language in the European Parliament [online]. High-Level Group on Gender Equality and Diversity. [Viewed 29 October 2022]. Available from: https://​www​.europarl​.europa​.eu/​cmsdata/​151780/​GNL​_Guidelines​_EN​.pdf. European Parliament., (2018b). Resolution of 13 March 2018 on gender equality in EU trade agreements (2017/2015(INI)). Official Journal C162, p. 9. European Parliament., (2021). Resolution of 14 December 2021 with recommendations to the Commission on combating gender-based violence: cyberviolence (2020/2035(INL)). Official Journal C251, p. 2. European Parliament and Council directive 2006/54/EC on the implementation of the principle of equal opportunities and equal treatment of men and women in matters of employment and occupation (recast) (2006) Official Journal L204, p. 23. European Parliament and Council directive 2010/41/EU on the application of the principle of equal treatment between men and women engaged in an activity in a self-employed capacity and repealing Council directive 86/613/EEC (2010) Official Journal L180, p. 1. European Parliament and Council directive (EU) 2019/1158 on work-life balance for parents and carers and repealing Council directive 2010/18/EU (2019) Official Journal L188, p. 79. (Work-Life Balance Directive) European Union., (2007). Declarations annexed to the final act of the intergovernmental conference which adopted the Treaty of Lisbon. Official Journal C326, p. 337. 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), 395–416. Foubert, P., (2017). The enforcement of the principle of equal pay for equal work or work of equal value: a legal analysis of the situation in the EU Member States, Iceland, Liechtenstein and Norway [online]. Luxembourg: Publications Office of the European Union. [Viewed 27 October 2022]. Available from: doi: 10.2838/672873. Gardiner, J., (2022). Hard Euroscepticism and constitutional conventions in the European Union – considering the role of the Court of Justice of the European Union in their protection and enforcement. Master thesis. Utrecht University. (on file with the authors) Georgieva, Z., (2016). The judicial reception of competition soft law in the Netherlands and the UK. European Competition Journal. 12(1), 54–86. Gerards, J. and Xenidis, R., (2021). Algorithmic discrimination in Europe: challenges and opportunities for gender equality and non-discrimination law [online]. Luxembourg: Publications Office of the European Union. [Viewed 27 October 2022]. Available from: doi: 10.2838/544956. Hartlapp, M. and Hofmann, A., (2021). The use of EU soft law by national courts and bureaucrats: how relation to hard law and policy maturity matter. West European Politics. 44(1), 134–154. Jacobsson, K., (2004). Between the deliberation and discipline: soft governance in EU employment policy. In: U. Mörth, ed. Soft law in governance and regulation. An interdisciplinary analysis. Cheltenham, UK: Edward Elgar. pp. 81–101. Láncos, P. L., (2018). A hard core under the soft shell: how binding is Union soft law for member states?. European Public Law. 24(4), 755–784. Mastenbroek, E., (2005). EU compliance: still a ‘black hole’?. Journal of European Public Policy. 12(6), 1103–1120. Mitchell, R. B., (2008). Compliance theory: compliance, effectiveness, and behaviour change in international environmental law. In: D. Bodansky, J. Brunnée and E. Hey, eds. Oxford handbook of international environmental law. Oxford: Oxford University Press. pp. 893–921.

Soft law  389 OECD., (2021). Caregiving in crisis: gender inequality in paid and unpaid work during Covid-19 [online]. Paris: Organisation for Economic Cooperation and Development (OECD). [Viewed 30 September 2022]. Available from: https://​www​.oecd​.org/​coronavirus/​policy​-responses/​caregiving​-in​ -crisis​-gender​-inequality​-in​-paid​-and​-unpaid​-work​-during​-covid​-19​-3555d164. Roggeband, C., (2021). Violence against women and gender-based violence. In: G. Abels, A. Krizsán, H. MacRae and A. van der Vleuten, eds. The Routledge handbook of gender and EU politics. Abingdon: Routledge. pp. 352–363. Savaşan, Z., (2019). Paris Climate Agreement: a deal for better compliance?. Cham: Springer. Scharpf, F., (1999). Governing in Europe: effective and democratic?. Oxford: Oxford University Press. Schmidt, V. A., (2013). Democracy and legitimacy in the European Union revisited: input, output and ‘throughput’. Political studies. 61(1), 2–22. Senden, L., (2004). Soft law in European Community law. London: Hart Publishing. Senden, L., (2013). Soft post-legislative rulemaking: a time for more stringent control. European Law Journal. 19(1), 57–75. Senden, L., (2014). The multiplicity of regulatory responses to remedy the gender imbalance on company boards. Utrecht Law Review. 10(5), 51–66. Senden, L. A. J. and Hesdahl, R., (2022). Gelijke beloning voor mannen en vrouwen: geen woorden meer maar daden. Nederlands tijdschrift voor Europees recht. 1-2. Senden, L. and Kruisinga, S., (2018). Gender-balanced company boards in Europe: a comparative analysis of the regulatory, policy and enforcement approaches in the EU and EEA Member States [online]. Luxembourg: Publications Office of the European Union. [Viewed 28 October 2022]. Available from: 10.2838/964451 Siedentopf, H. and Ziller, J., eds., (1988). Making European policies work: the implementation of community legislation in the member states. London: Sage. Snyder, F., (1993). The effectiveness of European Community law: institutions, processes, tools and techniques. The Modern Law Review. 56(1), 19–54. Ştefan, O., (2013). Soft law in court. competition law, state aid and the Court of Justice of the European Union. Alphen aan de Rijn: Kluwer Law International. Steunenberg, B., (2010). Is big brother watching? Commission oversight on the national implementation of EC directives. European Union Politics. 11(3), 359–380. Terpan, F., (2015). Soft law in the European Union—the changing nature of EU law. European Law Journal. 21(1), 68–96. Toshkov, D., (2011). Public opinion and policy output in the European Union: a lost relationship. European Union Politics. 12(2), 169–191. Treaty establishing the European Economic Community (1957) [online]. [Viewed 28 October 2022]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​?uri​=​celex:​11957E/​TXT (EEC). Treib, O., (2014). Implementing and complying with EU governance outputs. Living Reviews in European Governance. 9(1). UNDP., (2021). COVID-19 global gender response tracker [online]. United Nations Development Programme (UNDP). [Viewed 30 September 2022]. Available from: https://​ data​ .undp​ .org/​ gendertracker. United Nations., (1995). Beijing Declaration and Platform for Action, Fourth World Conference on Women, 15 September 1995 (A/CONF.177/20) [online]. Beijing: United Nations. [Viewed 27 October 2022]. Available from: https://​www​.un​.org/​en/​development/​desa/​population/​migration/​ generalassembly/​docs/​globalcompact/​A​_CONF​.177​_20​.pdf. UN Women., (2022). Poverty deepens for women and girls, according to latest projections [online]. UN Women. [Viewed 30 September 2022]. Available from: https://​data​.unwomen​.org/​features/​poverty​ -deepens​-women​-and​-girls​-according​-latest​-projections. Vibert, F., (1999). Fritz Scharpf, governing in Europe: effective and democratic? Oxford University Press, 1999. Journal of Public Policy. 19(2), 233–236. World Economic Forum., (2022). Global gender gap report 2022 [online]. Geneva: World Economic Forum. [Viewed 30 September 2022]. Available from: https://​ www​ .weforum​ .org/​ reports/​ global​ -gender​-gap​-report​-2022. Xenidis, R. and Senden, L., (2020). EU non-discrimination law in the era of artificial intelligence: mapping the challenges of algorithmic discrimination. In: U. Bernitz, X. Groussot, J. Paju and S. A.

390  Research handbook on soft law de Vries, eds. General principles of EU law and the EU digital order. Alphen aan de Rijn: Kluwer Law International. pp. 151–182. Zippel, K. S., (2006). The politics of sexual harassment: a comparative study of the United States, the European Union and Germany. Cambridge: Cambridge University Press.

25. Soft law and the rule of law crisis Joelle Grogan and Clara van Dam

1. INTRODUCTION The rule of law crisis has presented itself as one of the most significant crises to impact the constitutional foundations of the European Union (EU). Broadly, the crisis refers to the situation in an increasing number of EU States, paradigmatically Hungary and Poland, which have since the early 2010s been in a state of ‘backsliding’ from their commitment to the rule of law evidenced through the increasing consolidation of power within the executive, and reduction or removal of judicial independence and media freedom. Such a crisis has immediate impact not only on the EU citizens residing and working in these States, but also on the principle of mutual trust between Member States. This chapter considers soft law that has been adopted to enhance the rule of law in the EU and among the Member States. In Section 2, the chapter provides an overview of the recognition and development of the concept of the rule of law as a foundational value within the EU legal system. It then provides context to the current crisis, highlighting the challenges faced by Union institutions to challenge and mitigate concerns related to the increasing non-compliance with the rule of law. Section 3 then provides a contextual overview of soft law mechanisms in the EU regulatory space, providing the analytical framework for understanding the meaning of ‘effectiveness’ within the context of the application of soft law mechanisms. Section 4 analyzes the five main soft law mechanisms relevant in the context of the rule of law crisis in the EU: the European Semester, the EU Justice Scoreboard, the Rule of Law Framework, the annual rule of law dialogue, and the rule of law review cycle. The chapter concludes on a collective evaluation of these mechanisms within the context of the challenges to the rule of law within the EU.

2.

RULE OF LAW IN THE EUROPEAN UNION

2.1

The Concept and Development of the Rule of Law in the European Union

The rule of law has been commonly understood as a shared political ideal which underlies and provides the foundation for liberal democratic States. Within the European context, it is a legal principle of constitutional value which forms part of the common legal heritage of the Member States, as well as a foundational value of the EU (Pech et al. 2020a, p. 60; 2020b). The founding Treaties of the European Union did not explicitly reference the value, however, as examined elsewhere (Pech 2020a), the lack of explicit reference did not mean the lack of rule of law controls on the exercise of EU legal authority – most notably, judicial review on the exercise of EU legal authority. The first explicit reference was within the Court of Justice’s seminal decision in Les Verts (C-294/83), wherein the Court found that the then European Economic Community is a, ‘Community based on the rule of law, inasmuch as neither its 391

392  Research handbook on soft law Member States nor its institutions can avoid a review of the question whether the measures adopted by them are in conformity with the basic constitutional charter, the Treaty’ (para. 23). The rule of law was formally enshrined in the EU Treaties in the Treaty of Amsterdam in 1997, and following the 2007 Lisbon Treaty, it is now enumerated within Article 2 of the Treaty on European Union (TEU) as one of the values which are ‘common to the Member States’ and upon which the EU is founded. The concept within the EU should be understood to mean more than the mere obligation that all acts of EU institutions are subject to review. Based on the provisions of the Treaties and also building on the jurisprudence of the Court of Justice, the European Commission provided definition of the core components of the rule of law in 2014 as part of the Rule of Law Framework1 (European Commission 2014), and further offered a definition of the rule of law in the 2020 Rule of Law Report (European Commission 2020a, p. 1), according to which: public powers must be bound by constraints set out by law, in accordance with the values of democracy and fundamental rights, and under the control of independent and impartial courts. The rule of law includes principles such as legality, implying a transparent, accountable, democratic and pluralistic process for enacting laws; legal certainty; prohibiting the arbitrary exercise of executive power; effective judicial protection by independent and impartial courts, effective judicial review including respect for fundamental rights; separation of powers; and equality before the law.

The Lisbon Treaty reforms also encompassed a number of other provisions which created a more ‘robust’ legal environment for the protection and upholding of the rule of law, including a formal reinforcement of the conditions of eligibility for candidate countries with an obligation to show commitment to the promotion of these values; and the obligation under Article 21 TEU for the EU in its external policies to be guided by, and seek to promote, these values. A further reform, and one crucially important for understanding the soft law context of rule of law measures within the EU, was the introduction of Article 7 TEU. Article 7 TEU has been referred to, arguably inaccurately (Kochenov 2017), as the EU’s ‘nuclear option’, and is intended to enable sanctioning of EU Member States in a situation in which there was ‘serious and persistent breach’ of the values laid down in Article 2 TEU – notably in this context, the rule of law. It is a two-stage process: Article 7(1) – the so-called ‘preventive arm’, and Articles 7(2) and (3) TEU – the ‘sanctioning arm’. The scope of Article 7 TEU extends beyond areas covered by Union law to include any area of autonomous action by Member States where such action could ‘undermine the very foundations of the Union and the trust between its members, whatever the field in which the breach occurs’ (European Commission 2003). The introduction of this Article reflected the lack of internal EU institutional capacity or competences to effectively ensure compliance with the rule of law and other Article 2 TEU values by EU Member States, in sharp contrast to the high bars set by pre-accession conditionality for entry to the EU. In effect, there has been no clear prevention or sanctioning mechanism for Member States which are ‘backsliding’ (Pech and Scheppele 2017) from compliance and commitment with the rule of law. 2.2

The ‘Rule of Law Crisis’

In his 2012 State of the Union address (European Commission 2012), the then-Commission President Barroso identified the increasing awareness of ‘challenges to the rule of law in some Member States’ which presented ‘threats to the legal and democratic fabric in some of

Soft law and the rule of law crisis  393 our European states’ and referred to the ‘nuclear option’ of Article 7 TEU. This was not the first alarm, as in 2011 the European Parliament (2013) initiated monitoring of the situation in Hungary, culminating in the adoption of the Tavares Report in 2013. The concept of the rule of law, and the expectation of compliance with it as a common value among EU Member States, has come under increasing scrutiny in the near decade since with the observable increase of ‘rule of law backsliding’ within certain Member States, most notably, Hungary and Poland culminating in a so-called ‘rule of law crisis’ within the EU. Rule of law backsliding refers to ‘the process through which elected public authorities deliberately implement governmental blueprints which aim to systematically weaken, annihilate or capture internal checks on power with the view of dismantling the liberal democratic state and entrenching the long-term rule of the dominant party’ (Pech and Scheppele 2017, p. 7). It includes, inter alia, the dismantling of judicial independence and targeted attacks on NGOs and independent media among other cornerstones of liberal democracy leading to the increasing autocratization of certain Member States of the EU. At its end, it enables the establishment of elected autocracies and effectively one-party States. Rule of law backsliding is a critical issue for the EU, not only where it negatively impacts on the lives and livelihoods of EU citizens within these Member States, but also where autocratic governments have effective political and decision-making powers within the EU institutions, notably the Council of the European Union (2014). Further, the ‘constitutional capture’ or the undermining of judicial independence to bring the courts under direct political control effectively undermines the guarantees of consistent interpretation and application of EU law, as well as the enforcement of EU rights and obligations. As mentioned, Article 7 TEU had been introduced as a direct answer to the concerns for a lack of a control mechanism for Member States that are not complying with the fundamental values of the EU. It has, however, ostensibly proven so far ineffective in containing or limiting rule of law backsliding and resolving the rule of law crisis in the EU. In 2017, Poland became the first EU Member State to be subject to the Article 7 TEU procedure out of concern that the State was at clear risk of a serious breach of the rule of law (European Commission 2017a). This had followed the activation and (unsuccessful) application of the Rule of Law Framework2 against Polish authorities in 2016. In 2018, Hungary was referred to the Council by the Parliament, making it the second Member State subject to the Article 7 process. In addition to these processes, a number of infringement actions have been launched, in addition to preliminary references of direct relevance to rule of law backsliding (Pech and Kochenov 2021), most notably the Białowieża forest (C-441/17) and Portuguese Judges cases (C-64/16). These processes so far have shown limited, if any, success in addressing the pressing concerns for rule of law backsliding in these states (Pech and Scheppele 2017; Scheppele et al. 2020). The soft law mechanisms discussed in this chapter play out against this background, and both the impetus for their creation and their effectiveness must be understood in this context. The EU has adopted a number of soft law mechanisms both in direct response to the rule of law crisis and also as part of a larger commitment to core European values which have relevance to the resolution of the rule of law crisis, and which are analyzed as part of this chapter: the European Semester Process, the EU Justice Scoreboard (2013), the Rule of Law Framework (2014), the Annual Rule of Law dialogue (2014), and the Rule of Law Review Cycle (2020).

394  Research handbook on soft law

3.

A FRAMEWORK TO ANALYZE THE EFFECTIVENESS OF SOFT LAW INSTRUMENTS IN REMEDYING THE RULE OF LAW CRISIS

Soft, non-legally binding instruments have become part and parcel of the EU regulatory landscape (Ştefan 2013, p. 1). With various soft instruments to address the EU’s rule of law crisis, the ‘rule of law Toolbox’ is no exception. Before turning to the analysis of soft rule of law mechanisms that have been initiated to address the rule of law crisis, this section attempts to develop a framework that enables us to categorize ‘soft rule of law instruments’ as well as to assess their effectiveness. To this end, we first address the question whether soft rule of law instruments can be considered ‘soft law’ (sub-section 3.1) and reflect on different categories and types of soft instruments that have been discerned in the EU and on how the soft rule of law mechanisms fit into these categories and types (sub-section 3.2). Subsequently, we develop a conceptual framework to assess the effectiveness of soft rule of law instruments. To this end, we understand effectiveness as the degree to which rule of law instruments are able to achieve their objectives in remedying the rule of law crisis (section 3.3). Finally, in the light of existing literature and research on EU soft law, this section identifies some questions that we consider relevant when analyzing the variety of soft rule of law instruments as well as when assessing their effectiveness (sub-section 3.4). 3.1

Soft Rule of Law Instruments: Soft ‘Law’?

Even though the notion of soft law is widely used in legal literature, there is no definition of the concept of ‘soft law’ given in the EU Treaties.3 A definition that is commonly used is the one given by Senden (2004, p. 113), who defines soft law as ‘rules of conduct that have not been attributed legally binding force as such, but nevertheless may have certain (indirect) legal effects, and that are aimed at and may produce practical effects’. As this definition shows, one of the essential characteristics of soft law instruments is that it must concern rules of conduct. This means that the instruments must include a normative rule prescribing a certain behaviour or measures to be adopted by its addressees (Senden 2004, p. 113). Although soft ‘rule of law instruments’ sometimes include rules of behaviour,4 this is not always the case. Soft rule of law instruments may also be of a highly informational character – whilst not explicitly prescribing a normative rule to the Member States.5 As a result, not all soft instruments that are part of the rule of law toolbox may be considered EU soft law. Nonetheless, even when not including rules of behaviour, the instruments can still be called ‘soft governance tools’, as they are aimed at ‘political steering in which non-hierarchical modes of guidance, such as persuasion and negotiation are employed’ (Héritier 2002, p. 185; Strelkov 2019, p. 17). Hence, in the next paragraphs we refer to ‘soft rule of law instruments’ as an overarching category of all soft instruments (also those with a highly informational character) related to the rule of law toolbox. 3.2

Formal and Informal Rule of Law Steering Instruments

‘Soft rule of law instruments’ can be further characterized as fitting the so-called category of soft ‘steering instruments’ (also referred to as ‘soft regulatory rule-making’) in distinction

Soft law and the rule of law crisis  395 from ‘post-legislative guidance’ (or ‘soft administrative rule-making’). Characteristic for this category of ‘steering instruments’ is that it encompasses non-legally binding instruments that do not bear a close relationship to EU hard law. These instruments seek to steer behaviour of Member States or other actors in the direction of certain regulatory or policy objectives rather independently (Senden and Van den Brink 2012, p. 12) and are often issued in areas where the EU has few legislative competences or as an alternative to legislation (Hofmann 2021, p. 43). Other soft (law) measures taking the form of post-legislative guidance have a closer relationship to EU hard law, by giving guidance on the interpretation and application of EU law provisions by the Commission itself as well as on the implementation of EU law by the Member States (Van Dam 2020). The various soft rule of law instruments do not give post-legislative guidance, but rather seek to promote compliance with rule of law values through soft steering mechanisms that operate on the side of formal enforcement mechanisms. Within the broad category of ‘steering instruments’, Senden (2004) distinguishes among formal and informal steering instruments. Formal instruments have a basis in Article 288 TFEU and hence take the form of recommendations and opinions. Informal steering instruments do not have a basis in the Treaties and take the form of various other instruments such as conclusions, declarations, guidelines, or notices (Senden 2004, p. 158). This categorization between formal and informal steering instruments might also help to describe and analyze the soft rule of law tools. Indeed, a quick look at the rule of law instruments already reveals that in the context of the European Semester, the Commission and the Council adopt recommendations, whereas other soft rule of law tools are labelled differently (such as scoreboards and reports) and hence have a much more informal character. The above reflections on soft rule of law instruments as ‘formal and informal steering instruments’, not only help to understand the role of these instruments in the broader EU regulatory landscape. This categorization might also be relevant when analyzing the effectiveness of these instruments. Previous research suggests that soft instruments that bear a close relationship to EU hard law (i.e., post-legislative guidance) tend to be more regularly applied by national actors than soft instruments that do not bear such a close relationship to EU hard law (i.e., steering instruments) (Hofmann 2021, pp. 52–55). Hence, one might expect that soft rule of law instruments might not be applied as regularly as soft law acts that take the form of post-legislative guidance. What is more, it would be interesting to see whether differences can be observed between the effectiveness of formal rule of law steering instruments on the one hand, and the effectiveness of other more informal rule of law steering instruments on the other hand. Since in this contribution we study both formal and informal steering instruments, the analysis of the effectiveness of soft rule of law instruments might provide some further insights into this question. 3.3

Conceptualizing the Effectiveness of Soft Rule of Law Instruments

In terms of assessing the effectiveness of soft rule of law instruments, we understand the concept of effectiveness as the degree to which rule of law instruments are able to achieve their objectives in remedying the rule of law crisis. This means that we do not necessarily look at (formal) compliance with these instruments (e.g., transposition, application, and enforcement). Instead, we look into whether the (policy) goals of legal instruments are achieved in practice (compare Sarat 1985, pp. 23–31; Snyder 1993, pp. 19–54); This broad(er) perspective on effectiveness best fits our research aim, which is to provide insights in the degree as to which

396  Research handbook on soft law the rule of law instruments are able to remedy the rule of law crisis. It also fits the character of rule of law instruments, which often have as objective to promote ‘rule of law’ values rather than requiring strict compliance with specific rules or recommendations. Taking this broad perspective means that for each soft ‘rule of law instrument’, the (policy) objectives in remedying the rule of law crisis will need to be identified (Sarat 1985, p. 23). To this end, we identify which rule of law values the instruments seek to enhance, as well as how the instruments aim to do this. Subsequently, after having assessed the effectiveness of the selected rule of law instruments, the next question is how to explain a possible gap between the goals of these instruments and the results that have been observed (Sarat 1985, p. 23). 3.4

Assessing Effectiveness: Factors that Might Matter

The next section analyzes the soft rule of law instruments and assesses their effectiveness. This section introduces the factors and questions in light of which we analyze and assess the ability of soft rule of law instruments to remedy the rule of law crisis. In the first place, we look into the question whether the instruments contain precise normative recommendations or guidance, or whether they instead adopt a more suggestive or voluntary wording. Previous research on the role of post-legislative guidance (Van Dam 2020, pp. 235–238) and other forms of soft law (Saurugger and Terpan 2021) in the national legal order shows that the wording of guidance provisions matters for the way in which it is used in the implementation of EU law. We explore whether the degree of normativity also matters for the effectiveness of soft rule of law instruments. Second, we pay attention to the issuing process and in particular to whether Member States are involved in the design of soft rule of law instruments. Indeed, existing research suggests that an involvement of national actors contributes to a broader support for reforms in line with soft EU measures (Bekker 2021, pp. 114–133). In addition, we examine which EU institution eventually adopts the soft rule of law instruments. As we will see, as most rule of law instruments are issued by the Commission, rule of law instruments may also be adopted by the Council or endorsed by the European Council. Recommendations that have been endorsed by the Council or the European Council may become (more) politically binding on the Member States, leading to higher incentives to actually implement them in national policies. The third factor we look into is the relationship of soft rule of law instruments with EU hard law. As already noted above, previous research indicates that soft instruments that bear a close relationship to EU hard law tend be more regularly applied by national actors than soft instruments that do not bear such a close relationship to EU hard law (Hartlapp and Hofmann 2021, p. 149; Hofmann 2021, pp. 52–55). Hence, we explore whether there is a relationship with EU hard law instruments, and what this relationship entails. We find that soft rule of law instruments in the context of the European Semester have a strong legal underpinning by EU hard law, whereas other, more informational, rule of instruments have a loose connection with EU legislation. This then leads to the fourth and final factor that we will take into the equation, namely whether and to which extent pressures are being exerted at the European level to adopt policy changes in line with reforms and guidance suggested in the rule of law instruments (see also Van Dam 2020). Such pressures could, for instance, be the result of financial corrections or sanctions being imposed on Member States in case of non-compliance (Van Dam 2015, pp. 193–195), the construction of peer-pressure (Strelkov 2019, p. 15) or ‘naming and

Soft law and the rule of law crisis  397 shaming’. As we will see, strong steering pressures – even when backed up by sanctions – do not necessarily guarantee an effective role for soft rule of law instruments in remedying the rule of law crisis.

4.

THE ‘RULE OF LAW’ TOOLBOX: SOFT LAW MECHANISMS AVAILABLE TO THE EU

In light of the framework outlined in the previous section, this section considers the soft law instruments that have been adopted to enhance the rule of law in the EU and among the Member States. 4.1

The European Semester Process

The European Semester is one of the core mechanisms of EU governance that seeks to address rule of law issues in the Member States (Zeitlin and Vanhercke 2018, p. 149). Implemented in 2011, the European Semester is an annual governance cycle for the coordination and surveillance of economic, fiscal, and social polies of the EU Member States (Bekker 2021, pp. 114–133). As part of the European Semester national socioeconomic policies are being analyzed and reviewed (Zeitlin and Vanhercke 2018, p. 162), which results in so-called ‘country-specific recommendations’ (henceforth ‘CSRs’) addressed to the EU Member States. The CSRs are proposed by the European Commission and after being endorsed by the European Council, they are adopted by the Council (Verdun and Zeitlin 2018, pp. 137–148). Taking the form of ‘recommendations’, CSRs can be categorized as formal steering instruments. Indeed, as said above, recommendations are among the instruments listed in Article 288 TFEU, which explicitly states that recommendations lack legally binding force. The formal character of CSRs also follows from specific provisions in primary6 and secondary EU law7 that provide for the issuing of (country-specific) recommendations on economic and employment policies. In fact, the European Semester combines several coordination mechanisms resulting in CSRs underpinned by different legal instruments: the Stability and Growth Pact (SGP), the Macroeconomic Imbalance Procedure (MIP), as well as employment and social policy coordination (Verdun and Zeitlin 2018, p. 138; Bekker 2021, p. 115). Despite the lack of legally binding force, pressures to act in line with the CSRs are being exerted on the Member States in various ways. This is the case especially for recommendations underpinned by the SGP and the MIP, for the reason that these legal mechanisms both have a preventive and a corrective arm. This means that Member States having an excessive deficit or macroeconomic imbalances are subject to strict monitoring procedures and might eventually face sanctions or fines in case of failure to implement CSRs.8 When not in this corrective phase (yet), several monitoring mechanisms are in place that exert ‘soft’ pressure on Member States to act in line with the required reforms. On the other hand, recommendations underpinned by the employment policy have less coercive elements, for the reason that employment policy follows the logic of the ‘open method of coordination’ (Dawson 2018, pp. 191–209). Overall, the European Semester has been considered a governance instrument focused on rules and enforcement (Leino-Sandberg and Losada Fraga 2020, p. 9). Perhaps counter-intuitively, recent research shows that the effectiveness of the European Semester is rather modest, to put it lightly (Leino-Sandberg and Losada Fraga 2020, pp. 16,

398  Research handbook on soft law 17). The European Court of Auditors report (2020) finds a low substantial implementation rate of CSRs: only 1.6 per cent of the CSRs has been fully implemented and that only in 4.6 per cent of CSRs was ‘some progress’ made. Similarly, Efstathiou and Wolff (2018) find that in the last years overall implementation of CSRs by EU Member States has worsened. Remarkably, ‘low implementation’ of CSRs has been observed in particular in Member States with excessive macroeconomic imbalances (and hence are in the corrective arm of the MIP) (Efstathiou and Wolff 2018, p. 4).9 In the literature it has been argued that Member States tend to make their own ‘sovereign’ decisions on socioeconomic and fiscal policies (Leino-Sandberg and Losada Fraga 2020, p. 20). Not only do CSRs often provide specific and detailed guidance on these issues, they may be backed up by sanctions as well.10 This enforcement approach has triggered resistance and contestation, rather than acceptance and deliberation (Efstathiou and Wolff 2018; Leino-Sandberg and Losada Fraga 2020, pp. 19, 30). A more dialogical and deliberative approach is believed to be more effective than the rule-based and hierarchical approach that has dominated in the European Semester process for many years (Leino-Sandberg and Losada Fraga 2020, p. 30). In this regard, ‘improvements’ are already being made: more and more often Member States are being consulted during the issuing process of CSRs (Bekker 2021, p. 126). Eventually, this dialogical approach is expected to lead to a greater fit between the CSRs and national policies, as well as to generate a greater support of national actors for actually pursuing reforms (Bekker 2021, p. 129). This, however, remains to be seen. 4.2

The EU Justice Scoreboard

The second key component of the rule of law toolbox is the EU Justice Scoreboard published annually by the European Commission. First issued in 2013, the EU Justice Scoreboard gives a comparative overview of the independence, quality, and efficiency of justice systems in all EU Member States. The Scoreboard’s overall aim is to ensure that citizens and business can ‘fully enjoy their rights’, hence promoting effective justice which is crucial for upholding rule of law values (European Commission 2020b). The scoreboard takes the form of a Communication and has a highly informational character. Unlike CSRs the EU Justice Scoreboard does not have close ties with EU primary or secondary legislation. Instead, the EU Justice Scoreboard is a comparative tool which does not include concrete and precise recommendations or a ranking of performances of individual Member States. Furthermore, the Scoreboard draws on the information that is voluntarily provided by the Member States. In the words of the Commission, the Scoreboard evolves ‘in dialogue with the Member States’ (European Commission 2020b, p. 2). It does not envisage any sanctioning mechanisms (Strelkov 2019, p. 17). It follows that the Scoreboard relies on a dialogical approach, rather than on a top-down legal architecture. Despite the informational character, the Scoreboard might still exert soft (peer) pressures on the Member States to improve the workings of their judicial systems (Strelkov 2019, pp. 24, 25). Being a ‘soft governance tool’ (Strelkov 2019, p. 17), the Scoreboard aims to have a learning effect on the Member States and to identify good practices. In his foreword to the 2020 edition of the Scoreboard, Commissioner D. Reynders states that ‘[n]ow more than ever Member States need to learn from each other to improve the effectiveness of their justice systems’ (European Commission 2020b). Furthermore, the information provided by the EU

Soft law and the rule of law crisis  399 Justice Scoreboard may be used as a source of information for the drafting of CSRs in the context of the European Semester (Mak and Taekema 2016, pp. 36, 37). Thus far, the Scoreboard has not commonly used ‘normative reference point’ for national actors to improve their judicial systems (Strelkov 2019, pp. 23, 24). The 2020 EU Justice Scoreboard concludes that over the last years the efficiency of justice systems in EU Member States has improved, but also reveals that there is a continuing decrease in perceptions of judicial independence with interference from governments and politicians as the most stated reason (European Commission 2020b, pp. 19, 54). This observation, that problems concerning insufficient safeguards for judicial independence continue to exist, may be considered illustrative for the limited steering effect the Justice Scoreboard has had on this point. The finding that the Scoreboard has so far had a limited effect on resolving issues of judicial independence, may be explained by the fact that the judicial organization remains one of the key prerogatives of EU Member States (Strelkov 2019, p. 16). Hence, Mak and Taekema (2016, p. 48) question the effectiveness of the dialogical approach emphasized in the current rule of law initiatives, especially with regard to the Member States who ‘deliberately distance themselves from EU values’. And yet, there is hope that in the long term the Scoreboard’s non-hierarchical and deliberative approach will increase awareness of various national practices and rule of law problems, and eventually, increase trust between the EU institutions and the Member States (Strelkov 2019, pp. 23–25). 4.3

The Rule of Law Framework

The third key component of the rule of law toolbox was the adoption of the New Framework to Strengthen the Rule of Law in the EU in September 2014. It was a response to many recommendations made in 2013 by the Justice and Home Affairs Council (2013), the General Affairs Committee (Council of the European Union 2013), and the European Parliament to assess compliance with fundamental values of the EU, and to address cases of systemic11 threat to the rule of law. The mechanism in the Framework is intended to provide an alternative, informal gap-filling mechanism, and resolve threats to the rule of law before sanctioning conditions for Article 7 TEU are met. It was not meant, however, to prejudice targeted action on breach of EU law on the basis of Article 258 TFEU. The Commission emphasized the guarantee of equality of treatment between Member States as regards monitoring of rule of law compliance. The Framework also aims to provide clarity and enhance predictability in the Commission’s actions with regard to monitoring Member States. It will be triggered in situations where: the authorities of a Member State are taking measures or are tolerating situations which are likely to systematically and adversely affect the integrity, stability or the proper functioning of the institutions and the safeguard mechanisms established at national level to secure the rule of law (European Commission 2014, p. 6).

The Commission will act where the political or institutional order, the constitutional structure, the separation of powers, the independence or impartiality of the judiciary, or the system of judicial review is threatened, for example ‘as a result of the adoption of new measures or of widespread practices of public authorities and the lack of domestic redress’ (European Commission 2014, p. 7). Breaches of fundamental rights or miscarriages of justice do not trigger the mechanism, instead they are expected to be dealt with under existing national

400  Research handbook on soft law measures, or through the European Court of Human Rights, as all EU Members are party to the European Convention on Human Rights, and are members of the Council of Europe. The process envisaged by the Framework consists of three stages: (1) a Commission assessment; (2) a Commission recommendation, and (3) a follow-up to the recommendation given by the Commission. It is based on the following principles: 1. 2. 3. 4.

focusing on finding a solution through a dialogue with the Member State concerned; ensuring an objective and thorough assessment of the situation at stake; respecting the principle of equal treatment of Member States; indicating swift and concrete actions which could be taken to address the systemic threat and to avoid the use of Article 7 TEU mechanisms (European Commission 2014, p. 7) (Emphasis original).

Throughout the process, dialogue between the Commission and the Member State is emphasized. Additionally, the Commission will seek ‘external’, third-party guidance from the EU Agency for Fundamental Rights, as such expertise will ‘help to provide for a comparative analysis about existing rules and practices in other Member States in order to ensure equal treatment of the Member States, on the basis of a common understanding of the rule of law within the EU’ (European Commission 2014, p. 7). The Commission stated that the scope of intervention is not confined to areas covered by EU law, and that the EU is empowered to intervene even in instances where the Member State is acting autonomously. Intervention is justified on the basis that a systemic threat to the rule of law, not secured by national mechanisms which have ceased to function effectively, is a threat to the functioning of the EU as an ‘area of freedom, security and justice without internal frontiers’ (European Commission 2014, p. 5). There are both substantive and procedural weaknesses within the Framework, as Kochenov and Pech (2015) identify, including the absence of clear benchmarks against which the concept of ‘systemic threat’ can be assessed, and a further lack of clarification as to the nature of systemic threat as opposed to systemic violation. They also highlight the procedural inadequacies of the Framework as the confidential nature of Commission enquiries lack the ‘name and shame’ element of public disclosure (European Commission 2014, p. 5). From 2015, the European Parliament called upon the Commission to activate the Framework in respect of Hungary. The Commission however declined to do so on the ground that concerns could be addressed through either infringement procedures or national safeguards. However, in respect of Poland, the Commission activated the Framework for the first time in January 2016 on the basis of the measures adopted by the Polish government concerning the Constitutional tribunal and its functioning (European Commission 2016). However, despite issuing a formal Opinion and four Recommendations under the Framework, the Commission conceded in December 2017 that: the Polish authorities have adopted more than 13 laws affecting the entire structure of the justice system in Poland ... . The common pattern is that the executive and legislative branches have been systematically enabled to politically interfere in the composition, powers, administration and functioning of the judicial branch (European Commission 2017b).

Finding no other solution or resort, Article 7(1) TEU was triggered for the first time, and in respect of Poland, in December 2017.

Soft law and the rule of law crisis  401 4.4

The Council’s Annual Rule of Law Dialogue

The Annual Rule of Law Dialogue was introduced by the Council shortly following the adoption by the Commission of the new Framework for the rule of law in 2014. The adoption by the Council of a further dialogue mechanism between Member States (without Commission intervention nor reference to the new Rule of Law Framework) was interpreted as a rebuke of the Commission’s approach to the framework on the rule of law as a ‘pre-Article 7’ process, and the implication that it overstepped conferred competences to the Union (Pech 2020a). In this way, the Rule of Law Dialogue ostensibly creates a further distance from the ‘hard law’ enforcement mechanism of Article 7 TEU. In the Conclusions committing to the establishment of the Annual Rule of Law Dialogue, only seven provisions provide little by way of understanding the format, procedure, or substance of the Dialogue, despite the stress for the need for an ‘evidence-based approach’ giving it a more informal character than that of the Commission’s Rule of Law Framework. While it has been considered that direct engagement with the rule of law by the Council could lead to higher levels of (political) commitment to the value, the challenges posed become evident in examining the direction of the Council’s Rule of Law Dialogue from the point of its adoption. Between 2014–2018, Annual Dialogues focused on the themes of the rule of law in the digital era, migrant integration, medial pluralism and the rule of law in the digital age, and trust in public institutions. As an identified point of criticism (Pech 2020a, p. 24): none of these topics spoke directly to the critical and known rule of law issues including judicial independence in Hungary and Poland during this same period. A key reason was revealed in 2019, when the Council failed to adopt the conclusions following the evaluation of the dialogue mechanism under the Finnish Presidency: requiring unanimity of all countries – two of which were then subject to the Article 7(1) TEU procedure. Beyond the undefined scope and limited nature of the Annual Rule of Law Dialogue, there appears to be little external evidence of an effective connection between Council dialogues and action, or the amelioration of the rule of law situation in the EU. The difficulty of securing unanimity in order to adopt Conclusions which, while brief, and often only in the nature of best practices, highlights the challenge of meaningful action taken on the basis of a dialogue solely at Council level and between national governments. 4.5

Rule of Law Review Cycle

The Rule of Law Review Cycle was introduced in 2019, and a further initiative was led by the Commission, the key aim of which was to enable the development in the EU of ‘a stronger awareness and understanding of developments in the individual Member States’ (European Commission 2019). The Cycle aims to enhance the rule of law culture within the EU, and also to act as a ‘preventive tool’ by early identification of challenges within the Union, and to advocate best practices across the EU. Within this context, it may be better understood as less a ‘normative’ tool than perhaps might initially be understood. The Cycle is built on three pillars: promoting a rule of law culture, preventing rule of law problems, and responding effectively to the breaches of the rule of law (European Commission 2019). Partially in answer to the accusation by some Member States of ‘double standards’ (BBC News 2012), the Rule of Law Cycle involves annual reporting of all Member States, in addition to a transversal analysis to offer best practices and solutions. The first report was

402  Research handbook on soft law published in September 2020. The report adopted four main pillars as a point of focus: the justice system, the anti-corruption framework, media pluralism, and other institutional checks and balances essential to an effective system of governance. While still a new instrument at the time of writing, some initial issues may be identified which could hamper future effectiveness of this aspect of the Rule of Law Review Cycle. A criticism is that, where it is preventive, it fails to remedy existing systemic breaches of the rule of law: simply, an ex-post facto reporting mechanism may be unable to either deter or stop rule of law violations from first taking place. In sharp criticism of the first report, Pech (2020b) argues that the report, through the use of downplaying or euphemistic language, does not correctly correlate the reality of ‘attacks on and violations of the rule of law [that] have been deliberate, widespread and systemic in nature’ (Emphasis original). The collective effort of standardizing the rule of law reporting across all Member States presents not only a logistical challenge, but also an arguable diffusion of the Commission’s efforts from enforcement to reporting, obfuscating the most troubling developments among the false equivalence of there being rule of law issues (of significantly lesser degrees) in all EU Member States. The collective grouping of States and macro-level reporting of issues do not necessarily create meaningful peer-pressure, nor do ‘naming and shaming’ and the focus on prevention – rather than sanction or enforcement – aid a situation that is already in a negative position.

5. CONCLUSION This chapter has provided an overview of the core tools in within the rule of law toolbox available to the EU and on this basis offers key findings. Any analysis of the range of the soft rule of law tools available is that strong steering pressures – even when backed up by sanctions – do not necessarily guarantee an effective role for soft rule of law instruments in remedying the rule of law crisis. Nor have we found indications that formal soft rule of law instruments that have a strong underpinning in EU hard law have significantly more normative force than informal soft rule of law instruments. Indeed, the rule of law instruments in the form of recommendations – which are the most formal ones in the context of the European Semester – show a limited ability to remedy rule of law issues. Further, the preventive mechanisms cannot be effective in preventing a situation which has already happened. Dialogical approaches have characterized that a majority of the soft law tools are adopted in response to the crisis; however, such tools can only result in meaningful change when both sides are committed to action. Consistently evidenced by the tools primarily relying on mutual trust, solidarity and the commonality of fundamental values across the Union is the relative and arguable ineffectiveness of these mechanisms when governments are not committed to these values. Yet, there is (some) hope that in the long run, the soft rule of law instruments might still contribute to at the least promoting and shaping a debate on rule of law values whilst increasing the awareness of national rule of law practices and cultures (compare Strelkov 2019, p. 24). This might, eventually, help to (re)build trust between the Member States and the EU institutions, and improve the willingness of all EU Member States to comply with rule of law values. From the above analysis it follows that, in order to smoothen this process, improvements can be made in the way the soft rule of law instruments are designed. In the first place, soft rule of law mechanisms should be clearer in providing clear benchmarks against which national practices are being evaluated. Second, the naming-and-shaming function of these mechanisms

Soft law and the rule of law crisis  403 could be strengthened, for instance by not making the results of the Rule of Law Framework confidential. Thirdly, the instruments could have more normative force by focusing on the dissemination of best practices rather than on setting up mechanisms of ‘macro-level reporting of all Member States’. Nonetheless, these recommendations for improvement are only a first step forward. Further (empirical) research is needed on what features and mechanisms could improve the workings of (future) soft rule of law instruments, enabling these instruments to effectively remedy the EU rule of law crisis.

NOTES 1. For analysis, see infra section 4.3. 2. See infra section 4.3. 3. In its Resolution of 4 September 2007, the European Parliament considers the concept of soft law as ‘ambiguous and pernicious’ and advises that the notion of soft law ‘should not be used in any document of the Community institutions’. See European Parliament (2007). 4. As is the case for Country-Specific Recommendations. See infra section 4.1. 5. An example is the EU Justice Scoreboard. See infra section 4.2. 6. Articles 121 and 148 TFEU. 7. Council Regulation 1466/97 and Regulation 1176/2011. 8. Article 126 TFEU; Articles 5 and 6 Regulation (EU) 1173/2011; Article 3 Regulation (EU) 1174/2011. 9. Efstathiou and Wolff (2018) find the lowest implementation in Luxemburg, Slovakia, Hungary, Germany, and Bulgaria. 10. Namely, CSRs are underpinned by the SGP or MIP. 11. A ‘systemic’ problem is one which affects and exists throughout the entire system.

REFERENCES BBC News., (2012). Hungarian PM Viktor Orban denounces EU’s ‘colonialism’ [online]. BBC News. [Viewed 20 October 2021]. Available from: https://​www​.bbc​.co​.uk/​news/​business​-17394894. Bekker, S., (2021). Hardening and softening of country-specific recommendations in the European Semester. West European Politics. 44(1), 114–133. Case C-294/83 Parti écologiste ‘Les Verts’ v European Parliament ECLI:EU:C:1986:166. Case C-441/17 Commission v Poland (Białowieża forest) ECLI:EU:C:2017:877. Case C-64/16 Associação Sindical dos Juízes Portugueses ECLI:EU:C:2018:117. Council of the European Union., (2013). 3235th Council meeting. General Affairs [online]. Press release, 8577/13. Luxembourg: General Affairs Council. [Viewed 22 December 2021]. Available from: www​.consilium​.europa​.eu/​uedocs/​cms​_Data/​docs/​pressdata/​EN/​genaff/​136915​.pdf. Council of the European Union., (2014). Conclusions of the Council of the European Union and the member states meeting within the Council on ensuring respect for the rule of law [online]. Council Conclusions. Brussels: General Affairs Council. [Viewed 23 December 2021]. Available from: https://​www​.consilium​.europa​.eu/​media/​24875/​146323​.pdf. Council regulation (EC) no 1466/97 on the strengthening of the surveillance of budgetary positions and the surveillance of economic policies (1997) Official Journal L209, p. 1. Dam, C. van, (2015). EU administrative soft law and the separation of powers. In: H-M. ten Napel and W. Voermans, eds. The powers that be: rethinking the separation of powers: a Leiden response to Möllers. Leiden: Leiden University Press. pp. 191–213. Dam, J. C. A. van, (2020). Guidance documents of the European Commission in the Dutch legal order. PhD Thesis. Leiden University, Leiden Law School.

404  Research handbook on soft law Dawson, M., (2018). New governance and the displacement of Social Europe: the case of the European Semester. European Constitutional Law Review. 14(1), 191–209. Efstathiou, K. and Wolff G. B., (2018). Is the European Semester effective and useful?. Policy Contribution [online]. 9. [Viewed 20 October 2021]. Available from: https://​www​.bruegel​.org/​2018/​ 06/​is​-the​-european​-semester​-effective​-and​-useful. European Commission., (2003). Communication on Article 7 of the Treaty on European Union. respect for and promotion of the values on which the Union is based (COM/2003/606 final) [online]. Brussels: European Commission. [Viewed 15 December 2021]. Available from: https://​eur​-lex​.europa​.eu/​legal​ -content/​EN/​TXT/​?uri​=​celex:​52003DC0606. European Commission., (2012). José Manuel Durão Barroso President of the European Commission: state of the union 2012 address. plenary session of the European Parliament/Strasbourg, 12 September 2012 [online]. SPEECH/12/596. Strasbourg: European Commission. [Viewed 20 December 2021]. Available from: https://​ec​.europa​.eu/​commission/​presscorner/​detail/​en/​SPEECH​_12​_596. European Commission., (2014). Communication. A new EU framework to strengthen the rule of law (COM/2014/158 final) [online]. Brussels: European Commission. [Viewed 15 December 2021]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​?uri​=​celex:​52014DC0158. European Commission., (2016). Readout by the first Vice-President Timmermans of the college meeting of 13 January 2016 [online]. SPEECH/16/71. Brussels: European Commission. [Viewed 23 December 2021]. Available from: https://​ec​.europa​.eu/​commission/​presscorner/​detail/​en/​SPEECH​_16​_71. European Commission., (2017a). Proposal for a Council Decision on the determination of a clear risk of a serious breach by the Republic of Poland of the rule of law (COM/2017/835 final) [online]. Brussels: European Commission. [Viewed 15 December 2021]. Available from: https://​eur​-lex​.europa​ .eu/​legal​-content/​EN/​TXT/​?uri​=​celex:​52017PC0835. European Commission., (2017b). Rule of law: European Commission acts to defend judicial independence in Poland [online]. Press release, IP/17/5367. Brussels: European Commission. [Viewed 15 December 2021]. Available from: https://​ec​.europa​.eu/​commission/​presscorner/​detail/​en/​IP​_17​ _5367. European Commission., (2019). Communication. strengthening the rule of law within the union: a blueprint for action (COM/2019/343 final) [online]. Brussels: European Commission. [Viewed 15 December 2021]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​ALL/​?uri​=​COM:​2019:​ 343:​FIN. European Commission., (2020a). Communication. 2020 Rule of Law Report: the rule of law situation in the European Union (COM/2020/580 final) [online]. Brussels: European Commission. [Viewed 15 December 2021]. Available from: https://​eur​-lex​.europa​.eu/​legal​-content/​EN/​TXT/​?uri​=​CELEX:​ 52020DC0580. European Commission., (2020b). Communication. 2020 EU Justice Scoreboard (COM/2020/306 final) [online]. Brussels: European Commission. [Viewed 15 December 2021]. Available from: https://​eur​ -lex​.europa​.eu/​legal​-content/​EN/​TXT/​?uri​=​CELEX:​52020DC0306. European Court of Auditors., (2020). The European Semester – country specific recommendations address important issues but need better implementation [online]. Special report 16/2020. Luxembourg: European Court of Auditors. [Viewed 15 December 2021]. Available from: https://​ www​.eca​.europa​.eu/​en/​Pages/​DocItem​.aspx​?did​=​54357. European Parliament and Council regulation (EU) no 1173/2011 on the effective enforcement of budgetary surveillance in the euro area (2011) Official Journal L 306, p. 1. European Parliament and Council regulation (EU) no 1174/2011 on enforcement measures to correct excessive macroeconomic imbalances in the euro area (2011) Official Journal L 306, p. 8. European Parliament and Council regulation (EU) no 1176/2011 on the prevention and correction of macroeconomic imbalances (2011) Official Journal L 306, p. 25. European Parliament resolution of 4 September 2007 on institutional and legal implications of the use of ‘soft law’ instruments (2007/2028(INI)) (2007) Official Journal C 187E, p. 75. European Parliament resolution of 3 July 2013 on the situation of fundamental rights: standards and practices in Hungary (pursuant to the European Parliament resolution of 16 February 2012) (2012/2130(INI)) (2013) Official Journal C 75, p. 52. Hartlapp, M. and Hofmann, H., (2021). The use of EU soft law by national courts and bureaucrats: how relation to hard law and policy maturity matter. West European Politics. 44(1), 134–154.

Soft law and the rule of law crisis  405 Héritier, A., (2002). New modes of governance in Europe: policy-making without legislating?. In: A. Héritier, ed. Common goods: reinventing European and international governance. Oxford: Rowman & Littlefield. pp. 185–206. Hofmann, A., (2021). Types of EU soft law and their national impact. In: M. Eliantonio, E. Korkea-aho and O. Ştefan, eds. EU soft law in the Member States: theoretical findings and empirical evidence. Oxford, UK: Hart Publishing. pp. 39–56. Justice and Home Affairs Council., (2013). Council conclusions on fundamental rights and rule of law and on the Commission 2012 report on the application of the Charter of Fundamental Rights of the European Union [online]. Luxembourg: Justice and Home Affairs Council. [Viewed 20 October 2021]. Available from: www​.consilium​.europa​.eu/​uedocs/​cms​_data/​docs/​pressdata/​en/​jha/​137404​ .pdf. Kochenov, D., (2017). Busting the myths nuclear: a commentary on Article 7 TEU. EUI Working Paper, LAW 2017/10 [online]. Fiesole: European University Institute (EUI). [Viewed 20 October 2021]. Available from: https://​cadmus​.eui​.eu/​bitstream/​handle/​1814/​46345/​LAW​_2017​_10​.pdf​?sequence​=​ 1. Kochenov, D. and Pech, L., (2015). Monitoring and enforcement of the rule of law in the EU: rhetoric and reality. European Constitutional Law Review. 11(3), 512–540. Leino-Sandberg, P. and Losada Fraga, F., (2020). How to make the European Semester more effective and legitimate? Euro area scrutiny [online]. Brussels: European Parliament. [Viewed 20 October 2021]. Available from: https://​www​.europarl​.europa​.eu/​RegData/​etudes/​IDAN/​2020/​651365/​IPOL​ _IDA(2020)651365​_EN​.pdf. Mak, E. and Taekema, S., (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), 25–50. Pech, L., (2020a). The rule of law in the EU: the evolution of the treaty framework and rule of law toolbox. RECONNECT Working Paper, 7 [online]. [Viewed 20 October 2021]. Available from: https://​reconnect​-europe​.eu/​wp​-content/​uploads/​2020/​03/​RECONNECT​-WP7​-2​.pdf. Pech, L., (2020b). Doing more harm than good? A critical assessment of the European Commission’s first Rule of Law Report [online]. RECONNECT. [Viewed 20 October 2021]. Available from: https://​reconnect​-europe​.eu/​blog/​doing​-more​-harm​-than​-good​-a​-critical​-assessment​-of​-the​-european​ -commissions​-first​-rule​-of​-law​-report. Pech, L., Grogan, J., Bárd, P., Kochenov, D., Grabowska-Moroz, B., Beqiraj, J., Closa, C., Pirjatanniemi, E., Śniadach, O., Fernández-Albertos, J., Moxham, L., Koncewicz, T., Warylewska, K., Wenton A. and Podolska, A., (2020a). Meaning and scope of the EU rule of law. RECONNECT Work Package 7 – Deliverable 2 [online]. [Viewed 20 October 2021]. Available from: https://​reconnect​-europe​.eu/​ wp​-content/​uploads/​2020/​05/​D7​.2​-1​.pdf. Pech, L., Grogan., J., Bárd, P., Kochenov, D., Grabowska-Moroz, B., Beqiraj, J., Closa, C., Pirjatanniemi, E., Śniadach, O., Fernández-Albertos, J., Moxham, L., Koncewicz, T., Warylewska, K., Wenton A. and Podolska, A., (2020b). Unity and diversity in national understandings of the rule of law in the EU. RECONNECT Work Package 7 – Deliverable 1 [online]. [Viewed 20 October 2021]. Available from: https://​reconnect​-europe​.eu/​wp​-content/​uploads/​2020/​05/​D7​.1​-1​.pdf. Pech, L. and Kochenov, D., (2021). Respect for the rule of law in the case law of the European Court of Justice: a casebook overview of key judgments since the Portuguese Judges case [online]. Stockholm: SIEPS. [Viewed 20 December 2021]. Available from: https://​www​.sieps​.se/​en/​publications/​2021/​ respect​-for​-the​-rule​-of​-law​-in​-the​-case​-law​-of​-the​-european​-court​-of​-justice. Pech, L. and Scheppele, K. L., (2017). Illiberalism within: rule of law backsliding in the EU. Cambridge Yearbook of European Legal Studies. 19, 3–47. Sarat, A., (1985). Legal effectiveness and social studies of law: on the unfortunate persistence of a research tradition. Legal Studies Forum. 9(1), 23–31. Saurugger, S. and Terpan, F., (2021). Normative transformations in the European Union: on hardening and softening law. West European Politics. 44(1), 1–20. Scheppele, K. L., Kochenov, D. V. and Grabowska-Moroz, B., (2020). EU values are law, after all: enforcing EU values through systemic infringement actions by the European Commission and the Member States of the European Union. Yearbook of European Law. 39(1), 3–121. Senden, L., (2004). Soft law in European Community law. Oxford: Hart Publishing.

406  Research handbook on soft law Senden, L. and Brink, T. van den, (2012). Checks and balances of soft EU rule-making [online]. Brussels: European Parliament. [Viewed 20 December 2021]. Available from: https://​www​.europarl​ .europa​.eu/​thinktank/​en/​document/​IPOL​-JURI​_ET(2012)462433. Snyder, F. G., (1993). The effectiveness of European Community law: institutions, processes, tools and techniques. The Modern Law Review, 56(1), 19–54. Ştefan, O., (2013). Soft law in court: competition law, state aid and the Court of Justice of the European Union. Alphen aan den Rijn: Kluwer Law International. Strelkov, A., (2019). EU Justice Scoreboard: a new policy tool for “deepening” European integration?. Journal of Contemporary European Studies. 27(1), 15–27. Verdun, A. and Zeitlin, J., (2018). Introduction: the European Semester as a new architecture of EU socioeconomic governance in theory and practice. Journal of European Public Policy. 25(2), 137–148. Zeitlin, J. and Vanhercke, B., (2018). Socializing the European Semester: EU social and economic policy co-ordination in crisis and beyond. Journal of European Public Policy. 25(2), 149–174.

26. The role of soft law in the context of the financial crisis Alexander H. Türk

1. INTRODUCTION The financial crisis of 2008 exposed the weaknesses of many financial regulatory systems around the world. This was particularly the case in the European Union’s (EU) multilevel system of regulatory governance, where the financial crisis laid bare gaps in Union legislation and revealed the limits of supervision and enforcement at Union and Member State levels. It was therefore tempting, in particular by Union committees and agencies to use soft law as a means of crisis intervention. On the one hand, soft law instruments offered swift and flexible solutions for standard setting, for imposing market discipline, and for correcting gaps in the financial rule-book while the Union legislator would more slowly catch up by putting in place a more robust rule-book and financial architecture (Moloney 2014, pp. 880–888). This was all the more alluring given the considerable technical expertise of these administrative actors. On the other hand, concerns about the competence of the financial committees, the legal status of the quasi-regulatory tools they employed, questions about the legitimacy of their use as well as the danger of avoiding Treaty constraints made soft law a suspicious tool for crisis management. Undoubtedly, the reforms of the financial architecture and the more secure basis for soft law intervention in Union financial regulation have since alleviated some of those concerns. All the same, questions remain. The vast array of express or implicit bases for soft law interventions, the vastly different functions it is meant to fulfil, and the considerable differences in legal effects of such instruments, often complementing and on occasion supplanting Union legislation, require a careful analysis of the soft law landscape in EU financial regulation. This chapter will explore these themes and argue in favour of discarding the notion of ‘soft law’ and in favour of a more nuanced understanding of legal effects of regulatory instruments in financial law.

2.

EVOLUTION OF THE UNION’S REGULATORY SYSTEM FOR FINANCIAL SERVICES

The development of a comprehensive system of financial regulation has been a more recent phenomenon in Union law. While ideas for such a regime already existed much earlier (Commission 1996; Commission 1985), it was not until the Financial Service Action Plan (Commission 1999) that the European Commission seriously engaged with the pursuit of a more comprehensive system of financial regulation at Union level, setting out a legislative agenda to ensure the free movement of capital and financial services. The governance structure underpinning the substantive reforms followed the Lamfalussy report (Moloney 2014, 407

408  Research handbook on soft law pp. 942–1030). As part of a four-level structure aimed at harmonizing financial supervision and regulation in the EU, Level 1 provided for the adoption of framework legislation by the European Parliament (EP) and the Council in the context of financial regulation. Level 2 was envisaged for the drafting of implementing acts setting out the more technical aspects of financial legislation. This was the task of the Commission in cooperation with national administrations within the framework of the comitology regime. Level 3 was conceived as a ‘pole of cooperation’ between national authorities and took the form of committees where national competent authorities convened to discuss the application of European legislation. The committees, as established from 2001 to 2003, were the Committee of European Securities Regulators (CESR), the Committee of European Banking Supervisors (CEBS) and the Committee of European Insurance and Occupational Pensions Supervisors (CEIOPS). Both the Commission and the committees played a central role, with the latter acting on behalf of national supervisors and lacking intrusive powers of supervision, monitoring or sanctioning of national authorities or financial entities. Thus, their main tasks were the improvement of coordination among national regulators, their advisory role to the Commission and their assistance in the implementation of European legislation in the Member States. At Level 4 the Commission would ensure compliance with Union law. Although the Lamfalussy structure was a step in the right direction, shortcomings were identified in particular in relation to Level 3 committees (Moloney 2014, pp. 1107 et seq.). For instance, the strong ‘national imprinting’ of its functions or the absence of legally binding powers proved to be inadequate to allow for efficient coordination amongst national supervisors. These issues became particularly pronounced when Level 3 committees used soft law tools to further their objectives. In particular, CESR’s attempts to build a soft rule-book, while intended to offer a ‘pragmatic and effective response to the challenge of setting standards for a complex and evolving marketplace’ (Moloney 2014, p. 875), have been regarded as a challenge to the Union’s governance system. CESR used soft law instruments to underpin several Level 1 and 2 measures with soft law measures to support convergence and consistency for their implementation and application (Moloney 2014, p. 875). All the same, the extensive use of soft law, the lack of a clear legal basis and formal procedural constraints soon brought CESR in conflict with other institutional actors who were keen to protect their prerogatives at Levels 1 and 2 (EP 2005a, B.19; EP 2005b; Schaub 2005, p. 116; ECB 2005, pp. 7–8). This was compounded by the absence of a sufficiently well-developed accountability system. Opposition was particularly vocal when CESR used soft law during the financial crisis to address issues which arose in areas with little or no Union legislation (Moloney, 2014, p. 876). In addition to difficulties of the legal basis for CESR’s activities at Level 3, questions were also raised about the legal nature of its soft rules and their compatibility with Level 1 and 2 measures (Moloney 2014, pp. 877–878). CESR sought to address those issues by introducing mechanisms to enhance the accountability of its soft law measures not only in respect of the procedure for their adoption but also in respect of its decision-making processes (Moloney 2014, pp. 878–880). But it also introduced a ‘comply and explain’ approach to ensure greater compliance by national competent authorities (CESR 2006, Article 5; Moloney 2014, p. 879). In addition, the Commission adopted new founding decisions, which partially tried to resolve these problems (Commission 2009, Article 3). The financial crisis required, however, more profound changes to the Level 3 committee structure to put convergence activities on a more secure legal footing.

The role of soft law in the context of the financial crisis  409

3.

THE REFORM OF THE REGULATORY SYSTEM AND THE ROLE OF SOFT LAW POST CRISIS

The outbreak of the financial crisis dramatically changed financial regulation and supervision in the EU. The need for institutional reforms of EU financial regulation was highlighted by important European initiatives that showed how the established committee system was inadequate to tackle the risks of cross-border financial crises and to avoid spill-over effects. In February 2009, the Commission published the de Larosière Report (de Larosière 2009), which outlined the inefficiencies of the existing European financial supervisory model and called for a more institutionalized system of micro-prudential supervision and for the establishment of a macro-prudential system of supervision (de Larosière 2009, pp. 39–42). The Lamfalussy committee structure needed to be reformed in order to ensure financial stability and to provide more structured financial integration at the European level (de Larosière 2009, pp. 46–48). It appeared that the problems with the structure of Level 3 Committees were the absence of legally binding powers, insufficient level of accountability and transparency, and the lack of real independence from national and stakeholder interests (de Larosière 2009, pp. 54–55). The Council supported the de Larosière conclusions on the European System of Financial Supervisors (Council 2009, para. 7; European Council, 2009, para. 20) and paved the way for the creation of a new system of European financial supervision (ESFS) consisting of a European Systemic Risk Board (ESRB), which would monitor macro-prudential risks in the EU, and the three European Supervisory Authorities (ESAs), the European Banking Authority (EBA), the European Securities and Markets Authority (ESMA) and the European Insurance and Occupational Pensions Authority (EIOPA). As compared to the former Lamfalussy Level 3 committees, ESAs saw a considerable increase in tasks and powers in the regulation and supervision of the financial sector. In addition to providing technical advice on delegated acts and implementing acts adopted by the Commission, ESAs play a central role in the drafting of technical standards, which take the form of regulatory technical standards and implementing technical standards. ESAs can also adopt a vast array of soft law measures, which contribute to the shaping of EU financial regulation. And for the purpose of supervision and enforcement of EU financial law, they have been given the power to adopt binding decisions directly addressed to market participants allowing them, in exceptional circumstances, to bypass the national competent authorities. What emerges is a landscape of a wide range of soft law instruments that appear in different forms and are adopted in different procedures. 3.1

Soft Law Regulatory Tools: Draft Technical Standards and Their Constitutional Challenges

The preparation of draft technical standards is one of the most important regulatory powers the ESAs possess. The central role of the ESAs in the adoption of technical standards enhances the efficiency of the regulation of financial services, which is a technically complex and rapidly changing part of the Union’s financial market, by making use of the expertise of the agency. The extensive conferral of powers to draft technical standards1 requires however some reflections on their role in the EU financial architecture. ESAs can submit draft regulatory or implementing technical standards (hereinafter ‘DRTS’ and ‘DITS’) to the Commission for adoption in the form of delegated or implementing acts.

410  Research handbook on soft law ESAs’ role in the creation of technical standards is considerable as, in contrast with their predecessors, ESAs’ input will be direct and unmediated at the drafting stage (Schammo 2011, p. 1883). The categories of DRTS and DITS reflect the distinction between delegated acts and implementing acts under Articles 290 and 291 TFEU. The former are governed by Article 290 TFEU and shall consist of norms aimed at establishing ‘a single rulebook [and] a level playing field and adequate protection to depositors, investors and consumers across the Union’ (ESA Regulations 2010, recital 22). The latter follow the regime set out in Article 291 TFEU for the adoption of implementing acts. The power to draft regulatory and implementing technical standards has been set clear limitations as these ‘shall not imply strategic decisions or policy choices’ (ESA Regulations 2010, Articles 10(1)(3) and 15(1)(1)). While these limitations aim to provide political re-assurance that the conferral of such powers to the ESA is compatible with the Meroni doctrine,2 it may be argued that the ESAs enjoy a de facto regulatory power, often involving policy choices, in the adoption of such acts with the Commission having only limited opportunities to reject their draft measures (ESA Regulations 2010, recital 23). The procedure for the adoption of DRTS and DITS follows a largely similar path.3 The ESAs prepare technical standards in the form of a draft, which require for their adoption a qualified majority by the Board of Supervisors (ESA Regulations 2010, Article 44(1)(2)).4 Prior to the adoption of the draft, the ESAs need to conduct public consultations, provide a cost-benefit analysis, and seek the Opinion of a stakeholder group specific to each of the ESAs (ESA Regulations 2010, Article 10(1)(3) and Article 15(1)(2)). While making it more transparent and participatory, the consultation process seems to be dominated by large corporate actors and sees only a marginal involvement of consumer groups (Moloney 2014, p. 874; Schemmel 2016, p. 478). The draft is then submitted to the Commission, and, in the case of DRTS, has to forward it, at the same time, to the EP and the Council. The Commission’s adoption is the essential condition for the draft standards to become binding. In cases where the Commission intends not to adopt the draft technical standard, to adopt it in part or to adopt it with amendments, it shall send it back to the competent ESA together with an explanation of its objections (ESA Regulations 2010, Article 10(1)(5) and Article 15(1)(4)). Within a period of six weeks, the ESAs may amend the draft technical standard and resend it as a formal opinion to the Commission for adoption (ESA Regulations 2010, Article 10(1)(5) and Article 15(1)(4)). Crucially, the Commission cannot change the content of the draft technical standard without prior coordination with the ESAs (ESA Regulations 2010, Article 10(1) (7) and Article 15(1)(6)). If the ESAs do not submit an amended standard or have submitted a standard, which does not comply with the Commission’s amendment, the Commission may adopt the standard with the amendments it considers relevant or reject it (ESA Regulations 2010, Article 10(1)(6) and Article 15(1)(5)). The main differences in the process of the adoption of regulatory and implementing technical standards are the following. First, in the case of regulatory technical standards adopted under Article 290 TFEU the EP and the Council may object to the entry into force of the standard within three months from the notification of the DRTS by the Commission (ESA Regulations 2010, Article 13). This is essentially a veto power which prevents the technical standard entering into force. It shows that the two institutions have a role in the decision-making process for the regulatory technical standards. Second, the EP or the Council can revoke the delegation to adopt regulatory technical standards at any time (ESA Regulations 2010, Article 12). The role of the ESAs in the adoption of binding technical standards calls for a number of observations. In relation to the adoption of regulatory technical standards, it should, firstly,

The role of soft law in the context of the financial crisis  411 be noted that the Union legislation conferring such powers is often quite detailed. This might be the consequence of the concerns of the Union legislator about the politically sensitive nature of such legislation, but also might result from a better understanding of the technical issues of financial services regulation in the Union. It certainly seems a move away from the Lamfalussy report’s suggestion to limit Level 1 acts to core political principles. As a result, the powers delegated to the agency do not raise any concerns in respect of the ‘essential elements’ doctrine, which seeks to ensure that the important political choices of a subject matter are made by the Union legislator, in particular where they affect fundamental rights and relations with third countries (Case C-355/10 European Parliament v Council). All the same, the process for the adoption of regulatory technical standards and the powers granted to the ESAs raise Meroni concerns, even in its more relaxed interpretation after the Short Selling ruling (C-270/12 United Kingdom v Parliament and Council). Despite the protestations of Union legislation that DRTS ‘do not involve policy choices’ (EP and Council 2013a, recital 91; EP and Council 2013b, recital 27; ESA Regulations 2010, recital 22), it should be noted that many enabling provisions in EU banking legislation for the adoption of draft regulatory technical standards are broad enough to allow for different regulatory options reflecting a difference in reconciliation of competing policy objectives, such as the competitiveness of the Union’s financial services sector, the protection of financial stability, or the protection of investors (EP and Council 2013b, Article 124(4)). DRTS reflect often a particular choice as to how these competing objectives are reconciled.5 This is made evident by the cost-benefit analyzes, which the ESAs have to produce and which are attached to the DRTS (EBA 2013, section 5.1). Given the importance of some of the policy choices involved, it is argued that the process, which leaves the Commission often very limited room for manoeuvre, may be constitutionally problematic. The ESA Regulations make it clear that the Commission should make amendments only in ‘very restricted and extraordinary circumstances’ (ESA Regulations 2010, recital 23). While the Commission is formally charged with the adoption of regulatory technical standards, the ESA Regulations clarify that the Commission can amend or reject DRTS only where they were ‘incompatible with Union law, did not respect the principle of proportionality or run counter to the fundamental principles of the internal market for financial services’ (ESA Regulations 2010, recital 23). Moreover, the Commission cannot unilaterally revise the content of the ESA’s draft. The ESA Regulations state that ‘[the] Commission may not change the content of a draft regulatory standard prepared by the Authority without prior coordination with the Authority’ (ESA Regulations 2010, recital 23 and Articles 10(1) and 15(1)). The Commission has therefore the choice between adopting the standard (and thereby surrendering its political prerogative as Union executive) or objecting (thereby undermining the design for the adoption of such standards). This is even more delicate, as regulatory technical standards are adopted by the Commission and it takes formal legal and political responsibility for them. This fragmentation of the Union’s executive detaches actual from formal responsibility. Taking into account the institutional design of the ESAs with the Board of Supervisors as decision-maker composed of national supervisors, the formulation of the Union interest seems removed from the Commission as the constitutionally responsible forum in Article 290 TFEU and also undermines democratic legitimacy. The fact that the Commission is represented on the Board of Supervisors, without voting rights, alleviates to some extent its detachment from the actual decision-making forum, but further obscures responsibility in the process for the adoption of regulatory technical standards.

412  Research handbook on soft law In relation to the adoption of implementing technical standards, the first issue concerns the distinction between regulatory acts (following the regime of Article 290 TFEU) and implementing acts (Article 291 TFEU). The distinction is no doubt fraught with difficulty (Türk 2020, pp. 421–422) but needs to be made as a matter of constitutional law. It is therefore confusing when the Court in its ruling in the Biocides case held that ‘the EU legislature has discretion when it decides to confer a delegated power on the Commission pursuant to Article 290(1) TFEU or an implementing power pursuant to Article 291(2) TFEU’ (Case C-427/12 Commission v European Parliament and Council, para. 40). Notwithstanding the constitutional importance of the distinction between Article 290 and 291 TFEU the Court held that it would restrict judicial review ‘to manifest errors of assessment as to whether the EU legislature could reasonably have taken the view’ that the requirements of Article 291 TFEU were met. Even on this softer approach to judicial review, it should be observed that the provision for DRTS and DITS in the recently adopted Capital Requirements Regulation and Directive (EP and Council 2013a and 2013b) is apparently based on political sensitivity of the issues but is questionable from a legal point of view. While the adoption of DITS for the format and timing of reporting (EP and Council 2013b, Article 101(4)) seems politically uncontroversial, such standards can supplement the Union legislation in the same way as other DITS, such as those that are to determine the operational functioning of the colleges of supervisors (EP and Council 2013a, Article 116(5)). The second problem with DITS is that the process for their adoption perverts the premise of Article 291 TFEU, which is based on the Commission adopting implementing acts subject to ‘control mechanisms by Member States’. The procedure, to a considerable extent at least, eliminates thereby the deliberative interaction between the Commission and the administrations of the Member States, which characterizes the comitology regime (Türk 2015, p. 327). Finally, the same objection as against the DRTS can be made in that the endorsement process for DITS detaches formal from actual responsibility. This analysis has shown that the Commission is limited in the activities it can conduct to draft, amend or discard technical standards. The requirement for the adoption of RTS and ITS is perhaps the most significant element in the limitation of the exercise of discretionary powers attributed to the ESAs to ensure that the Meroni doctrine is not bypassed. Yet, the adoption process provides the Commission only with a negative and exceptional power to discard DRTS and DITS. The existence of strict conditions for the Commission to amend or discard the technical standards shows that the ESAs could play a powerful role in shaping the content of technical standards. The system of adoption seems to go beyond a strict reading of the Meroni doctrine. In fact, should the Commission agree on the draft technical standard, the ESAs have the ‘first word’ to make choices that can have the potential of going beyond the rule-making process of the delegator (Busuioc 2013, p. 117). 3.2

Soft Law Regulatory Tools: The Wide Array of ‘Quasi-hard’ Law Powers

The ESAs can also adopt measures in the form of guidelines and/or recommendations addressed to competent national authorities and/or to market players. As stated in Article 16 of the ESAs Regulations, this prerogative serves as a way ‘to ensure efficient and effective supervisory practices’ and ‘to ensure the common, uniform and consistent application of Union law’. These acts are subject to certain procedural requirements: open public consultations, proportionate in relation to the scope, nature and impact and respect cost-benefit concerns.

The role of soft law in the context of the financial crisis  413 ESAs soft law powers appear to be more stringent than mainstream soft law powers. The ESA Regulations clearly specify that the addressees shall ‘make every effort to comply’ (ESA Regulations 2010, Article 16(3)) with these measures. National competent authorities are obliged to report in a clear and detailed way whether they have complied or intend to comply with the guideline or recommendation. In cases where the national authorities do not comply or intend to comply, they are further obliged to state reasons for non-compliance. The ESAs are entitled to publish the fact of non-compliance and the reasons given by the competent authority. As such, peer pressure and disclosure of information to other parties is a system of enforcement with a view to compliance. It is worth noting that guidelines and recommendations are only one part of the regulatory ‘soft law’ tools at the disposal of the ESAs. A wide array of tools also include the supervisory handbook (ESA Regulations 2010, Article 29(2)(2)), the systemic risk tools (ESA Regulations 2010, Article 22), the peer review process (ESA Regulations 2010, Article 30), information management and reporting tools (ESA Regulations 2010, Article 35), methodological tools, and warnings (ESA Regulations 2010, Article 9(3)). Against this background, it is important to analyze further the power of adoption of soft law measures by the ESAs. For instance, the Capital Requirements Regulation and Directive extensively confer the power to adopt guidelines to the EBA, often with reference to Article 16 of the EBA Regulation, but not always (EP and Council 2013a, Article 74(3)). Article 16 of the ESA Regulations may therefore constitute a general enabling provision for the adoption of guidelines (irrespective as to whether a specific legislative provision exists), given that, in contrast to Articles 10 and 15 of the ESA Regulations, no reference is made to powers conferred in acts mentioned in Article 1(2) of the ESA Regulations. It is also interesting to note that a survey of Union legislation enabling the EBA to adopt guidelines reveals a vast amount of different types of such guidelines. One can distinguish between practice-enhancing guidelines, methodology guidelines, benchmark-setting guidelines, implementation/application guidelines, and interpretative guidelines. In particular, the implementation guidelines are often indistinguishable in content from regulatory technical standards (EP and Council 2013a, Article 74(3)). Furthermore, in practice many technical standards that have recently been adopted have as their basis CEBS guidelines and certain legislative provisions make guidelines expressly a precursor to the adoption of technical standards (EP and Council 2013b, Articles 243(6) and 244(6)). Given their practical importance as quasi-technical standards it is important to assess Article 16 Guidelines against wider constitutional principles of Union law. Such guidelines do not raise issues in respect of the powers that can be conferred on Union agencies, where such guidelines do not impose obligations of their own. One could argue that the only binding legal effects of guidelines are those set out in Article 16, which do not however compel the national competent authority to follow the guidelines, but it does impose an obligation to state reasons. ESAs’ soft law powers might act as de facto hard law powers as they produce legal effects which can go beyond the non-binding force of these acts (Schemmel 2016, p. 465; Möllers 2010, p. 392, for CESR guidelines). The Court has acknowledged that soft law instruments might still have legal effects, in that they have to be considered by national courts in deciding national disputes (Case C-322/88 Grimaldi). What is more, Article 16 Guidelines, when followed by national competent authorities, can result in binding national law (Schemmel 2016, p. 474). This is all the more likely given the high level of compliance with Article 16 Guidelines (Schemmel 2016, p. 464; Eliantonio et al. 2021 for country analyzes), which, it has

414  Research handbook on soft law been argued, can be attributed to the public instances of ‘naming and shaming’, which act as an element to ensure enforcement (Busuioc 2013, p. 118). Therefore, even without apparent legally binding effect, ESA guidelines can have indirect legal effects, which can raise concerns, in particular in respect of accountability and legal certainty (Tridimas 2012, p. 72). Such effects could also arise in relation to Articles 17 (breach of Union law) and 19 (mediation) of the ESA Regulations. While guidelines do not constitute a body of law, the breach of which could trigger the application of Article 17, it is possible to envisage a situation in which an ESA in its interpretation of Union law (including DRTS and DITS) relies on its guidelines.6 For example, Article 11 of Directive 2006/48/EC requires that the competent authorities shall only grant an authorization where the persons directing the business of the credit institution are of ‘sufficiently good repute [and have] … sufficient experience’. This requirement has been the subject of Article 16 guidelines by the EBA (EBA, 2012). In this case, the guidelines, while not binding on the competent authority, will have a prejudicial effect on the outcome of the case, while any final decision rests with the Court of Justice. It should however be made clear that a guideline cannot, in and of itself, impose directly applicable requirements in the sense of Article 17(6) of the ESA Regulations. Conversely, however, guidelines bind the ESAs in the exercise of their power under Article 17. Where the ESAs have published a guideline, the European Central Bank (ECB) and national authorities (and financial institutions) should be entitled to rely on the guidelines. While the ESAs can amend those guidelines for the future, the principle of legitimate expectations would, in principle, preclude any deviation in a single case decision (such as Article 17). As regards Article 19, it is submitted that also in this case guidelines cannot form part of the body of Union law, compliance with which the EBA may ensure under this provision against national authorities and financial institutions (ESA Regulations, Article 19(3) and (4)). Even if the Short Selling judgment of the Court (Case C-270/12 United Kingdom v Parliament and Council) has somewhat alleviated the constitutionally precarious nature of Articles 17 and 19 of the ESA Regulations, it is not entirely clear why the power granted to the ESAs in those provisions should not lie with the Commission according to Article 291 TFEU, raising concerns about the institutional balance and democratic legitimacy. Finally, some specific mention should be made of other soft law measures such as the European supervisory handbook that, as a collection of supervisory practices, plays a fundamental role in soft law regulation in financial markets (ESA Regulations 2010, Article 29(2)). In fact, such a tool has the potential to harmonize supervisory practices in the EU. This may happen in two ways. First, in the short term, it can enhance supervisory practice. All the same, while a valuable tool for the ECB and national authorities, the concerns are here the same as for guidelines. Second, in the long term, its provisions will no doubt act as precursors to guidelines, technical standards, and even legislative provisions. The issue will be to identify which provisions should be included in legal rules and which should be best left to informal practices. A trend towards more legislation is not necessarily always the best option. Also, on the path from practice to guideline to a technical standard, the issue of political judgement and the importance of process values might become relevant. 3.3

Soft Law Supervisory and Enforcement Tools: The Mix of Soft and Hard Law

ESAs have a wide range of market control and supervisory powers over market operators and national supervisors. These powers are characterized by being embedded in lengthy

The role of soft law in the context of the financial crisis  415 procedures that contain soft law and ultimately hard law components. In this case, soft law measures operate in the shadow of the hard law decisions that can be taken at the end of those procedures. One of the essential powers of ESAs is to ensure the consistent application of EU rules. Article 17 of the ESA Regulations contains provisions that allow them to make use of a structured enforcement procedure which is parallel to the general enforcement procedure under Article 258 TFEU. The procedure is divided into different phases that resemble the procedure under Article 258 TFEU. In the first phase, the provision provides for an investigative formal phase where the competent ESA carries out an investigation on its own initiative7 or on the request of an institution or a competent authority. The competent authority against which the investigation is carried out has the obligation to provide the ESA with all the information which it considers necessary for the investigation. The power of investigation concerns the application of banking, securities and occupational pension scheme law or the regulatory and implementing technical standards adopted by the ESA. During the second phase, the ESA will, no later than two months from the initiation of the procedure, address recommendations to the competent authority with a view to setting out the necessary action to comply with EU law.8 This phase can result in an informal agreement between the acting parties, which would avoid the issuance of the formal recommendation by the competent ESA. The third phase includes the participation of the Commission. In cases where the competent authority has not followed the ESAs’ recommendations, the Commission will be involved in the process and may issue a formal opinion to the competent authority. Where a competent authority does not comply with the formal opinion of the Commission, the ESA, ‘where it is necessary to remedy in a timely manner such non-compliance in order to maintain or restore neutral conditions of competition in the market or ensure the orderly functioning and integrity of the financial system’ (ESA Regulations, 2010, Article 17(6)), is empowered to adopt decisions with binding effect for financial institutions to ensure compliance with directly applicable Union legislation. The ESA Regulations also provide in Article 18 for a system of intervention powers in emergency situations. These powers are triggered in cases of adverse developments which may seriously jeopardize the orderly functioning and the integrity of the whole or part of the financial system. The ESAs must actively facilitate and coordinate actions undertaken by the relevant competent supervisory authorities. Any formal decisions of the ESAs are however dependent on a formal decision by the Council as to the existence of an emergency situation. It appears that discretion is only partial and the Council’s emergency decision constitutes the essential boundary of the exercise of emergency powers by the ESAs (Tridimas 2012, p. 76). Where the relevant competent authority fails to address the ESAs’ decision, the ESA can adopt an individual decision addressed also to market actors, including the cessation of any practice. The arrangements contained in Articles 17 and 18 of the ESA Regulations show that the process of adoption of these decisions has been the object of extensive negotiations (Moloney 2014, pp. 882–884). The result appears as a ‘patchwork’ procedure where the legislator has strived for a sound institutional balance. Given their cumbersome, and in the case of Article 18 highly exceptional, nature, combined with the sensitive point that the Board of Supervisors will have to investigate, and decide, a case against one of its own members, the ESAs are likely to be reluctant to proceed to hard law decisions and will instead aim for an informal resolution. Further, the ESAs’ Regulations contain provisions for the settlement of disagreements between national supervisory authorities (ESA Regulations 2010, Article 19). The role of the

416  Research handbook on soft law ESAs in this case is threefold as they are mediators, decision-makers and enforcers. When competent authorities disagree on the procedure or the content of an action or inaction of a competent authority of another Member State, ESAs’ may be called on to settle the disagreement. ESAs can intervene at the request of one or more of the competent authorities or on their own motion. The system to settle disagreements could play an important role, even more than the enforcement actions discussed above. The role of the ESAs in this case is stronger than in the enforcement procedure as no reference is made to the Commission. It is submitted that this situation makes the ESAs more powerful than would appear acceptable under the Meroni doctrine, as the ESAs can use some discretion in evaluating the disagreement between competent authorities.9 Nonetheless, in the context of these powers, the safeguard for fiscal responsibilities constitutes a strong limitation for ESAs. It provides that ESAs’ decisions shall not impinge on fiscal responsibilities of the Member States (ESA Regulations 2010, Article 38).10 This provision gives ground to Member States to contest the validity of an ESA’s decision if this impinges on the national fiscal responsibilities. The provision does not define the term ‘fiscal responsibility’. It appears that this safeguard procedure expresses the limited powers that ESAs can have in practice vis-à-vis Member States as it acts as an ex post form of control to the adoption of ESAs’ decisions under Articles 18 and 19.

4.

TOWARDS A MORE NUANCED UNDERSTANDING OF LEGAL EFFECTS OF SOFT LAW IN THE FINANCIAL SECTOR

The wide array of soft law that has emerged from the financial crisis requires a renewed effort to understand the differentiated legal effects of soft law in this field. Consequently, a more nuanced understanding of the legal effects of the various instruments that come together under the umbrella of ‘soft law’ will emerge that will obviate the need to refer to the notion of ‘soft law’ at all. In this final section, a framework for the assessment of legal effects (Türk 2017; Türk and Xanthoulis, 2019, pp. 154–156) will be set out and then be applied, as a test case, to the particularly thorny issue of the legal effects of Article 16 guidelines. A proper understanding of legal effects for the purposes of analyzing ‘soft law’ instruments is particularly important for the issue of judicial accountability. Given the general difficulties of other accountability mechanisms for ‘soft law’, the answer to what avenues for judicial review are available for ‘soft law’ is of particular pertinence. 4.1

A Legal Effects Framework for Judicial Review

Any assessment as to the appropriate avenue for challenging Union acts and the allocation of jurisdiction between the Union and national courts depends on one’s understanding of the legal effects of Union acts. A Union act that is not considered as reviewable in the meaning of Article 263 TFEU cannot be challenged directly in the Union courts but is only open to indirect challenge, either on the basis of a plea of illegality under Article 277 TFEU in proceedings before the Union courts or through a reference from a national court under Article 267 TFEU. The same is true where an applicant is not deemed to be directly concerned in the meaning of Article 263(4) TFEU. Where an act entails implementing measures the applicant must, in

The role of soft law in the context of the financial crisis  417 addition to direct concern, also show individual concern, an often high threshold that, if not met, leads to the inadmissibility of a direct action under Article 263 TFEU and means that the applicant has to pursue indirect means of review. As will be shown below, the determination of such issues will depend on the legal effects of the Union acts in question. Conceptual clarity as to the determination of legal effects in Union law is therefore central to any understanding of judicial review of Union acts generally and ‘soft law’ in particular. 4.1.1 The notion of legal position and primary legal effects A person’s legal position consists of rights and obligations created within a legal system. Any legally binding alteration of a person’s rights and obligations constitutes therefore a change in a person’s legal position and therefore produces legal effects. Legal obligations could be defined as duties to do something or refrain from doing something. In this respect, one can distinguish between direct legal obligations and indirect legal obligations. The former arise where an act itself imposes new or modifies existing legal obligations.11 The latter are created where an act triggers obligations, which have their source in other rules or principles of the legal system.12 As regards the determination of rights, a similar distinction between direct and indirect determinations of rights can be made. Such determinations can take various forms, such as the grant of a right, or its modification, or its denial.13 Finally, it should be noted that an act can have more than one legal effect (Case T-393/06 Makhteshim-Again and Others v Commission, at para. 37). 4.1.2 Primary and secondary legal effects The classification above encapsulates legal effects that either directly or indirectly result from a legal act either by imposing an obligation on a person and/or determining their right. Such legal effects can be considered as primary legal effects in a legal system, as they lead to a change in the legal position of a person without the intervention of another competent authority. Changes in the legal system can however also occur as a result of secondary legal effects of an initial act. This is the case where a primary act provides the basis for a subsequent act, which produces legally binding effects. There is a final category of acts, namely those that do not entail any legal effects at all. This can be difficult to determine in individual cases, as the case law demonstrates.14 Where an act has primary legal effects only a direct action under Article 263 TFEU would be available, while acts that have secondary legal effects can only be reviewed indirectly under Articles 277 and 267 TFEU. On the other hand, where acts that do not have any legal effects Union institutions could only be liable under the more stringent requirements imposed for compensation claims under Article 340(3) TFEU. 4.2

Article 16 Guidelines and Legal Effects

Article 16 Guidelines (ESA Regulations, Article 16) are considered by the Court as not legally binding and therefore do not constitute reviewable acts under Article 263 but can be reviewed as part of a validity reference from a national court under Article 267 TFEU (Case C-911/19 FBF).15 As discussed above, the essential feature of Article 16 Guidelines is that they do not require their addressees, national competent authorities and financial institutions, to comply with them. But crucially, while they do not impose an obligation to comply, they do impose other

418  Research handbook on soft law obligations. Article 16(3) requires both national competent authorities and financial institutions to ‘make every effort to comply’. This exhortation is then followed with concrete legal obligations, but it is important to note that they are only imposed on national competent authorities. Article 16(3) requires authorities to confirm whether they comply or intend to comply with the guideline. Where they do not comply or do not intend to comply, they shall inform the ESA, stating their reasons. These obligations are then followed by a ‘name and shame’ mechanism, which obliges the ESA to publish the fact that a national competent authority does not comply or intend to comply with the guidelines. A more limited reporting obligation is imposed on financial institutions, which have to report their compliance, ‘in a clear and detailed way’ only where the guideline imposes this duty. It seems obvious that Article 16 Guidelines have primary legal effects on national competent authorities by imposing legal obligations. While not imposing an obligation to comply with the substantive provisions of the guideline, such guidelines trigger a general duty to confirm compliance and in case of non-compliance impose a duty to state reasons. Similarly, where guidelines impose such a duty, financial institutions also have a reporting obligation and therefore the guideline, in this case, will also have primary legal effects on them. As a consequence, where such a duty exists the guideline constitutes a reviewable act under Article 263 TFEU. Article 16 guidelines do, however, also have primary legal effects for national competent authorities as they constitute the basis for the adoption of national measures creating legal effects.16 Article 16 guidelines do, however, also produce secondary legal effects. This can be the case where the Guideline (as a primary act) constitutes the basis for a national competent authority (constituting primary legal effects for the national competent authority) to adopt measures (as secondary acts) that bind individuals. In this case, the individual can raise the invalidity of the guideline indirectly in proceedings in national courts in which it challenges the national measure adopted by the national competent authority.

5. CONCLUSION Soft law instruments constitute an important part of the ESAs’ repertoire to contribute to ‘the establishment of high-quality common regulatory and supervisory standards and practices’ (ESA Regulations 2010, Article 8(1)(a)) and to a ‘consistent, efficient and effective application’ (ESA Regulations 2010, Article 8(1)(b)) of Union financial legislation. Without them supervision and enforcement of Union financial legislation would be hard to achieve. Many of the currently used ‘soft law’ instruments have been developed by CESR and other committees during the financial crisis often with doubtful legal bases and often in the absence of robust accountability mechanisms. As this chapter has shown, while some of these defects have been cured with the establishment of the ESAs, questions remain as to whether the accountability mechanisms are sufficient. This is all the more so, as a bewildering array of ‘soft law’ instruments are available often with very different objectives. This chapter has argued for a more nuanced assessment of the legal effects of such soft law measures and has offered a legal effects framework for the purpose of finding appropriate avenues of their judicial review. While the application of this framework has been limited, due to time and space constraints, to Article 16 guidelines as a test case, the framework can be applied to other ‘soft law’ instruments as well. Such a more

The role of soft law in the context of the financial crisis  419 nuanced assessment will avoid the sui generis escape from what Advocate General Bobek called the Court’s binary view of ‘soft law’ as either binding or not (Advocate General Bobek, Case C-911/19 FBF, at para. 53), not to mention the more robust Games of Thrones metaphors of his Opinion.

NOTES 1. Regulation 575/2013 contains some 59 legal bases for the adoption of DRTS and 21 for DIRS; Directive 2013/36 contains 13 and 11 respectively. 2. As re-interpreted in Case C-270/12 United Kingdom v Parliament and Council. 3. See Articles 10–14 of the ESA Regulation for DRTS and Article 15 of the ESA Regulations for the DITS. 4. Article 44(1)(2) of the EBA Regulation (Regulation 2010/1093) requires, in addition, a simple majority of participating and of non-participating members. 5. See e.g., pp. 20 and 25 of EBA/RTS/2013/07 on large exposures. The EBA has of course to respect the limits of the enabling provision. Doubtful is therefore EBA/RTS/2013/10, which extends the information categories beyond those listed in Article 50(6) of Directive 2013/36. 6. It is, however, clear from the ruling in Case T-660/14 SV Capital OÜ v EBA, paras 66–72, that Article 17 of the ESA Regulations can only be employed in case of non-application or a breach of Union law set out in Article 1(2) of the ESA Regulations, as well as any DRTS and DITS. 7. The initial position of the Board of Appeal has been that, while the ESAs have discretion as to whether to investigate breaches of Union law by competent authorities on request by a private party, they must nevertheless exercise this discretion properly. See ESA Board of Appeal, Decision of the Board of Appeal of 24 June 2013, SV Capital OÜ v EBA, para. 34; ESA Board of Appeal, Decision of the Board of Appeal of 14 July 2014, SV Capital OÜ v EBA. This view has, however, not been shared by the General Court which, in Case T-660/14 SV Capital OÜ v EBA, at paras. 48 and 49, ruled that the lack of procedural rights in the investigation procedure means that they cannot challenge a decision by the EBA not to investigate an alleged breach by a competent authority. It is similarly clear from that judgment, in particular at paras 64–72, that the Board of Appeal cannot review a rejection by an ESA of a request by a private party to employ its power under Article 17 of the ESA Regulations. This position has not substantially changed with the recent amendment to Article 17(2), which sought to enhance the role of private parties by allowing their own initiative application by the ESA, including when this is based on well-substantiated information from natural or legal persons. See ESA Board of Appeal, see Decision of the Board of Appel of 12 March 2021, A v ESMA. 8. Such recommendations while not subject to a review under Article 263 TFEU can be subject to a validity review under Article 267 TFEU, see Case C-501/18 Balgarska Narodna Banka, at paras 82 and 83. 9. See, however, the reference in Article 19(3) of the ESA Regulations to the use of the power to ensure ‘compliance with Union law’ indicating a mere ‘legality’ review of competent authorities. However, a wider reading of the scope of Article 19 can be justified on the ground that Article 19, as a procedure, in addition to Article 17 would otherwise be superfluous. 10. The omission of Article 17 in Article 38 provides a further argument for a wider reading of the scope of Article 19. 11. For example, a Commission act ordering the Member State to recover aid declared incompatible with the internal market, as in Case C-274/12 Telefónica v Commission. 12. For example, the act identifying a substance as a result of the procedure set out in Article 59 of the ECHA Regulation. While the act itself is merely a factual determination, it triggers information obligations laid down in other provisions of the ECHA Regulation. See Case T-94/10 Rütgers and Others v European Chemicals Agency, at para. 33. 13. For example, a direct determination of rights is made by the act to consider aid granted by a Member State incompatible with the internal market, as in Case T-154/10 France v Commission, at para. 40.

420  Research handbook on soft law 14. Case T-507/13 SolarWorld and Others v Commission, at para. 48; Case T-37/11 Hungary v Commission, at paras 41–43. See also Case T-42/13 1. garantovaná v Commission, at para. 27. But see Case T-275/94 Groupement des Cartes Bancaires v Commission, at para. 32. Case T-22/07 US Steel Košice v Slovak Republic, at para. 47. Case C-261/13P Peter Schönberger v European Parliament, at para. 24. 15. For a more differentiated view, see the Opinion of AG Bobek in this case, Case C-911/19 FBF. 16. In certain instances, national law imposes an obligation for the NCA to comply with an Article 16 Guideline (Schemmel 2016, p. 474). In this case it can be argued that the guideline indirectly imposes an obligation to comply. Indirectly, because the obligation does not result from the guideline or Article 16 but is triggered by national law.

REFERENCES Busuioc, M., (2013). Rule Making by the European Financial Supervisory Authorities: Walking a Tight Rope. European Law Journal. 19, 111–125. Case C-322/88 Grimaldi ECLI:EU:C:1989:646. Case T-275/94 Groupement des Cartes Bancaires v Commission ECLI:EU:T:1995:141. Case T-393/06 Makhteshim-Again and Others v Commission ECLI:EU:T:2008:531. Case T-22/07 US Steel Košice v Slovak Republic ECLI:EU:T:2009:158. Case T-94/10 Rütgers and Others v European Chemicals Agency ECLI:EU:T:2013:107. Case T-154/10 France v Commission ECLI:EU:T:2012:452. Case C-355/10 European Parliament v Council ECLI:EU:C:2012:516. Case T-37/11 Hungary v Commission ECLI:EU:T:2012:310. Case C-270/12 United Kingdom v Parliament and Council ECLI:EU:C:2014:18. Case C-274/12 Telefónica v Commission ECLI: EU:C:2013:852. Case C-427/12 Commission v European Parliament and Council ECLI:EU:C:2014:170. Case T-42/13 1. garantovaná v Commission ECLI:EU:T:2013:621. Case C-261/13 P Peter Schönberger v European Parliament ECLI:EU:C:2014:2423. Case T-507/13 SolarWorld and Others v Commission ECLI:EU:T:2015:23. Case T-660/14 SV Capital OÜ v EBA ECLI:EU:T:2015:608. Case C-501/18 Balgarska Narodna Banka ECLI:EU:C:2021:249. Case C-911/19 FBF ECLI:EU:C:2021:599. Committee of European Securities Regulators., (2006). Charter of the Committee of European Securities Regulators. CESR/05-758. Committee of European Securities Regulators. Council., (2009). Council conclusions on strengthening EU financial supervision. 2948th Economic and Financial Affairs, Luxembourg, 9 June 2009. Brussels: Council. [Viewed 7 July 2022]. Available from: https://​www​.consilium​.europa​.eu/​uedocs/​cms​_data/​docs/​pressdata/​en/​ecofin/​110617​.pdf. Eliantonio, M., Korkea-aho, E., and Ştefan, O., eds., (2021). EU Soft Law in the Member States Theoretical Findings and Empirical Evidence. Oxford: Hart. European Banking Authority., (2012). Guidelines on the assessment of the suitability of members of the management body and key function holders. EBA/GL/2012/06. Paris: European Banking Authority. [Viewed 7 July 2022]. Available from: https://​www​.eba​.europa​.eu/​sites/​default/​documents/​files/​ documents/​10180/​106695/​7d1a34fc​-0a74​-4020​-a14c​-358ade55237d/​EBA​-GL​-2012​-06​-​-Guidelines​ -on​-the​-assessment​-of​-the​-suitability​-of​-persons​-​.pdf. European Banking Authority., (2013). EBA FINAL draft Regulatory Technical Standards on the determination of the overall exposure to a client or a group of connected clients in respect of transactions with underlying assets under Article 390(8) of Regulation (EU) No 575/2013. EBA/RTS/2013/07. Paris. European Banking Authority. [Viewed 7 July 2022]. Available from: https://​www​.eba​.europa​.eu/​ sites/​default/​documents/​files/​documents/​10180/​513001/​dc888876​-cab6​-43b8​-92e9​-bc563d9b95b8/​ EBA​-RTS​-2013​-07​%20​%28Determination​%20of​%20exposures​%20​%29​.pdf. European Central Bank., (2005). Review of the Application of the Lamfalussy Framework to EU Securities Market Legislation. The Governing Council, 17 February 2005. Frankfurt: European

The role of soft law in the context of the financial crisis  421 Central Bank. [Viewed 7 July 2022]. Available from: https://​ www​ .ecb​ .europa​ .eu/​ pub/​ pdf/​ other/​ lamfalussy​-reviewen​.pdf. European Commission., (1966). Segré report. The development of a European Capital Market. Brussels: European Commission. [Viewed 28 May 2022]. Available from: http://​ ec​ .europa​ .eu/​ economy​ _finance/​emu​_history/​documentation/​chapter1/​19661130​en382devel​eurocapitm​_a​.pdf. European Commission., (1985). Completing the internal market. COM(85) 310 final. Brussels: European Commission. European Commission., (1999). Communication from the Commission – Implementing the Framework for Financial Markets: Action Plan. COM(1999) 232 final. European Commission., (2009). Commission Decision 2009/77/EC establishing the Committee of European Securities Regulators, OJ [2009] L 25/18. European Council., (2009). Presidency conclusions – 18/19 June 2009. 11225/2/09 REV 2. Brussels: European Council. [Viewed 7 July 2022]. Available from: https://​www​.consilium​.europa​.eu/​uedocs/​ cms​_data/​docs/​pressdata/​en/​ec/​108622​.pdf. European Parliament., (2005a). European Parliament resolution on the current state of integration of EU financial markets. OJ [2006] C 45E/140. European Parliament., (2005b). European Parliament Resolution on Clearing and Settlement in the EU. OJ [2006] C 157E/485. European Parliament and Council., (2010). Regulation 1093/2010, [2010] OJ L 331/12; Regulation 1094/2010, [2010] OJ L 331/48; Regulation 1094/2010, [2010] OJ L 331/84 (cited as ESA Regulations, 2010). European Parliament and Council., (2013a). Directive 2013/36, [2013] OJ L 176/338. European Parliament and Council., (2013b). Regulation 575/2013, [2013] OJ L 176/1. European Securities Authority ESA Board of Appeal, Decision of the Board of Appeal of 24 June 2013, SV Capital OÜ v EBA (Decision EBA C 2013 002). European Securities Authority ESA Board of Appeal, Decision of the Board of Appeal of 14 July 2014, SV Capital OÜ v EBA (Decision BoA 2014 - C1 02). European Securities Authority ESA Board of Appeal, Decision of the Board of Appeal of 12 March 2021, A v ESMA (Decision D 2021 02). Moloney, N., (2014). EU Securities and Financial Markets Regulation. 3rd ed. Oxford: Oxford University Press. Möllers, T., (2010). Sources of Law in European Securities Regulation – Effective Regulation, Soft Law and Legal Taxonomy from Lamfaussy to de Larosière. European Business Organization Law Review. 11, 379–407. Opinion of AG Bobek in Case C-911/19 FBF ECLI:EU:C:2021:294. Schammo, P., (2011). The European Securities and Markets Authority: Lifting the Veil on the Allocation of Powers. Common Market Law Review. 48(6), 1879–1913. Schaub, A., (2005). The Lamfalussy Process Four Years On. Journal of Financial Regulation and Compliance. 13(May), 110–120. Schemmel, J., (2016). The ESA Guidelines: Soft Law and Subjectivity in the European Financial Market – Capturing the Administrative Influence. Indiana Journal of Global Legal Studies. 23(2), 455–504. The De Larosière Group., (2009). Report of the High-Level Group on Financial Supervision in the EU. Brussels: The De Larosière Group. [Viewed 28 May 2022]. Available from: https://​ec​.europa​.eu/​ economy​_finance/​publications/​pages/​publication14527​_en​.pdf. Tridimas, T., (2012). Financial Supervision and Agency Power: Reflections on ESMA. In: N. Shuibhne and L. Gormley, eds. From Single Market to Economic Union. Oxford: Oxford University Press, pp. 55–83. Türk, A., (2015). Comitology, In A. Arnull and D. Chalmers, eds. The Oxford Handbook of European Law. Oxford: Oxford University Press, pp. 327–349. Türk, A., (2017). Liability and accountability for policies announced to the public and for press releases. In: European Central Bank, ed. ECB Legal Conference 2017, Shaping a new legal order for Europe: a tale of crises and opportunities. Frankfurt, European Central Bank. Türk, A., (2020). Legislative, delegated acts, comitology and interinstitutional conundrum in EU law. European Law Journal. 26(5-6), 415–428,

422  Research handbook on soft law Türk, A. and Xanthoulis, N., (2019). Legal accountability of European Central Bank in bank supervision: A case study in conceptualizing the legal effects of Union acts. Maastricht Journal of European and Comparative Law. 26(1), 151–164.

27. Soft law governance in the field of AI: a European perspective Frederik Schade and Mikkel Flyverbom

1. INTRODUCTION Soft law and other governance initiatives in the area of artificial intelligence (AI) remain scattered and under-developed. However, the importance of this issue area is increasingly acknowledged by researchers, policymakers, and the public at large. As digital technologies, data and algorithms become integrated into more and more parts of society, it raises questions about their consequences for a wealth of issues, such as competition, surveillance, copyright, freedom of expression, violent content, workers’ rights, and the protection of children. As we outline in this chapter, the governance of AI is a regulatory domain in the making – marked by a complexity of initiatives and mechanisms that mobilize both hard and soft forms of law in the attempt to make novel technologies governable and amenable to democratic control. To make sense of these developments, we rely on both academic literature and policy papers addressing AI as an object of governance. There is a growing literature on the societal importance of digital technologies, data, and algorithms, and on the emergence of global governance efforts and institutions in this domain (Bowrey 2005; DeNardis 2009; Flyverbom 2011; Gillespie 2017; Hofmann et al. 2016; Mueller 2010). In policy circles, particularly in the European Commission and related institutions, policymakers and national governments increasingly focus on governance measures. These range from hard laws such as the EU’s General Data Protection Regulation (GDPR), put into effect in 2018, and multiple soft law efforts ranging from nearly legally mandatory practices and sanctions to voluntary initiatives, guidelines and certification programs, as well as proposals for new legislation. In the public sphere, there is a growing interest in the need to regulate tech companies that have become extremely powerful and more focused on profits than societal responsibilities, made clear via the exposure of problematic practices such as those of Cambridge Analytica and corporate tactics exposed in US congressional reports (Nadler 2020). Taken together, these developments attest to the increasingly central role that digital technologies, platforms and data-driven, algorithmic approaches play in contemporary society. It is against this backdrop that we articulate the emergence and current shape of soft law governance in the area of artificial intelligence. For a working definition of ‘soft law’, we rely on the work of Abbott and Snidal (2000) and others distinguishing between ‘hard law’ as a set of legally binding obligations versus ‘soft law’ as a broader class of norm development and political arrangements where the element of legalization is increasingly absent. With regards to AI, the space of ‘hard’ and ‘soft’ governance emerging around these technologies is constituted by a transnational, cross-sectoral and otherwise complex socio-political set of arrangements that have yet to be fully institutionalized in global governance. For these technologies, the policy space cuts across many institutions, and involves both policy and technical specialists, human rights activists, government officials, business leaders and directors of 423

424  Research handbook on soft law international organizations. As such, the global governance of digital technology remains under construction although a lot has happened over the past two decades (Flyverbom 2011). Furthermore, the current vocabulary of and framing for the technological developments under scrutiny remain somewhat loosely defined. Artificial intelligence has emerged as an umbrella term for the growing reliance on data, algorithms, and digital systems in both the public and the private sector. But the term ‘artificial intelligence’ is mostly shorthand for a long list of technological solutions and developments that we are accustomed to understand in different terms. As Alaimo and Kallinikos (2020) suggest, we are really talking about a socio-technical apparatus consisting of data, algorithms, and machineries with a long historical lineage as sources of digital transformations and automation. Generally, what makes these uses of digital solutions worthy of the descriptions as ‘artificial’ and ‘intelligence’ is the way in which they are perceived or envisioned to replace human processes of understanding and intervention (Smith 2019). Still, however, most projects in this domain suffer from high error margins, operational difficulties, problems with bias, and a very reduced ability to deliver human-like forms of intelligence beyond simply copying what humans have done in the past (O’Neil 2018). But emergent forms of generative artificial intelligence, such as ChatGPT, certainly point to the need for reflections about the future of technology and how technology can serve democratic societies and responsible societal developments. Recent years have seen increased attention being paid to AI as an object of potential regulation. In 2020, the European Commission issued a White Paper on Artificial Intelligence: a European approach to excellence and trust which, in 2021, was followed by a regulatory proposal on harmonized rules regarding such technologies (European Commission 2020; European Commission 2021). As we will describe in this chapter, the regulatory approach suggested by the European Commission is a risk-based approach, which entails a particular logic of power and set of governance rationalities. This is likely to entail the prohibition of certain select uses of AI technology followed by ‘hard’ regulatory requirements for use cases deemed ‘high risk’. These developments suggest – as do research (e.g., Marchant, et al. 2020; Marchant 2019) – that the field of AI technologies is likely to be only partially regulated by law, leaving what some scholars call a ‘governance gap’ for a significant variety of ‘low’ or ‘medium’ risk systems. Such a gap, they suggest, is likely to be addressed through various ‘soft law’ initiatives, i.e., non-binding (in a regulatory sense) norms, standards, and guidelines which nonetheless enact certain professional – or even moral and ethical – expectations and norms across the professional field of AI. In this chapter, we aim to provide a partial overview of these developments by considering (1) the ways in which AI is increasingly construed as an object of regulation, relying on both ‘harder’ to ‘softer’ modes, and (2) the ways in which such regulatory and/or norm-setting interventions are sought to be carried out in practice through a set of concrete governance mechanisms involving diverse actors and rationalities of governance. This analysis is not meant to be fully encompassing or take into account all recent developments in the field of AI regulation and their implications. Rather, we focus on certain trends and tendencies to describe what we find to be some relatively general characteristics of an emergent governmental regime which targets AI technologies and their particular usages from different points of view. Finally, we undertake a discussion of the current state of this field including some seeming ambiguities, certain paradoxes and complexities related to AI regulation which seem to us to constitute some prevailing challenges even in the context of – and some of them perhaps even caused or problematized further by – new regulatory initiatives.

Soft law governance in the field of AI  425

2.

THEORETICAL FRAMEWORK AND APPROACH

Our theoretical framework builds on insights from Foucauldian governmentality studies which provides a valuable entry point for analyzing and discussing governance and acts of government which go beyond the enforcement of state-based regulation and ‘hard’ law only (Foucault 1991). By conceptualizing power as a productive rather than a limiting force, such approaches focus more on the ‘conditions of possibility’ (Foucault 1994), political rationalities and implicit ways of thinking that shape governance efforts, and less on explicit intentions and concrete instruments of the actors involved. By seeing governance not as acts of continuous limitation, but rather as particular way of directing action, a governmentality perspective provides a theoretical vantage point from which to analyze a variety of modes of government (including both ‘harder’ and ‘softer’ regulatory modalities). Following Foucault, we can think of governance in general as a particular form of power enactment or steering which a variety of institutions (of various kinds, not merely states) engage in, through both direct and indirect means of steering and directing action. Thus – in line with a governmentality perspective – we will be thinking of governing broadly as the conduct of conduct which works not merely to limit the action of certain actors, but rather to structure the field of possible action of those actors (Foucault 1982). More concretely, we will be considering emergent forms of governance measures and instruments arising in the field of AI with a focus on how and according to which governmental rationalities such technologies of government seek to structure a particular field of (future) social action (Miller and Rose 1990). This allows us to cast a rather wide net when it comes to discussing both how this policy domain is being constructed as an object of governance and how it is getting populated with concrete governance initiatives with consequences for technologies, citizens, and societal institutions (see Djelic and Sahlin-Andersson 2006). As our analyzes below unpack, much of emergent AI governance is underpinned by political rationalities and logics that revolve around risk management and the balancing of opportunities and dangers associated with technology development. These logics also shape the concrete ways in which governance gets institutionalized and practiced. Our attention to how the governance of AI is constructed through arrangements of particular (ideational) political rationalities and more (concrete) technologies of governance is clearly indebted to governmentality studies (Miller and Rose 1990). Thus, we approach the governance of AI with a dual focus on the construction of AI as an object of governance and on the shape, actors, and dynamics of the governance mechanisms being institutionalized. This allows us to develop an early mapping and conceptualization of soft law in the area of AI and digital technology. Given that such instruments in the context of AI are still very young and still undergoing formulation and redefinition this work may come across as slightly hypothetical and speculative in character. However, it does seem worthwhile to try to ascertain the current state of this field of governance and consider the direction taken by institutions of government to construct and re-organize a field of ‘responsible AI’ at this early stage. Such work will allow us to discuss the current rationalities and directionalities taken up by emergent policy initiatives while these are still undergoing formulation and thus create new spaces for ethical reflection which may yet affect the direction taken by future AI governance initiatives.

426  Research handbook on soft law

3.

AI AS AN OBJECT OF REGULATION

The global governance of the internet, digital technologies, and data has emerged as an increasingly salient question in politics over the last decade. In the early 2000s, many national governments and international organizations such as the United Nations realized that digital technologies and their consequences for societal development had to be established as a policy domain and addressed through institutional processes (Flyverbom 2011). These conversations played out at the intersection of governments, civil society, businesses, and other stakeholders, largely in the context of intergovernmental institutions with a history of addressing issues related to telecommunications and other technological developments. In recent years, we have seen a rise of tech giants as dominant forces in the tech space. Such companies provide platforms for communication and other forms of exchange, offer possibilities for information searches, and possibilities for social connections. Also, these companies are largely setting the conditions for the development and operations of digital spaces, including the dominance of advertisement-driven, free services and widespread data harvesting that challenges privacy and other established rights. Such processes of digitalization and datafication have paved the way for a wide-reaching set of transformations that impact not only users, but also companies, public sector organizations and society at large. And while data-driven, algorithmic operations early on were largely limited to social media, search engines and other digital platforms, they are increasingly being integrated into the rest of society, including in the provision of public sector services and other settings that citizens and organizations depend on and cannot withdraw from. These developments raise the stakes and pave the way for demands about more careful and regulated approaches to the use of digital, datafied solutions. Taken together, these digital transformations give rise to deep, societal developments, such as the emergence of new forms of capitalism, new modes of production, new approaches to the spread and editing of information, and the need for new forms of governance. It is against this backdrop that we highlight a range of attempts to develop new forms of governance and soft law initiatives focusing on artificial intelligence and other digital transformations. With the ubiquity of digital solutions, the tech domain has become more central in policymaking and is emerging as a policy area that many governments and international organizations care about. The result is both an increase in regulatory initiatives, a focus on the need for more active enforcement of existing and relevant legislation, and a widespread and a growing focus on ethics and responsibility in the tech space. While these developments are important, they remain scattered and take rather limited forms. Governance initiatives in the realm of digital technology, data and algorithms are often rather abstract, focusing on principles. Or they consist of patchy solutions that address separate and individual parts of the domain in a somewhat piecemeal fashion. These efforts include solutions such as informed consent, cookie regulation, ethics initiatives, labelling schemes, and other ways of addressing a wealth of problems propelled by the rapid development of digital solutions. But the deeper concerns raised by data-driven approaches, algorithms and artificial intelligence are often left to the side. In this context, AI has been widely discussed as a regulatory object for several years, but it is only recently with the European Commission’s proposal for the regulation of AI that these discussions have become manifest in concrete (ultimately ‘hard’) regulatory text. Predating (and in several ways inspiring) this regulatory proposal one will find a number of ‘softer’ initiatives mandated by the EU itself as well as certain international governance institutions

Soft law governance in the field of AI  427 such as the Organisation for Economic Co-operation and Development (OECD) and the Institute of Electrical and Electronics Engineers (IEEE), as well as certain national and private sector initiatives. For example, in the years between 2016 and 2019, we saw a stark increase in the number of various ‘ethics guidelines’ related to the use of AI (Jobin, et al. 2019; Fjeld et al. 2020). Examples of such initiatives include the Ethics Guidelines for Trustworthy AI published by the EU High Level Expert Group on AI (AI HLEG 2019), the 23 Asilomar Principles for the development of AI proposed by the Future of Life Institute (Future of Life Institute 2017), a Recommendation of the Council on Artificial Intelligence as well as a set of AI Principles published by the OECD (2019a; 2019b), and a document on Ethically Aligned Design published by the IEEE as part of an initiative to produce no less than fourteen standards related to AI (IEEE 2019). Additionally, some countries are currently developing initiatives such as standards and labelling schemes in the field of AI. For example, in an American context, the National Institute for Standards and Technology (NIST) is working to develop a selection of AI standards (NIST 2022), in Singapore the Personal Data Protection Commission (PDPC) has published a Model Artificial Intelligence Governance Framework (PDPC 2020), and, in Europe, Malta (Malta.AI 2019), Denmark (D-Seal 2022) and Switzerland (Swiss Digital Initiative 2022) have presented or are in the process of developing certification or labelling schemes for digital technologies, AI and related fields (for standards see also Gustafsson and Tamm-Hallström’s as well as Micklitz’s chapters in this volume). Importantly, the regulation of AI is nothing new in the sense that even before the emergence of these relatively recent initiatives, AI technologies and their particular uses were already regulated under existing laws at both European, national, and international levels. Examples include EU primary law (e.g., EU Treaties and its Charter of Fundamental Rights), EU secondary law (e.g., the GDPR, product liability law, anti-discrimination directives, consumer law, etc.), the UN human rights treaties, the Council of Europe Conventions (e.g., the European Convention on Human Rights), and a multitude of national (i.e., member state) laws. Among most of the existing initiatives aimed specifically at the governance of AI, one can identify a number of common aspects. First of all is a tendency to objectify ‘AI’ systems as a deeply complex technology of a rather uncertain ontology. Second is a tendency to describe AI as both a benefit and a potential hazard to human societies (or democracies specifically). Third is a formal aspiration to govern AI in a way such as to simultaneously motivate an increased usage of the technology in virtually all domains of society while at the same time limiting the potential ‘riskiness’ of AI to the extent possible to protect citizen rights – leading, ultimately, to the potential banning of certain AI use cases through hard regulation (as proposed by the European Commission). In this context, soft law, it seems, might come to serve as a way to filter other kinds of ‘risk’ (medium or low) related to AI which are deemed problematic without resulting in prohibitions or other mandatory requirements imposed by law. In the following, we will elaborate on each of these aspects or dimensions of the constitution of AI as an object of regulation. Firstly, the complexity of AI is made visible both in the way that its definition is usually quite broad and unspecific as well as the fact that AI typically gets objectified in the form of certain ‘use cases’ rather than as isolated technologies as such. To exemplify, one may consider the commonly referenced definition by the EU AI HLEG in 2019 (p. 36) which states:

428  Research handbook on soft law Artificial intelligence (AI) refers to systems designed by humans that, given a complex goal, act in the physical or digital world by perceiving their environment, interpreting the collected structured or unstructured data, reasoning on the knowledge derived from this data and deciding the best action(s) to take (according to predefined parameters) to achieve the given goal. AI systems can also be designed to learn to adapt their behavior by analyzing how the environment is affected by their previous actions.

Apart from conceptualizing AI in a highly anthropomorphic manner – what does it mean for a technical system to ‘perceive’, ‘interpret’, ‘reason’, let alone ‘decide’ things? – the issue commonly attributed to such definitions is that they might in principle include within their scope quite simple technologies and algorithms that can be claimed to do all these things without being considered ‘intelligent’. AI thus constitutes a highly moving target in that many systems which were by experts considered to be AI ten years ago would not qualify today as the bar has moved and the complexity and capacities of systems are improved all the time. A more recent definition of AI proposed by the EU Commission attempts to deal with such uncertainty by basing its definition on a list of more specific technologies (machine-learning approaches, logic- and knowledge-based approaches, and various statistical approaches listed in Annex I of the Commission’s regulatory proposal for an Artificial Intelligence Act) adding only that such systems ‘[...] can, for a given set of human-defined objectives, generate outputs such as content, predictions, recommendations, or decisions influencing the environments they interact with [...]’ (European Commission 2021, p. 36). Where such an approach to regulation does accommodate definitional uncertainty to some extent, it necessarily entails a new uncertainty regarding which technologies might be added to or removed from such a list in the future. This further entails an uncertain space of politics and potential lobbying efforts on the part of companies seeking to avoid their products being listed in Annex I. In other words, the remarkable definitional uncertainty surrounding AI – which might itself be central to the preservation of the myth of artificial intelligence – is far from being resolved through existing regulatory initiatives. Secondly, that AI is viewed as both a potential benefit and a hazard to society is reflected in the way that AI tends to be described and evaluated in terms of its ‘riskiness’ towards human beings and society. Systems, in other words, are to be evaluated – and governed accordingly – as either ‘high’ or ‘low’ risk systems depending on the degree of ‘harm’ they might cause humans. Importantly, this ‘riskiness’ of AI is not necessarily viewed as inherent to the technology itself which tends to be described as morally neutral. Rather, the ‘riskiness’ of certain particular AI applications is seen to result from their mobilizations and uses in particular contexts and for particular purposes – i.e., as the object of regulation becomes certain AI ‘use cases’ rather than particular technologies in and of themselves. This approach is exemplified in EU’s proposal for an Artificial Intelligence Act in the way that ‘high risk’ AI is defined as systems used for a variety of specific purposes from biometric identification and management of critical infrastructure to education, employment, access to essential private and public services and benefits, and law enforcement. The reasoning behind associating ‘high risk’ with AI technologies used for these exact purposes and in these institutional contexts stems from the idea that a person’s claim to fundamental rights as well as his or her health and safety is particularly vulnerable in these contexts. Thus, it is not the individual technology itself which is tied to a relative ‘riskiness’ but rather the mobilization of any AI system in particularly public sector contexts and situations.

Soft law governance in the field of AI  429 Thirdly, it is important to note that virtually all existing governance initiatives not only seek to limit but also seek to motivate an increased use of AI in virtually all domains of society. This is due to certain (mainly economic) ‘potentials’ deemed inherent to AI, which in many ways contrast with the potential ‘riskiness’ of the particular use cases described above. The governmental strategy therefore tends to take the form of a dual exercise in simultaneously motivating the uptake of AI among both public and private organizations while at the same time limiting its ‘riskiness’ as much as possible. The approach thus constitutes a kind of simultaneous governmental encouragement and filtering of AI use across societal sectors and institutional contexts. At one end of the scale, one finds proposals of de facto prohibitions of certain AI use cases. That is, the EU’s recent regulatory proposal (European Commission 2021, p. 43) suggests the prohibition of: 1. AI for subliminal techniques to materially distort a person’s behaviour in a manner likely to cause physical or psychological harm; 2. AI aimed at exploiting the vulnerabilities of certain exposed groups; 3. AI used by public authorities to evaluate or classify the trustworthiness of natural persons (social credit scoring) leading to particular forms of detrimental or unfavourable treatment; 4. The use of real time remote biometric identification in public spaces by law enforcement (unless necessitated by certain legitimate objectives). Even if these proposed prohibitions seem relatively specific, it is worth noting the seeming uncertainty around e.g., what it means for a system to ‘materially distort’ a person’s behaviour, what it means to be ‘vulnerable’ or to ‘exploit vulnerability’, what it means to ‘evaluate trustworthiness’, as well as what might constitute ‘legitimate objectives’ for biometric surveillance of public spaces (ibid.). In this context, the relative ‘riskiness’ of AI discussed above becomes the key issue. The regulatory proposal from the EU of course suggests that ‘high risk’ AI should constitute the object of hard regulation, leaving, it would seem, any ‘medium’ and ‘low risk’ systems to be dealt with and governed primarily by various forms of non-legal, soft regulation (such as voluntary standards and certifications, labelling schemes, and guidelines). Besides imposing a number of requirements upon different AI use cases, the regulation also proposes so-called ‘AI sandboxes’ to be mobilized in EU national contexts in order for local governments to encourage the uptake of AI technologies and spur AI-based innovation across all societal sectors. So again, the regulation proposes a filtering mechanism and the governance of AI according to its risks to human rights and welfare, as well as certain mechanisms (e.g., national AI sandboxes) to accelerate the uptake of AI among businesses and public authorities within the EU. Despite its groundbreaking nature, this regulation is still merely a proposal by the EU Commission. Today, the professional field of AI and its usage is populated by a selection of soft law initiatives such as guidelines, recommendations, and so on, which have already in many ways informed the regulatory proposition from the EU. These existing frameworks tend to take the form of principles rather than concrete standards for step-by-step implementation. They also tend to take a more generalist approach to both potentially ‘high’ and ‘low’ risk (i.e., simply ‘risk’ in general) and to address, according to a study from Harvard University’s Berkman Klein Center, a variety of topics from privacy, transparency and explainability, accountability, safety and security, fairness, and non-discrimination, to professional responsibility, human control, and the promotion of human values (Fjeld et al. 2020). Importantly,

430  Research handbook on soft law some of these focus areas seem to be recommended from a rights-based perspective (e.g., safety, privacy, and non-discrimination), whereas others may not be directly rights-related and still be considered ‘best practice’ more broadly (e.g., human control and promotion of values). Again, it is worth noting the significant degree of vagueness adherent to such principles (e.g., notions of ‘human values’).

4.

GOVERNANCE MECHANISMS IN THE MAKING

While political and regulatory developments in the area of AI are largely driven by institutional actors such as the European Commission, a much wider range of stakeholders are involved, including national governments, non-governmental organizations (NGOs), expert groups, and tech companies. Digital technology has for long been approached as an area requiring ‘multistakeholder’ approaches (Flyverbom 2011; Hasselbalch 2021) and has been the subject of extensive discussions, and the instalment of expert groups and policy development processes. Meanwhile, in other ways and other circles, tech companies increasingly engage in a wide range of lobbying efforts, such as engaging in policy deliberations, knowledge production, reporting and visualization of the working and consequences of digital technology (Flyverbom 2017). Specifically, the soft law initiatives outlined above come in a number of different forms and institutional constellations. On the one hand, one finds expert recommendations, principles, and guidelines, that is, a number of documents issued by government institutions, business organizations or expert bodies (including universities) which outline a number of issues and point to a set of very general solutions. On the other hand, one also finds organizations specializing in the development of certifications or labels to other organizations, based on audits of the organizations being assessed. In the field of AI, these certification and labelling schemes (a format we are already familiar with in other areas like ecology, sustainability, fair working conditions, etc.) are still under development, for example in national contexts like Malta and Switzerland. In Denmark, the first label of its kind has been established, dealing with a variety of issues related to cybersecurity and responsible use of data, where AI is treated as one out of eight overall criteria. From a governmentality perspective, such initiatives are interesting because they actively seek to govern organizational uses of AI, not through a state-mandated juridical-legal approach, but from the idea that the certification and labelling of responsibility, and the visibility of such labels in a given market, can function as a driver for both responsible behaviour and competitive advantages amongst businesses. In the following sections, based on our governmentality approach, we elaborate on the concrete shape and underlying rationalities of this overall and emerging governmental regime, which seems likely to include both hard and soft law initiatives. First, we will explore further the general regime of risk management emerging in the field of AI. Second, we will consider some of the more general, principle-based guidelines which have gained traction in and served to frame the governmental field of AI. Finally, we briefly consider some of the more concrete certification and labelling schemes looking specifically at the projects undertaken in Malta and Denmark.

Soft law governance in the field of AI  431 4.1

AI Risk Management as Governance Mechanism

Virtually all existing approaches to the regulation of AI – whether through legal or non-legal measures – rely on the perspective outlined above of AI as an always potentially risky phenomenon. The governmental approach outlined by the AI HLEG (and described above) of ‘[…] seeking to maximize the benefits of AI systems while at the same time preventing and minimizing their risks’ (AI HLEG 2019, p. 4) is effectively echoed by most if not all other initiatives dealing with AI. Thus, most existing guidelines on AI tend to speak the language of ‘risk’ and the more concrete initiatives such as the labelling and certification schemes we will consider below all rely on a risk assessment and risk management approach to AI governance. The general approach to risk management is similar to that described in the regulatory proposal on ‘high risk AI’ put forth by the European Commission insofar as distinctions are drawn between ‘high’, ‘medium’, and ‘low’ risk use cases with more substantial governance requirements imposed on ‘high risk’ applications. The proposal defines ‘high risk’ based on the idea that people’s health and safety, as well as their claim to fundamental rights, is particularly at stake in certain contexts where AI might be mobilized (primarily in the public sector and regarding its ‘essential services’). This does not, however, entail the prohibition of AI use in such contexts; rather the ‘riskiness’ of AI use should in these situations be reduced through a number of concrete initiatives such as risk and quality management systems, new forms of data governance, documentation and user transparency, conformity assessments, corrective actions, and human oversight in general. The proposed law would further impose certain transparency requirements upon AI systems intended to interact with natural persons (such as chatbots) so that people are informed that they are interacting with a technology rather than a human being (European Commission 2021). Also, whereas the regulatory proposal tends to frame AI risk as a question of potential impact to fundamental rights, the more generalist guidelines and forms of certification emerging around AI often speak of risk in a broader sense. For example, the preliminary material on the Maltese approach to AI certification speaks of the risks of ‘direct and indirect effects on different stakeholders’ (Malta.AI 2019, p. 23) including a number of potential ‘risks’ which have hitherto remained unaddressed by regulation. One example is the inclusion of very general risks such as ‘different types of harm or damage’ and broad-spectrum risks such as risks ‘to the environment and to animals’ (ibid.). Hereby, what may count as AI-related risks clearly extend well beyond what regulators have until now sought to take into account. Thus, while risk management appears as a key to most if not all existing approaches to AI governance, we can identify a continuum between relatively narrow approaches focusing, for example, on fundamental rights specifically versus inclusive forms of risk management which, in principle, seeks to address any conceivable risk of societally undesirable implications of AI usage. Such differences in perspective on what may count as ‘risk’ naturally involve different ideas of the ways in which risk mitigation is conducted. Again here, we may locate a continuum between forms of risk and impact assessment which focus more or less exclusively on rights and legal entitlements to the kinds of more general stakeholder involvement proposed by initiatives such as the Maltese and Danish certification initiatives. Rather than focusing exclusively on rights, the approach to risk assessment appears as much more general. For example, the official language of these initiatives proposes a form of risk management which mitigates and avoids ‘societally unacceptable results’ (Malta.AI 2019, p. 23) or ensures that

432  Research handbook on soft law particular AI applications ‘can be accepted by those they affect’ (D-Seal, 2022). Both of these approaches to risk mitigation are entirely general in scope and imply the inclusion, in principle, of any conceivable risk if deemed sufficiently problematic. Of course, the very generalizing language also entails the potential issue that fundamental rights may never be taken into account specifically, whereby the approach to risk management becomes so general that it might lead companies to neglect the specificity and particularity of some risk types relative to others. In other words, the conceptualization of AI in terms of risk has consequences for both the forms of political intervention and the concrete mechanisms being developed. We will consider various limitations of this paradigm in more detail in the discussion below. 4.2

Principles and Guidelines as Governance Mechanisms

The number and variety of principle-based documents and guidelines on AI is becoming increasingly extensive. As described briefly above, examples include the Ethics Guidelines for Trustworthy AI published by the EU High Level Expert Group on AI, the 23 Asilomar Principles for the development of AI proposed by the Future of Life Institute, the Recommendation of the Council on Artificial Intelligence and a set of AI Principles both published by the OECD, and a document on Ethically Aligned Design published by the IEEE. Also, companies such as Google and IBM have adopted principles for the responsible use and deployment of AI-based solutions in recent years (Google 2022; IBM 2022). While such documents are informative as to the general debate around AI and the overall issues emerging from the politics concerning this technology, the language usually remains quite vague and general with regard to the ways in which issues might be addressed in a concrete fashion. For example, guidelines on privacy tend to point to the importance of gaining ‘informed consent’ to AI-driven data processing but rarely (if ever) suggest concrete measures for doing so. Similar points could be made about ideals and principles such as fairness, non-discrimination, and the promotion of human values which all constitute important areas, but where best practice solutions remain scarce and often difficult to implement in practice. Thus, while the emergence of such guidance documents is arguably important to the development of a more concrete governance regime in the field of AI, the degree to which many of these principles or ideals are put into practice is currently very low. By turning the governance of AI into a matter of developing principles and guidelines, it becomes rather abstract and distanced from concrete settings and interventions. As we shall see, however, a few certification and labelling schemes are emerging which in various ways seek to translate abstract principles formulated by such guidance documents into relatively concrete measures for organizations to undertake and potentially become certified. 4.3

Labelling and Certification Programs as Governance Mechanisms

At least two AI certification schemes have been proposed within the EU in the context of particular member states, namely Malta and Denmark. In 2019, the Maltese authorities issued a document entitled ‘Strategy and Vision for Artificial Intelligence in Malta 2030’ centring around the establishment of a certifying body for ‘trustworthy AI’ (Malta.AI 2019). At approximately the same time, the Danish authorities announced the establishment of a privately funded labelling scheme for ‘IT security and responsible use of data’ which came to

Soft law governance in the field of AI  433 include criteria for ‘trustworthy algorithms and AI’ (D-Seal 2022). While the current stage of development of the Maltese certification body is uncertain, the Danish label was launched in September 2021. As mentioned above, both initiatives appear to require companies wishing to become certified to establish a formal governance regime and conduct risk assessments, certain forms of stakeholder involvement to determine risk, to mitigate against errors and potential bias by ensuring the quality of data and mathematical models, as well as to reorganize the development of new technologies to ensure both responsibility and auditability. With such governance mechanisms as labelling and certification schemes, it is important to consider the differences in governmentality between legalist, regulatory approaches on the one hand and these, in a sense, ‘non-’ or ‘extra-legal’ initiatives on the other. For example, in the Danish context, the certifying body is a private organization rather than a governmental institution which entails important implications for the overall mode of governance. In this case, the certifying organization is itself first and foremost an economic actor seeking to become economically sustainable or even profitable by selling certifications to other organizations. These organizations, in turn, are likely to only invest in certification if the perceived (economic) benefits of becoming certified outweigh the costs of implementation, and this decision to become certified or not remains entirely in the individual organization’s discretion. This form of governance-by-market dynamics rather than through the imposition by the government of legal obligations is exactly what we tend to characterize as neoliberal governance. More exactly, the success of such certification and labelling initiatives relies on the degree to which the economic self-interest of organizations by itself manages to become a driver for widespread implementation. On the one hand, if the implementation of criteria for the ‘trustworthiness’ of AI applications becomes perceived as too costly and difficult by organizations, the initiative risks being unsuccessful. On the other hand, however, if implementation becomes too easy and the relevant criteria too watered down, these initiatives equally risk failing, as public scepticism regarding the value of organizational certification causes the potential marketing value perceived by organizations to decrease. Thus, initiatives such as these lead to a difficult situation where a regulator has to find the right balance between providing concrete, implementation-ready criteria, while at the same time ensuring that implementation also does not become too easy as, for example, a merely formal exercise of ticking a few boxes to earn the certificate. In the cases of both the Maltese and Danish initiatives, their relative uptake by organizations remains to be seen, yet it is worthwhile to note the distinct governmentality of such projects in comparison with potentially ‘hard’ regulation. By turning the governance of AI into a matter of often commercial, voluntary labelling and certification based on self-reporting and documentation, these interventions entail a logic or political rationality that may not always align with those of democratic institutions and more binding forms of governance.

5. DISCUSSION Our overview of the emergence of AI as an object of governance and the concrete mechanisms being developed gives insight into the making of a relatively new domain of governance that is still in the making. As we have shown, the strong focus on risk assessments paves the way for a particular configuration of ‘hard’ and ‘soft’ forms of governance. At the one end, you have an acknowledgement from the EU that some measure of hard regulation (including

434  Research handbook on soft law certain prohibitions) is needed to tame what is seen to be the riskiest AI use cases to human welfare. This does, however, leave a wealth of potentially ‘medium’ and ‘low risk’ use cases to go unregulated or regulated solely through voluntary (‘soft’) initiatives, whereby societal responsibility becomes a matter of organizational discretion rather than legal obligation. As such, the emerging governmental field is characterized by elements of a traditional regulatory state as well as characteristics of a relatively neoliberal order. As we know from the writings of Foucault, neoliberal governance often entails – and does indeed in this case entail – posing seemingly paradoxical questions such as: How much risk to human wellbeing are we happy to tolerate in order to ensure innovation? And yet, how do we regulate that which we insist on not regulating? We certainly find such logics instantiated in the field of AI, in the spectrum between ‘harder’ and ‘softer’ forms of governance we have described in this chapter. Such dynamics are further reflected in the way that compliance with various soft law initiatives becomes a question of whether market forces may function to drive responsibility, or whether the minimization of risk associated with medium or low risk systems become economically unfeasible – and thus may go virtually unregulated. For regulators, this seems to amount to a question of either regulating AI intensely and thereby forgoing significant economic potentials from its unlimited innovation versus regulating risk primarily through voluntary initiatives, thus leaving the door open for innovation and economic benefits. One way of resolving this dilemma seems to be an attempt to turn ‘ethical AI’ into a means of competitive advantage; that is, allowing businesses to differentiate themselves as ‘responsible’ actors through various forms of certification and labelling schemes. Indeed, we see the governments from Malta, Denmark and Switzerland developing or assisting in the development of certification or labelling of ‘ethical AI’, allowing businesses to market themselves as morally superior providers of AI solutions. The hope is that responsibility can be mobilized by market actors as a certain kind of commodity and thus drive rather than hinder the creation of economic value. As we argue, such mechanisms provide fine examples of what we commonly understand as neoliberal governance, that is, the governance of conduct through the exploitation of market dynamics and economic value as its primary instruments. It is the exact same driver we associate with non-mandatory forms of corporate social responsibility (CSR), corporate sustainability, and other such concepts taken up by companies themselves to reshape their marketing and modes of production in the hope of differentiating themselves positively and gaining competitive advantages in a given market by acting ‘responsibly’. While such a mode of governance may indeed lead to desirable changes in corporate conduct, it is also worth noting how the effectiveness of this neoliberal mode of regulation is inherently limited in certain ways. For example, in principle, if all businesses managed to live up to all relevant criteria for responsible conduct and achieved a certificate or a label, the marketing value of differentiating oneself as responsible is effectively annulled. Thus, with such a governmental strategy, the goal is not and cannot be ‘global compliance’ across the entire population of businesses. Further, the effectiveness of the governmental strategy is further problematized by the principle that the economic benefits achieved through subsequent marketing of being ‘responsible’ usually must outweigh the costs associated with achieving the certification or label in the first place (costs of developing compliance, of engaging in various auditing activities, etc.). In other words, being ‘responsible’ must – often or at least in principle – be profitable. This principle is likely to entail the exclusion of certain criteria (e.g., those that are particularly costly for organizations to comply with) and prioritization of other (e.g., low cost) criteria. It also means that the auditing activities involved in controlling

Soft law governance in the field of AI  435 whether organizations live up to the relevant criteria before achieving the certification or label might come to be scaled down or reduced so as to make control activities less costly. Whether such developments will indeed come to characterize the certification or labelling of ‘ethical AI’, of course, in practice remains to be seen. It might also be the case that such initiatives indeed manage to prompt new levels or forms of responsibility within certain (e.g., national) populations of organizations. However, the elevation of responsibility across the entire population through certification or labelling is, as noted above, very unlikely in the short term. Limiting hard regulation only to a select list of ‘high risk’ AI use cases and leaving other use cases to market dynamics seems to constitute a way for regulators today to aim for a middle way between regulating AI and ensuring its continuous innovation and diffusion across the economy. Such a strategy, however, begs the question whether innovation will continue to happen at the expense of rights? If ‘medium’ or ‘high risk’ indeed implies a risk to fundamental rights, what are the implications when governments focus exclusively on ‘high risk’? An important question in this regard today seems to be: which phenomena tied to AI development are actually considered as ‘risks’ to human wellbeing and which are not? On the one hand, we have seen how the list of issues associated with AI usage in organizational settings already spans widely from transparency to responsibility, bias and discrimination, surveillance, and so on. On the other hand, a number of other issues are usually not taken up by regulatory initiatives whether hard or soft in character. Perhaps unsurprisingly, several of these unarticulated issues are those that neoliberal policymaking has particular difficulties dealing with – or which, to neoliberalism, do not constitute problems at all but could be taken as inescapable consequences of capitalism or even a sign of a well-functioning economy. That is, issues such as job uncertainty, constant competition, economic inequality, product addiction, and overconsumption are rarely – if ever – addressed by either existing hard or soft law initiatives; perhaps because these topics often constitute the flip side of the positive potentials commonly associated with AI usage (e.g., automation, profitability, personalization, consumer loyalty, enhanced marketing, and so on). In other words, such potential issues associated with AI which conflict directly with the most common aspirations associated with AI usage, it would seem, tend to escape regulatory initiatives, since regulating these areas might constitute a direct attack on techno-capitalist ideology. Or so one might speculate. One noticeable attempt to regulate the transition from ‘living’ to ‘dead labour’ (i.e., human versus machine labour) can be observed as some countries begin to experiment with universal basic income (UBI) programs. However, the questions as to the degree by which a transition is taking or will take place from a primarily human to an increasingly machinic and automated ‘workforce’, and how regulators will deal with the associated questions of especially unemployment and inequality, currently stand largely unanswered. Such questions quickly turn into fundamental questions about the future of capitalist societies and as such may require debates and discussions as serious as the issues that prompted the debate in the first place.

6.

CONCLUDING REMARKS

Our chapter has unfolded two dimensions of artificial intelligence and soft law, namely how this issue area has come into existence and how it is currently being institutionalized in the European context through concrete governance mechanisms. This dual focus contributes to

436  Research handbook on soft law emergent scholarship on AI and soft law by situating the policy domain in relation to other discussions about the regulation of digital technology and in relation to broader discussions of the role of soft law in democratic societies. In particular, our approach highlights the amorphous and emergent character of contemporary efforts to develop regulation in relation to AI and shows how particular ways of organizing and conducting governance becomes part and parcel of the institutionalization of particular logics and rationalities of governance. By approaching AI in terms of risks and potentials, and with a focus on use cases rather than concrete technologies, the governance of this domain takes a very fuzzy and amorphous character at present, which leaves a lot of room for uncertainty and interpretation by the commercial and governmental actors involved. As we have argued, the governance of AI is a regulatory domain marked by uneasy and potentially misaligned configurations of interventions and rationalities propelled by democratic institutions and commercially driven efforts based on voluntary self-regulation. In the years to come, the governance of AI should be at the top of minds for not only policymakers but also scholars interested in contemporary transformations in the making and consequences of soft law.

REFERENCES Abbott, K. W. and Snidal, D., (2000). Hard and soft law in international governance. International organization. 54(3), 421–456. AI HLEG., (2019). Ethics Guidelines for Trustworthy AI. High Level Expert Group on Artificial Intelligence [online]. European Commission. [Viewed 25 August 2022]. Available from: https://​ec​ .europa​.eu/​futurium/​en/​ai​-alliance​-consultation​.1​.html. Alaimo, C., and Kallinikos, J., (2021). Managing by Data: Algorithmic Categories and Organizing. Organization Studies, 42(9), 1385–1407. [Viewed 12 July 2023]. Available from https://​doi​.org/​10​ .1177/​0170840620934062. Bowrey, K., (2005). Law and Internet Cultures. Cambridge: Cambridge University Press. D-Seal., (2022). Denmark’s new labelling program for IT security and responsible use of data [online]. D-Seal. [Viewed 25 August 2022]. Available from: https://​d​-seal​.eu/​. DeNardis, L., (2009). Protocol Politics: The Globalization of Internet Governance, MIT Press. Djelic, M.-L., and Sahlin-Andersson, K., (2006). Transnational Governance: Institutional Dynamics of Regulation. Cambridge, UK; New York, Cambridge University Press. European Commission., (2020). White Paper on Artificial Intelligence – A European approach to excellence and trust. COM(2020) 65 final. European Commission., (2021). Proposal for a Regulation of the European Parliament and the Council Laying Down Harmonised Rules on Artificial Intelligence (Artificial Intelligence Act) and Amending Certain Union Legislative Acts. COM(2021) 206 final. Fjeld, J., Achten, N., Hilligoss, H., Nagy, A., and Srikumar, M., (2020). Principled artificial intelligence: Mapping consensus in ethical and rights-based approaches to principles for AI. Berkman Klein Center Research Publication. No. 2020-1. [Viewed 12 July 2023]. Available from: https://​dx​.doi​.org/​10​ .2139/​ssrn​.3518482. Flyverbom, M., (2011). The Power of Networks: Organizing the Global Politics of the Internet. Cheltenham: Edward Elgar. Flyverbom, M., (2017). Corporate advocacy in the Internet domain: shaping policy through data visualizations. In: C. Garsten and A. Sörbom, eds. Power, Policy and Profit: Corporate Engagement in Politics and Governance. Cheltenham: Edward Elgar. pp. 115–133. Foucault, M., (1982). The Subject and Power. Critical Inquiry. 8(4), 777–795. Foucault, M., (1991). Governmentality. In: G. Burchell, C. Gordon and P. Miller, eds. The Foucault Effect: Studies in Governmental Rationality. London: Harvester Wheatsheaf. pp. 87–105. Foucault, M., (1994). The Order of Things: An Archaeology of the Human Sciences. London: Vintage.

Soft law governance in the field of AI  437 Future of Life Institute., (2017). Asilomar AI Principles [online]. Future of Life Institute. [Viewed 25 August 2022]. Available from: https://​futureoflife​.org/​2017/​08/​11/​ai​-principles/​. Gillespie, T., (2017). Governance of and by platforms. In: J. Burgess, T. Poell and A. Marwick, eds. SAGE Handbook of Social Media. London: SAGE. pp. 254–278. Google., (2022). Artificial Intelligence at Google: Our Principles [online]. Google. [Viewed 25 August 2022]. Available from: https://​ai​.google/​principles/​. Hasselbalch, G., (2021). Data Ethics of Power – A Human Approach in the Big Data and AI Era, Cheltenham: Edward Elgar. Hofmann, J., Katzenbach, C. and Gollatz, K., (2016). Between coordination and regulation: Finding the governance in Internet governance. New Media & Society. 19(9), 1406–1423. IBM., (2022). Everyday Ethics for AI [online]. IBM. [Viewed 25 August 2022]. Available from: https://​ www​.ibm​.com/​design/​ai/​ethics/​everyday​-ethics/​. IEEE., (2019). Ethically Aligned Design. A Vision for Prioritizing Human Well-being with autonomous and Intelligent Systems [online]. Institute of Electrical and Electronics Engineers. [Viewed 25 August 2022]. Available from: https://​ethicsinaction​.ieee​.org/​wp​-content/​uploads/​ead1e​.pdf. Jobin, A., Ienca, M., and Vayena, E., (2019). The Global Landscape of AI Ethics Guidelines. Nature Machine Intelligence. 1(9), 389–399. Malta.AI., (2019). A strategy and vision for artificial intelligence in Malta 2030 [online]. Malta.AI. [Viewed 25 August 2022]. Available from: https://​malta​.ai/​wp​-content/​uploads/​2019/​11/​Malta​_The​ _Ultimate​_AI​_Launchpad​_vFinal​.pdf. Marchant, G., (2019). ‘Soft Law’ Governance of Artificial Intelligence, in AI Pulse. [Viewed 13 July 2023]. Available from https://​escholarship​.org/​content/​qt0jq252ks/​qt0jq252ks​_noSplash​_1f​ f6445b4d4e​fd438fd6e0​6cc2df4775​.pdf​?t​=​po1uh8. Marchant, G., Tourna, L., and Gutierrez, C. I., (2020). Governing emerging technologies through soft law: Lessons for artificial intelligence—An introduction. Jurimetrics. 61(1), 1–18. Miller, P. and Rose, N., (1990). Governing economic life. Economy and Society. 19(1), 1–31. Mueller, M., (2010). Networks and States: The Global Politics of Internet Governance. Cambridge, MA: MIT Press. Nadler, J., (2020). Congressional Report: Investigation of Competition in Digital Markets [online]. US House of Representatives. [Viewed 25 August 2022]. Available from: https://​judiciary​.house​.gov/​ uploadedfiles/​competition​_in​_digital​_markets​.pdf. NIST., (2022). Artificial Intelligence [online]. NIST [Viewed 25 August 2022]. Available from: https://​ www​.nist​.gov/​artificial​-intelligence. OECD., (2019a). Recommendation of the Council on Artificial Intelligence. OECD/LEGAL/0449. Available from: https://​legalinstruments​.oecd​.org/​en/​instruments/​OECD​-LEGAL​-0449. OECD., (2019b). OECD AI Principles [online]. Organization for Economic Co-operation and Development. [Viewed 25 August 2022]. Available from: https://​oecd​.ai/​en/​ai​-principles. O’Neil, C., (2018). Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy. New York: Crown. PDPC., (2020). Model Artificial Intelligence Governance Framework. Second Edition [online]. Personal Data Protection Commission Singapore. [Viewed 25 August 2022]. Available from: https://​www​.pdpc​ .gov​.sg/​-/​media/​files/​pdpc/​pdf​-files/​resource​-for​-organisation/​ai/​sg​modelaigov​framework2​.ashx. Smith, B. C., (2019). The Promise of Artificial Intelligence. Cambridge, Massachusetts: MIT Press. Swiss Digital Initiative., (2022). Swiss Digital Initiative: Advancing Practice-Oriented Ethics In The Digital Age [online]. Swiss Digital Initiative. [Viewed 25 August 2022]. Available from: https://​www​ .swiss​-digital​-initiative​.org/​.

Index

Abbott, K. 11, 43, 45, 77, 133, 138, 184, 209, 210, 305, 320, 322, 435 abuse of powers, courts and rule of law, EU 193, 203 accountability business and social norms in transnational governance 138–9 democracy 9, 91, 92, 93–4, 96 economic governance, EU 212–14, 218 international standards and the dilution of responsibility 185–6 US administrative guidance 243–4 Acuto, M. 337, 338, 340, 345, 346 administrative discretion, courts and rule of law, EU 194–8, 203 administrative guidance, US see US administrative guidance African Court on Human and Peoples’ Rights (ACtHPR), Anudo v United Republic of Tanzania 355–6 agencies decentralized see decentralized agencies, EU economic governance, EU 208, 211–13, 214, 217, 218 US administrative guidance 238–9, 240, 241, 242–3 AI governance, EU 423–37 digitalization and datafication processes 426 economic potentials 429, 433, 434–5 employment and universal basic income (UBI) programs 435 ethics guidelines 427 future direction 430–33 future research 435–6 governance structure 425 High Level Expert Group (HLEG) 427–8, 431, 432 human rights issues 429–30, 431, 432, 435 international governance institutions 426–7 labelling and certification schemes 427, 430, 431, 432–3, 435 local governments and AI sandboxes 429 national and private sector initiatives 427 neoliberal governance 433, 434–5 principle-based documents and guidelines 432 privacy issues 426, 429, 432 regulation, AI as object of 424, 426–30 regulatory prohibitions 429

risk, regulatory approach 424, 427, 428, 429–30, 431–2, 435 stakeholder involvement 430, 431 steering 425 system objectification issues 427–8 tech companies and lobbying 430 tech giants, rise of 426 technical standards and European private law 147, 155–6 transparency requirements 431, 435 see also technology Alemanno, A. 261, 263, 264 Allio, L. 119, 125 Andone, C. 256, 372, 383 annulment cases 109–10, 197, 198, 213, 215, 217, 218 Anthony, R. 237, 238, 241 anthropology of soft law 71–86 customary village law and international law, similarities between 73–4 and denationalization process 75–6 enforceability issues 73, 75, 79, 80, 82 future direction 78–83 and global legal environment 71, 72, 75–6, 77, 80, 81 human rights legal order 81–2 and legal pluralism 74–5, 76, 81 neoliberal principles 71, 72, 75, 77 racial discrimination and private sector action 80–81 self-regulation 81 social ordering in small-scale societies 73 soft law concept 71–3 soft law, legal reception and opinions 76–8 soft law as non-law 77–8 and state sovereignty 76 supranational capitalism 75 transnational legal processes 74–6, 81 Armstrong, K. 13, 18, 19, 107, 202, 225 Arnull, A. 110, 196, 197 Arthurs, H. 14, 19 artificial intelligence see AI governance, EU Ashiagbor, D. 47, 106 Ausfelder, A. 5, 304–18 authorship criterion, courts and rule of law, EU 198 autocracies, rise of, rule of law crisis 393 backsliding, rule of law crisis 261, 391, 392, 393

438

Index  439 Bair, J. 133, 138 Baldwin, R. 87, 88, 92, 93, 263 Barron, D. 338, 340 Bartley, T. 133, 138 Baumgärtel, M. 337, 340 Bekker, S. 50, 128, 290, 294, 308, 396, 397, 398 Benda-Beckmann, F. 75, 76, 83 Bergkamp, L. 227, 229 Berman, P. 14, 16, 19, 71, 75 Bernstein, L. 56, 59, 68, 69 better regulation agenda, EU 116–31 and Business Environment Simplification Task Force (BEST) 119 consultation and IA requirements, hard law versus soft law 122–3 criticisms 117 domestic level 122–6 future research 128 history 119–22 impact assessment (IA) 119, 121, 122–3, 124–6, 127–8 institutional confrontation 120–21 Mandelkern Report 119 OECD guidance 119, 124–5 and open method of coordination 120 policy process and politics 118, 120, 121, 122 political priorities 122, 128 regulatory burdens 116 regulatory indicators 120 regulatory offsetting mechanism 122 Regulatory Oversight Body (ROB) 125–6 Regulatory Scrutiny Board (RSB) 121 stakeholder consultation 123–4 Beveridge, F. 43, 305, 372 Bexell, M. 88, 92, 93, 94, 183 bindingness citizenship regimes 354 and civil society 274 and climate change law 321, 322, 326–7, 332 decentralized agencies 226–7, 230, 233 domestic factors and soft law cycle 305, 306, 307, 310, 313 public law, 101, 102–3, 105, 106, 107, 108–9, 110, 111–12 technical standards and European private law 146–7 US administrative guidance 240, 241, 242, 244, 246 Blauberger, M. 47, 262 Bodansky, D. 324, 325, 326, 327, 328 Bodle, R. 325, 326–7, 329 Böök, B. 6, 368–90 Borchardt, G. 10, 43, 101, 165, 201 Börzel, T. 88, 306

Bourdieu, P. 11, 14, 80 Bovens, M. 184, 208 Boyle, A. 275, 320, 322, 354 Braithwaite, J. 13, 15, 18 Brown, E. 247, 279 Brummer, J. 134, 141, 173 Brunnée, J. 273, 322–3, 324, 325, 329 Brunsson, N. 11, 138, 139, 177, 178, 179, 182, 183, 184, 185, 187, 346 Business Environment Simplification Task Force (BEST), EU 119 business and social norms in transnational governance 132–44 brand loyalty effects 139 civil society and corporation interaction 138–40 economic globalization effects 132 enforceability 136, 137, 138 NGOs 132, 132–3, 136, 137, 138, 139, 140 organizational cultures and societal expectations 134 public accountability 138–9 pursuit of profit 133, 134, 135 business and social norms in transnational governance, corporate responsibility 133–8 due diligence duties 136–7 and enlightened self-interest 134–5 ethical considerations, priority of 135 financial considerations 135–6 group interests and stakeholder engagement 135–6 and hard law 136, 137 legitimacy of decisions 137–8 responsibility attribution 136–8 sanctions 136 and supply chains 132, 136–7, 138 voluntary self-regulation 135, 136 Busuioc, M. 211, 212, 213, 218, 224, 412, 414 Büthe, T. 10, 12, 87, 138, 149, 162 Cancun Agreements, climate change law 324, 326 Cappellina, B. 253, 305, 307, 308 Capra, F. 14, 19 Carmichael, S. 134, 141 Cassese, S. 19, 75 Chalmers, D. 17, 18, 19 Chamon, M. 110, 213, 224, 226, 227, 261 Chayes, A. 63, 69 children, citizenship regimes 355, 356, 359, 360, 361 China, Belt and Road Initiative (BRI) 12, 16 Chinkin, C. 45, 47, 166, 191, 203, 320, 321–2, 363 Chiti, E. 224, 225, 226 Cini, M. 44, 254

440  Research handbook on soft law citizenship regimes 353–67 bindingness 354 children 355, 356, 359, 360, 361 citizenship acquisition 354 and democracy 90, 92, 93, 94 EU investor citizenship 358, 362–3 EU Open Method of Coordination 358 future research 361, 363 Global Citizenship Compact consideration 362 Global Citizenship Observatory’s (GLOBALCIT) Citizenship Law Dataset 360, 361 governance indicators 359–62 immigrant integration and security 360–61 indigenous peoples 358–9, 360, 361 and international human rights norms 355–6, 357, 358–9, 361 international law-making 354–7 Migrant Integration Policy Index (MIPEX) 360 multi-level governance 358, 359–60 multicultural citizenship 354, 356–7, 358–9, 360 and national sovereignty 353 nationality right 354, 355–6 and non-discrimination 356 refugee rights 361 statelessness 353, 355, 356, 359, 360, 361 city regulation and governance 337–52 agency and participation 340 cities as global-local entities 338–41 and COVID-19 pandemic 337 economic power 340 enforceability 341, 344 future research 339, 346, 347 global dimension 339–41 Human Rights Cities 340 hybridization of governance 346, 347 international charters for local self-government 340 new stakeholders 346–7 sanctuary cities 340 technological innovation challenges 338, 339, 340–41 transnational city networks (TCNs) 337–8, 339, 340, 345, 346 city regulation and governance, smart cities 338, 340–46 and big data analytics 342 Cities Coalition for Digital Rights Declaration (CC4DR) 342, 343 definition 341 digital rights 342–3, 344 and economic marketing 341–2

ethical governance of algorithms 342 EU Digital Single Market Strategy 344–5 Global Observatory on AI 342 Global Smart Cities Alliance on Technology Governance 345–6 hard law enforcement issues 344 home-sharing and Airbnb 344 human rights in digital environments 343 privacy issues 338, 342–3, 347 sharing (collaborative) economy 343–5 standards and performance indicators 346–7 sustainability and environmental policies 341, 345–6 technology reliance 341–2 United for Smart Sustainable Cities (U4SSC) 346 civic and rights-based citizenship, and democracy 90, 92, 93, 94 civil society 272–87 anti-whaling discourse 278 and bindingness 274 business and social norms in transnational governance 138–40 community of practice principle 273–6 community of practice principle, problems and cognitive framing 275 compliance with soft law 274–5 environmental law 276–7, 278 and EU Open Method of Coordination (OMC) 297–8 global anti-corruption regime 278–80 global social problems, response to 278–80 hard law and soft law, relationship between 275, 280–83 human rights 280–83 international arbitration 277–8 and international lawmaking process 275–6 legal obligation 273–4 making appeals to normativity 276–8 NGOs 276, 278, 279, 283 ‘norm entrepreneurs’ framework 272 soft law as tool in strategic litigation 282–3 climate change law 320–36 and bindingness 321, 322, 326–7, 332 Cancun Agreements 324, 326 and collaboration 323, 326 common but differentiated responsibilities and respective capabilities (CBDRRC) 323–4, 329 developing countries 323, 324, 325 greenhouse gas emissions 324, 329 hard and soft law, distinction between 322–3, 325, 326 Kyoto Protocol 63, 320, 324, 325, 326, 327, 328, 329, 330, 332

Index  441 legal obligation 326–7, 330 and legal validity 322 new legally-binding climate treaty debates 323–35 Paris Agreement 63, 320, 325, 326–7, 328–32 soft law instruments 321–2 UN Climate Change Conference and Copenhagen Accord 324, 326, 328 see also environmental law; UN Framework Convention on Climate Change (UNFCCC) climate change law, climate change mitigation issues 324, 325, 327–32 commitments and obligations 330–32 flexibility 328 global average temperature increase 329 Global Stocktake 329 Montreal Protocol 327–8 Nationally Determined Contributions (NDCs) 329–30 provision comparisons 330–32 top-down versus bottom-up approach 327–8 climate and energy policy, EU Open Method of Coordination (OMC) 290–91, 292, 294, 296–8 co-regulation, and economic analysis of soft law as regulatory tool 61 Coase Theorem 58, 59 codifications, history of soft law 31–3, 34, 35, 36, 37–8 collaboration and cooperation city regulation and governance, smart cities 343–5 climate change law 323, 326 European Commission and soft law making 253, 254, 258–9, 261, 262, 265 global governance 166, 170 sincere cooperation principle 48, 111, 195, 258, 265 Colombian Constitutional Court 280–83 Coman-Kund, F. 224, 256 common but differentiated responsibilities and respective capabilities (CBDRRC), climate change law 323–4, 329 community of practice principle, civil society 273–6 competition law 18, 64, 108, 110, 194–5, 196, 254, 261, 264–5 compliance assessment civil society 274–5 gender equality, EU 371, 372 international standards and dilution of responsibility 179–82 rule of law crisis 392, 393, 398

‘comply-or-explain’ approach 111, 408, 417–18 conferral principle, public law 12, 102–3, 110 contract law, technical standards and European private law 148–51, 156 Convention on Biological Diversity 277 Convention on the Elimination of All Forms of Discrimination against Women (CEDAW) 356 cooperation see collaboration and cooperation Cooter, R. 58, 68 Copeland, P. 293, 294, 296 copyright protection, technical standards and European private law 153 corporate responsibility see business and social norms in transnational governance, corporate responsibility corruption, global anti-corruption regime 278–80 Cosma, H. 47, 254 cosmopolitan democracy 92, 94 Cotterrell, R. 14, 19 country-specific recommendations (CSRs) 107, 290, 294, 295, 397–8, 399 courts and rule of law, EU 191–207 access to justice and right to fair trial 194, 203 annulment cases 197, 198 authorship criterion 198 Commission guidelines 195, 201 competition and State aid 194–5, 196 criticisms 192 enforceability 195, 202, 203 environmental protection 199 equality before the law and non-discrimination 194, 203 hard and soft law, distinction between 200–201 hybridity of legislative framework and judicial engagement 199–200 interpretative support role 200 judicial review and administrative discretion control 194–8, 203 legal certainty 193, 194, 195, 197–8, 201, 203 legality principle 193, 194 legitimate expectations principle 195, 201 national authorities and courts 195–7, 198, 199, 200 new governance context 202, 204 prevention of abuse/misuse of powers 193, 203 sincere cooperation principle 195 COVID-19 pandemic and city regulation and governance 337 EU Open Method of Coordination (OMC) 291, 295

442  Research handbook on soft law European Commission and soft law making 254, 259–60, 261–3, 264 gender equality, EU 368 and global governance 168 post-Covid effects, governance through soft law 19–20 and public law 106, 108 US administrative guidance 237, 245, 248 Craig, E. 80, 82 Craig, P. 13, 18 Crane, A. 134, 141 Curtin, D. 214, 255 Cutler, A. 16, 92, 93 Cyert, R. 139, 141 Czempiel, E.-O. 12, 138 Dahl, R. 87, 92 Daly, M. 293, 294, 296 D’Aspremont, J. 43, 101, 305, 320, 321, 322, 340 Davies, G. 11, 148 Davis, K. 11, 173, 359 Dawson, M. 193, 208, 216, 217, 258, 288, 305, 397 De Búrca, G. 13, 231, 358 De Filippi, P. 19, 347 De Francesco, F. 120, 125 De la Porte, C. 289, 291, 294, 296, 297 De Larosière Report, financial crisis effects, EU 409 De Witte, B. 3, 17, 101–15, 197, 200, 288 decentralized agencies, EU 223–36 Agency for the Cooperation of Energy Regulators (ACER) 226, 232, 233 bindingness 226–7, 230, 233 ‘Common Approach’ 224 EU Intellectual Property Office (EUIPO) 226 EU Maritime Safety Agency (EMSA) 226 European Medicines Agency (EMA) 226–7 European Union Aviation Safety Agency (EASA) 226–7 future research 233 gatekeeping exercises 227 guidance-making 223, 225–6 influence of 224 institutional and judicial frameworks 225 judicial review 227 and national regulatory authorities (NRAs) 226–7 and private sector 227 proceduralization 226 soft law rule-making 223–32 decentralized agencies, EU, European Chemicals Agency (ECHA) and REACH 199, 223, 225, 227–32 bindingness 230

Guidance Documents 228–32 Guidance Documents, data collection 230–31 Guidance Documents, Partner Expert Group (PEG) 231 interest group involvement 231 Legal Notices 229, 232 mandate 228 and standardization 231–2 and translation 231, 232 deliberative democracy 89–90, 91, 92, 93, 289 democracy 87–99 civic and rights-based citizenship 90, 92, 93, 94 cosmopolitan democracy theories 92, 94 deliberative democracy 89–90, 91, 92, 93, 289 discursive democracy 94 electoral democracy at global level 92, 94 European Commission and soft law making 258 future research 95–6 global stakeholder democracy 94 legitimate participants 90 legitimate power and accountability 9, 91, 92, 93–4, 96 liberal democracy 88, 89, 90–92, 93, 94, 212 networked governance 94 ‘one person one vote’ principle 90 output legitimacy 90, 91, 93–4 and political decision-making 96 and political system, level of 91–2 and private governance 92–4, 95 republican and communitarian view and governance by the people 89, 90, 91, 92 and rule of law 91 social and cultural self-understanding 91 and transnational corporations 92–4 democratic accountability, US administrative guidance 243–4 Depledge, J. 325, 327, 328, 330 developing countries climate change law 323, 324, 325 economic analysis of soft law as regulatory tool 60, 65, 66, 67 Di Robilant, A. 72, 75, 76, 77, 202, 383 Diamond, L. 87, 94 Dietz, T. 16, 92 digital economy see under technology Dingwerth, K. 88, 92, 93 direct effect exclusion, public law 111–12 discrimination issues 194, 203, 342, 356 discursive democracy 94 Dixit, A. 11, 14 Djelic, M.-L. 10, 12, 19, 87, 92, 183, 425

Index  443 domestic factors and soft law cycle, EU 304–18 actors and organized interests 309 agenda-setting 306, 309, 310 and bindingness 305, 306, 307, 310, 313 communautarization of Justice and Home Affairs 307 domestic usage and soft norms 308–10 enforceability 307, 310, 311, 313 and EU multilevel system 306–11 EU, mutual influences and usage 308–9 EU policy fields 306–8 European Banking Authority (EBA) guidelines 304, 305 federalism and state aid 311–12 feedback in EU policy-making 310–11 future research 309–10, 314, 315 hard law and soft law, distinction between 305, 306–8, 309, 310 hybrid incorporations of hard and soft law 309 implementation assessment 308, 310–11 legalization process 307 major stakeholders 309 Open Method of Coordination (OMC) 305 oversight and control activities 310 regional aid maps 311–12 state features, effects of 311–14 see also national headings Drahos, P. 13, 15, 18 Dryzek, J. 90, 91, 92, 93, 94, 95, 96 due diligence duties, corporate responsibility 136–7 Dupuy, R. 43, 44, 49 Durmus, E. 338, 346 Džankić, J. 358, 361 Earth Summit 277 economic analysis of soft law as regulatory tool 56–70 Coase Theorem 58, 59 cooperative relationships and behavioural patterns 59–60 customary law 56, 58–9, 65–6 customary law and social norms, relationship between 59–60, 65 developing countries and recognition of soft law 60, 65, 66, 67 enforceability 56, 57, 58, 60, 61, 62, 63–4, 66–7 and game theory 58, 59, 60, 62–3 hybrid regulation 62 market failures, remedying 56, 57, 64, 65 policy guidelines and communications notices 57, 64, 66

property rights in close-knit societies, customary allocation 56, 58–9 social norms and customs 57–60 soft law disadvantages 66 stakeholder preferences 57, 65–7 stakeholders and interest group behaviour 64 state institutions, lack of trust in 60 economic analysis of soft law as regulatory tool, international soft law 57, 62–4 free rider issues 62–3 and new approach to sovereignty 63 and rational choice theory 57, 63 regime organization 63 and state cooperation 62 and treaty compliance 62 economic analysis of soft law as regulatory tool, self-regulation 56–7, 60–62 and co-regulation 61 cost factors 60–61 private regulation comparison 61–2, 66–8 and private sector 60–61 public interest perspective 60, 61, 66 and rent-seeking behaviour 61, 68 economic governance, EU 208–22 and agencification process 208, 211–13, 214, 217, 218 annulment cases 213, 215, 217, 218 Euro crisis effects 209, 210, 215, 216, 217 and European Central Bank (ECB) 211, 213–14, 217–18 European Commission and soft law making 254 European Parliament (EP) and democratic accountability 212–13 European Supervisory Authorities (ESA) 217, 218 European System of Financial Supervision (ESFS) 208–9, 213, 214, 217–19 future research 220 hard law and soft law, distinction between 209, 217 legal accountability 213, 214, 218 liberal democracy perspective 212 multiple accountability framework 212–14 multiple policy areas 210 Open Method of Coordination (OMC) 210–11 peer pressure as sanction mechanism 210–11 preliminary reference proceedings 213, 215, 217, 218–19 public law 107, 111 Single Resolution Mechanism (SRM) 217 Single Supervisory Mechanism (SSM) 217–18 social accountability 213–14

444  Research handbook on soft law soft law definition 209–10 transparency and accountability 212–14 economic governance, EU, Economic and Monetary Union (EMU) 208–9, 213, 214–19 Excessive Deficit Procedure (EDP) 214, 215 Excessive Imbalance Procedure (EIP) 215 and intergovernmentalism 216–17, 218 legal oversight 214, 215–16, 217, 218 legally binding nature of decisions, analysis of 215 Macroeconomic Imbalance Procedure (MIP) 215 and political oversight 214–17 public sector purchase (PSPP) 215–16 recommendations 214–16 economic marketing, smart cities 341–2 economic potentials, AI governance, EU 429, 433, 434–5 economic power, city regulation and governance 340 Edelman, M. 74, 164 effectiveness assessment gender equality, EU 369–73 and rule of law see rule of law crisis, remedies and effectiveness analysis soft law definition 47–8, 50 Efstathiou, K. 398, 403 Ehrlich, E. 11, 34, 35 Eick, A. 5, 304–18 electoral democracy at global level 92, 94 Eliantonio, M. 4, 11, 13, 19, 44, 47, 48, 108, 109, 110, 156, 191–207, 225, 226, 227, 232, 233, 248, 254, 256, 258, 263, 264, 305, 309, 310, 337, 413 Ellickson, R. 56, 58–9, 65, 68 Elliott, E. 239, 241 Elster, J. 11, 14, 88, 90, 91 Emerson, B. 4–5, 7, 203, 237–52 employment policy 106–7, 290, 292, 293, 435 energy sector 226, 232, 233, 261 enforceability 7, 14, 45–6, 48 anthropology of soft law 73, 75, 79, 80, 82 business and social norms in transnational governance 136, 137, 138 city regulation and governance 341, 344 courts and rule of law, EU 195, 202, 203 domestic factors and soft law cycle, EU 307, 310, 311, 313 economic analysis of soft law as regulatory tool 56, 57, 58, 60, 61, 62, 63–4, 66–7 EU Open Method of Coordination (OMC) 294, 298 European Commission and soft law making 257, 260

financial crisis effects, EU 409, 413, 414–17 gender equality, EU 372, 375, 376, 377, 378 public law 108, 111, 112 rule of law crisis 395, 397, 398, 401, 402 US administrative guidance 239, 240, 241, 242, 243 environmental law civil society 276–7, 278 courts and rule of law, EU 199 EU Open Method of Coordination (OMC) 289, 291, 298 European Commission 256, 263 smart cities 341, 345–6 US administrative guidance 240 see also climate change law epistemic governance, authority 4, 6, 162, 167, 170–72, 202 equality, gender see gender equality, EU Eriksen, O. 87, 93 ethics guidelines 135, 264, 342, 427 EU AI governance see AI governance, EU better regulation agenda see better regulation agenda, EU courts and rule of law see courts and rule of law, EU decentralized agencies see decentralized agencies, EU domestic factors and soft law cycle see domestic factors and soft law cycle, EU economic governance see economic governance, EU Economic and Monetary Union (EMU) see economic governance, EU, Economic and Monetary Union (EMU) Energy Union 291, 294, 296–7 financial crisis effects see financial crisis effects, EU gender equality see gender equality, EU Green Deal 289, 291 Intellectual Property Office (EUIPO) 226 ‘investor citizenship’ schemes 358 Istanbul Convention 381, 382 Lamfalussy report 263–4, 407–8, 409, 411 Maritime Safety Agency (EMSA) 226 Next Generation EU package 261 public law see public law, EU Recovery and Resilience Facility (RRF) 292, 294, 295, 296, 297 Single Resolution Mechanism (SRM) 217 Single Supervisory Mechanism (SSM) 217–18 Social Europe 291, 293, 296 Stability and Growth Pact (SGP) 290, 397

Index  445 technical standards see technical standards and European private law EU, cases 1. garantovaná v Commission 420 Akzo Nobel Chemicals and Akcros Chemicals v Commission 194 Amministrazione della finanze dello Stato v Salumi 194 Balgarska Narodna Banka 419 Banco Privado 110 Belgium v Commission [2015] 197 Belgium v Commission [2017] 232 Belgium v Commission [2018] 18, 110, 198, 256 Berlusconi 217 Białowieża forest 393 Bilka 374 Biocides 412 Birds Eye Walls 374 Bosman 154 Cassis de Dijon 253, 255, 259, 260, 261 Chrysostomides 216 Commission v CAS Succhi di Frutta 194 Commission v Council [2004] 17, 215 Commission v Council [2017] 102 Commission v Czech Republic [2016] 48 Commission v European Parliament and Council [2014] 412 Convention d’Istanbul 383 Dansk Rørindustri 110, 195 Defrenne 375–6 Dekker 374 Deufil 201 E and F 194 EBA’s Guidelines 218 ERTA 197 ESMA (UK v Parliament and Council) 212, 224, 232 European Parliament v Council [2012] 411 Expedia 195, 196, 265 FBF 417, 419, 420 FCD and FMB 199 Fédération bancaire française 200, 304 Fra.bo 154, 155, 156 France v Commission [1990] 198 France v Commission [1993] 109, 198 France v Commission [2004] 233 France v Commission [2012] 419 Gauweiler 17, 215, 216, 219 Grimaldi 48, 110, 196, 197, 199, 413 Groupement des Cartes Bancaires v Commission 420 Hauer 15 Hoechst v Commission 194 Hungary v Commission [2012] 420

Ijssel-Vliet 196, 197 James Elliott Construction 12, 154–5, 157, 198 Johnston 374 Koninklijke KPN 196, 197, 199, 265 Kotnik 110, 195, 196, 200 Kronofrance SA v Commission 311 Landerskreditbank 217 Laval 15 Ledra Advertising 216 Les Verts 391–2 Makhteshim-Again and Others v Commission 417 Marshall II 374 Meroni 211–12, 224, 232, 410, 411, 412, 416 Momentive Specialty Chemicals v ECHA 229–30 N.V. Elektriciteits v ECHA 228–9 Parliament v Commission [2022] 261 Peter Schönberger v European Parliament 420 Pfleiderer 195 Poland v European Parliament and Council [2018] 106 Polska Telefonia Cyfrowa 109, 255 Portuguese Judges 393 Pringle 17 Public.Resource.Org and Right to Know v Commission 153 Romano 211–12 Rütgers and Others v European Chemicals Agency 419 Seymour 374 Short Selling (United Kingdom v Parliament and Council) 411, 414, 419 Slovak Republic and Hungary v Council [2017] 106 SolarWorld and Others v Commission 420 SV Capital OU. v EBA 419 Telefónica v Commission 419 Test-Achats 374 US Steel Košice v Slovak Republic 420 Von Colson 374 Weiss 215–16, 219 EU, regulations Aarhus Convention 298 Artificial Intelligence Act 147, 156 Aviation Safety Agency (EASA) 226–7 Capital Requirements Regulation and Directive 412, 413 Committee of European Banking Supervisors (CEBS) 408, 413 Committee of European Securities Regulators (CESR) 408

446  Research handbook on soft law Common European Sales Law (CESL) 38, 149 Common Foreign and Security Policy (CFSP) 112 Conference on Security and Cooperation in Europe (CSCE) 50 Consumer Rights Directive 39 Consumer Sales Directive 150 Council of Europe Framework Convention for the Protection of National Minorities (FCNM) 357 Digital Single Market Strategy 344–5 E-Commerce Directive 344 Emissions Trading System (ETS) Directive 291 European Stability Mechanism (ESM) Treaty 17 Excessive Deficit Procedure (EDP) 16–17 Framework Convention for the Protection of National Minorities 80 Gender-balanced Boards Directive 373–4 Habitats Directive 263 Integrative Pollution Prevention and Control Directive 307 Justice Scoreboard 398–9 Minimum Wages Directive 111 National Energy and Climate Plans (NECPs) 291, 294, 297–8 Non-Financial Reporting Directive 135 Omnibus Directive 156 Package Travel Directive 259 Principles of European Contract Law (PECL) 37 Product Liability Directive 152, 186 Product Safety Directive 151–2 Return Directive 262 Services Directive 147, 154, 260, 344 Single European Act 146 Stability and Growth Pact 16–17 Timber Regulation 62 Transparency in Equal Pay Directive 310 Unfair Terms Directive 153–4, 155 Water Framework Directive 199, 202, 231 Work-Life Balance Directive 373 EU Open Method of Coordination (OMC) 13, 18–19, 50, 72, 107, 191, 288–303 and better regulation agenda 120 citizenship regimes 358 and civil society 297–8 climate and energy policy 290–91, 292, 294, 296–8 coercive conditionality 297 country-level recommendations (CSRs) 290, 294, 295 Covid-19 pandemic effects 291, 295

and deliberative democracy 289 design 288–9 domestic factors and soft law cycle, EU 305 and domestic policy change 292–3 economic governance, EU 210–11 education sector 296 employment policy 290, 292, 293 Energy Union 291, 294, 296–7 enforceability 294, 298 environmental law 289, 291, 298 European Commission and soft law making 253, 262, 263 European Pillar of Social Rights (EPSR) 291–2, 293, 294, 295, 297 European Semester 290, 291, 292, 294, 297 European Social Fund (ESF) 295, 298 and Eurozone crisis 292, 296 Governance Regulation 291, 292–4 hybridization 292, 294–5 Juncker Commission and flagship initiatives 291, 297 Macroeconomic Imbalance Procedure (MIP) 290 monitoring and benchmarks 294 National Energy and Climate Plans (NECPs) 291, 294, 297–8 neoliberal values 293 policy coordination 289–90, 291, 297 as political leverage tool 295–8 PROGRESS-peer reviews in social inclusion 290 and public law 107 reciprocal influence on Member States 295–7 Recovery and Resilience Facility (RRF) 292, 294, 295, 296, 297 and Social Europe 291, 293, 296 social inclusion 295 Stability and Growth Pact (SGP) 290, 397 steering instruments 292, 295, 297, 298 European Accreditation (EA) 181–2 European Asylum Agency 105 European Banking Authority (EBA) 105, 304, 305, 413, 414 European Central Bank (ECB) 211, 213–14, 217–18 European Chemicals Agency (ECHA) see decentralized agencies, EU, European Chemicals Agency (ECHA) and REACH European Civil Code 145 European Commission and soft law making 253–71 Agency for the Cooperation of Energy Regulators (ACER) 261 Christmas notices 253

Index  447 collaborations 253, 254, 258–9, 261, 262, 265 competition law 254, 261, 264–5 Cooperation and Verification Mechanism (CVM) 253 Court of Justice interaction 261 COVID-19 pandemic effects 254, 259–60, 261–3, 264 democracy principle 258 economic policy coordination 254 energy sector and network guidelines 261 enforceability 257, 260 environmental law 256, 263 Ethics Guidelines on Trustworthy Artificial Intelligence (AI) 264 European Data Protection Board 261–2 financial sector and Lamfalussy Committees 263–4 flexibility in soft law decision making 254, 255, 263 future research 259, 265 gender equality 374–5 general EU law application 254–8 and harmonization 255 interpretative guidance 254, 263 intra-institutional exchanges 259–62 Joint Research Centre 259 legal basis and Commission competence 256–8 legitimacy concerns 258 Member State interactions 262–3 migration 254, 262 mutual recognition principle 253, 255 Next Generation EU package 261 online gambling 256, 260 Open Method of Coordination (OMC) 253, 262, 263 Parliamentary involvement 260 public law 104–5, 106, 110, 111 Recommendations 260–62 regulation by information 253, 255–6, 259 regulation by information, translations 255–6 revision/transformation 264–5 and rule of law 258, 261 sincere cooperation principle 258, 265 soft law transformed into hard law 264–5 stakeholder involvement and consultation 263–4 State aid measures 262, 264 telecommunications sector 265 Ukraine war effects 262–3 European Data Protection Board 261–2 European Electronic Communications Code 265 European Equality Law Network 377 European Medicines Agency (EMA) 226–7

European Pact for Gender Equality 380 European Pillar of Social Rights (EPSR) 291–2, 293, 294, 295, 297 European Semester 395, 396, 397–8, 399 European Standardisation System (ESS) 146 European Supervisory Authorities (ESAs) 217, 218, 409–16, 417–18 European System of Financial Supervision (ESFS) 208–9, 213, 214, 217–19, 409 European Systemic Risk Board (ESRB) 409 expert advice 167–9, 171, 184–5 fair trial access 194, 203, 245 Falk, R. 77, 326, 329 Falke, J. 150, 151, 152, 156 Falkner, G. 46, 305, 308, 326, 329, 383 Faure, M. 2–3, 56–70 federalism and state aid 311–12 Ferguson, J. 71, 76, 78 Ferrarese, M. 75, 83 financial crisis effects, EU 407–22 Capital Requirements Regulation and Directive 412, 413 Committee of European Banking Supervisors (CEBS) 408, 413 Committee of European Securities Regulators (CESR) 408 ‘comply and explain’ approach 408, 417–18 De Larosière Report 409 draft technical standards 409–12 economic governance, EU 209, 210, 215, 216, 217 enforceability 409, 413, 414–17 and EU Open Method of Coordination (OMC) 292, 296 European Banking Authority (EBA) 413, 414 European financial supervision (ESFS) 409 European Supervisory Authorities (ESAs) 409–16, 417–18 European Supervisory Authorities (ESAs), Article 16 Guidelines 412–14, 416, 417–18 European Supervisory Handbook 414 European System of Financial Supervisors 409 European Systemic Risk Board (ESRB) 409 Financial Service Action Plan 407 hard and soft law mix 414–16 intervention powers in emergency situations 415 judicial review 412, 416–17 Lamfalussy report 407–8, 409, 411 legal obligation 417, 418

448  Research handbook on soft law and national competent authorities (NCAs) 413–14, 418 quasi-hard law powers 412–14 regulatory system evolution 407–8 regulatory system reform 409–16 settlement of disagreements between national supervisory authorities 415–16 Finck, M. 338, 343 Finnemore, M. 47, 272 Fishkin, J. 89, 90, 93, 96 Fjeld, J. 427, 429 flexibility climate change mitigation issues 328 European Commission and soft law making 254, 255, 263 Fligstein, N. 11, 13 Flyverbom, M. 6, 7, 423–37 Fon, V. 56, 57, 62, 66 food 242, 245–6, 313–14 Fossum, J. 87, 89 Foucault, M. 7, 14, 78, 425, 434 free trade agreements, public law, hard/soft law distinction 111–12 free-rider issues, economic analysis of soft law 62–3 Friedman, M. 92, 133 Frug, G. 338, 340 fundamental rights see human rights future research AI governance, EU 435–6 better regulation agenda, EU 128 citizenship regimes 361, 363 city regulation and governance 339, 346, 347 courts and rule of law, EU 205 decentralized agencies, EU 233 domestic factors and soft law cycle, EU 309–10, 314, 315 economic governance, EU 220 European Commission and soft law making 259, 265 governance through soft law 19 rule of law crisis 403 soft law definition 43–4, 51 Galland, J.-P. 180, 186 gambling 256, 260 game theory, and economic analysis of soft law 58, 59, 60, 62–3 Garben, S. 116, 117, 262 gatekeeping exercises, decentralized agencies 227 gender equality, US transgender rights 246–7 gender equality, EU 368–90 Action Plan on Gender Equality and Women’s Empowerment 374

assessment framework and benchmarks 369–70, 371–3 Commission, Council and Parliament instruments 374–5 compliance assessment 371, 372 COVID-19 pandemic and gender gap 368 decision-making processes and bodies, balanced participation 370, 377–9 effectiveness assessment 369–73 enforceability 372, 375, 376, 377, 378 equal pay case study 375–7 European Equality Law Network 377 European Pact for Gender Equality 380 gender biases and stereotypes 368, 369 Gender Equality Strategy for 2020–2025 375, 381 gender-balanced company boards 378 gender-based violence 379–82 gender-based violence, cyber violence 381 gender-based violence, legislation, issues with 381–2 gender-neutral language in European Parliament 375 Global Gender Gap Report 368 guidance documents 370 hard law 373–4 implementation assessment 371, 372 interpretative acts 370 Istanbul Convention 381, 382 para-law function of soft law 370–71, 372, 376, 378–9, 380, 381, 382 Pay Transparency Recommendation 375, 376 private sector 368, 378, 379, 380 recommendations, role of 370–71, 372, 375–7, 378–9, 380 sex discrimination directives 373–4 soft law concept 369–71, 374–5 steering acts 370–71, 372 under-representation in political decision-making 368 work-life balance 373, 374–5 Gentile, G. 197, 227, 304 Georgieva, Z. 108, 198, 265, 305, 309, 310, 372 Gersen, J. 237, 239 Gleckman, H. 87, 92 global anti-corruption regime, civil society 278–80 Global Citizenship Compact consideration 362 Global Citizenship Observatory’s (GLOBALCIT) Citizenship Law Dataset 360, 361 global dimension, city regulation and governance 339–41 Global Gender Gap Report 368 global governance 162–76 cooperation role 166, 170

Index  449 and Covid crisis 168 epistemic authority 170–72 epistemic authority, disclaimers 172 expert governance, attitudes towards 167–9, 171 and idealism 163, 164 and indicators, use of 170 judicial review 166, 168 language and labels 163–5 law and politics, relationship between 166–7 and legalism 163–4, 165 liberal institutionalists 163 and political priorities 168–9 and rankings, use of 170 and realism 163, 164, 165 and recommendations and guidelines 168 and rule of law 167 soft law benefits 165–7 soft law pitfalls 167–9 and UN resolutions 169–70 global legal environment, and anthropology of soft law 71, 72, 75–6, 77, 80, 81 Global Observatory on AI 342 Global Smart Cities Alliance on Technology Governance 345–6 global social problems, response to, civil society 278–80 Goldmann, M. 77, 170 Goldsmith, J. 57, 62, 63, 164 Goodin, R. 88, 89, 94 Gornitzka, A. 107, 288, 293, 296 governance, epistemic 4, 6, 162, 167, 170–72, 202 governance, global see global governance governance indicators, citizenship regimes 359–62 governance through soft law 10–30 algorithmic governance 19 executive management and MoU 15, 16–18, 19 fundamental rights 15, 18 future research 19 global financial and product markets and ‘new global rulers’ 12 and legal pluralism 10, 11, 14–15, 19 national courts and administrations 13–14 post-Covid effects 19–20 and public interest 15 and rule of law 15 and social control 11 soft law definition 10–14 soft law and hard law, distinction between 10–11, 12, 14–16 sustainable governance 18–20 and transnational governance 11–12, 15–16

Graziano, P. 209, 296, 297 Greco, S. 256, 372, 383 greenhouse gas emissions see under climate change law Grogan, J. 6, 194, 391–406 group interests and corporate responsibility 135–6 Guérin, E. 324, 329 Gupta, A. 71, 76 Gustafsson Nordin, I. 4, 88, 147, 177–89, 427 Gutmann, J. 56, 60, 65, 66, 68 Guzman, A. 43, 63–4, 69 Habermas, J. 89, 90, 91, 92, 93, 94, 96, 137, 238 Hagemann, R. 338, 341 Halliday, T. 10, 16, 75 Handler Chayes, A. 63, 69 hard law better regulation agenda 122–3 business and social norms in transnational governance 136, 137 city regulation and governance, smart cities 344 gender equality, EU 373–4 rule of law crisis, remedies and effectiveness analysis 396 see also hybridization hard law/soft law distinction civil society 275, 280–83 climate change law 322–3, 325, 326 courts and rule of law, EU 200–201 domestic factors and soft law cycle, EU 305, 306–8, 309, 310 economic governance, EU 209, 217 European Commission and soft law making 264–5 financial crisis effects, EU 414–16 governance through soft law 10–11, 12, 14–16 public law see public law, EU, hard/soft law distinction, mitigation of soft law definition 44–6, 47, 48–50 Hartlapp, M. 5, 47, 198, 304–18, 372, 396 Harvey, D. 75, 77 Held, D. 15, 74, 88, 92 Herbertson, K. 328, 330 Héritier, A. 91, 394 Hervey, T. 191, 204 history of soft law 31–42 American restatements 34, 36–7 codifications 31–3, 34, 35, 36, 37–8 courts, rise of, and judge-made law 35–6 Decretum Gratiani 32 free law movement 35 Justinian’s Digest 32

450  Research handbook on soft law pre-modern law and non-legislative codifications 31–3 pre-modern law and political domination 33–4, 35 re-discovery of soft law 34–8 Saxon Mirror 32 Hodson, D. 214, 254 Hofmann, A. 47, 198, 305, 308, 372, 395, 396 Hofmann, H. 201, 253 Holzer, B. 3–4, 87, 92, 93, 132–44 home-sharing and Airbnb, smart cities 344 human rights 15, 18, 43, 50 AI governance, EU 429–30, 431, 432, 435 anthropology of soft law 81–2 citizenship regimes 355–6, 357, 358–9, 361 city regulation and governance 340 civil society 280–83 EU Charter of Fundamental Rights 198, 200, 216 smart cities 343 Huovila, A. 341, 346 hybridization city regulation and governance 346, 347 courts and rule of law, EU 199–200 economic analysis of soft law as regulatory tool 62 EU Open Method of Coordination (OMC) 292, 294–5 and soft law definition 50 see also hard law idealism, and global governance 163, 164 immigration see migration impact assessment (IA), better regulation agenda 119, 121, 122–3, 124–6, 127–8 indigenous peoples, citizenship regimes 358–9, 360, 361 information, regulation by, European Commission 253, 255–6, 259 institutional balance principle, public law 103–6 institutional practice, public law, hard/soft law distinction 111–12 Inter-American Court of Human Rights Ivcher-Bronstein v Peru 355 Yean and Bosico Girls v Dominican Republic 355 interest group involvement 61, 64, 231, 314 see also stakeholder involvement Intergovernmental Panel on Climate Change 277 intergovernmentalism, and economic governance 214, 216–17, 218 International Accreditation Forum (IAF) 182, 186 international arbitration, civil society 277–8 International Civil Aviation Organization (ICAO) 172

International Convention on the Elimination of All Forms of Racial Discrimination (CERD) 356 International Court of Justice 43, 62, 354 Namibia Advisory Opinion 321 International Covenant on Civil and Political Rights (ICCPR) 355, 356, 357 international governance institutions, AI governance, EU 426–7 International Labour Organization 50 International Organization for Migration 172 International Organization for Standardization (ISO) 177 international standards and the dilution of responsibility 177–89 characteristics of standards 178–9 choice and decision-making 184–6 compliance monitoring systems 179–82 compliance monitoring systems, accreditation of certification 181–2, 186 compliance monitoring systems, certification 180–81 EU New Approach and harmonized standards 180–82 European Accreditation (EA) 181–2 expert advice 184–5 future research 182 history 178–9 International Accreditation Forum (IAF) 182, 186 International Organization for Standardization (ISO) 177 management system standards 177, 179, 180–81 organization theory 178, 182–4, 185 and public interest 178, 181 responsibility issues 184–6 responsibility issues, choice and decision-making, accountability issues 185–6 soft law guidelines, recommendations or standards comparison 183–4 standards as private rules 178 verification markets 179 voluntary status 178, 179, 180, 183, 184, 185–6 International Telecommunication Union (ITU) 172, 345–6 International Union for Conservation of Nature and Natural Resources (IUCN) 277 International Whaling Commission (IWC) 278 interpretative role courts and rule of law, EU 200

Index  451 European Commission and soft law making 254, 263 gender equality, EU 370 soft law definition 49 US administrative guidance 240, 243 intra-institutional exchanges, European Commission 259–62 investor citizenship, EU 358, 362–3 Jacob-Owens, T. 6, 353–67 Jacobsson, B. 11, 178 Jacobsson, K. 47, 210, 288, 298, 383 Jansen, N. 2, 31–42, 68, 72, 145 Jasanoff, S. 172, 275 Joerges, C. 12, 17, 18, 146, 151 Jordan, A. 298, 320 Jordana, J. 4, 92, 96, 107, 208–22 judicial review courts and rule of law, EU 194–8, 203 decentralized agencies, EU 227 financial crisis effects, EU 412, 416–17 global governance 166, 168 public law, hard/soft law distinction, mitigation of 109–10 and rule of law crisis 391–2, 393 and soft law definition 48 technical standards and European private law 153–5 US administrative guidance 241 Jütersonke, O. 163, 166 Kant, I. 137, 242 Kantorowicz, H. 35–6 Katcherian, J. 72, 78, 79, 80, 82 Kennedy, D. 166, 171 Keppenne, J. 107, 111 Kessler, J. 203, 245 Kilpatrick, C. 13, 19 Kingsbury, B. 19, 274 Kirton, J. 72, 76, 77 Klabbers, J. 4, 7, 19, 44, 68, 72–3, 75, 162–76, 191, 200–201, 225, 237, 272, 305, 320, 322, 323 Knodt, M. 291, 294, 299, 308 Kochenov, D. 392, 393, 400 Komesar, N. 11, 199, 202 Korkea-aho, E. 4, 13, 19, 47, 48, 50, 87, 109, 110, 191–207, 225, 227, 231, 233, 237, 265, 305, 309, 310 Koskenniemi, M. 15, 320, 322 Kruisinga, S. 379, 383 Kulovesi, K. 5, 320–36, 354 Kymlicka, W. 354, 356 Kyoto Protocol, climate change law 63, 320, 324, 325, 326, 327, 328, 329, 330, 332

labelling AI governance, EU 427, 430, 431, 432–3, 435 global governance 163–5 US administrative guidance 245–6 Ladeur, K.-H. 19, 153 Láncos, P. 256, 260, 266, 372 Lavergne, B. 43, 45 legal accountability, economic governance, EU 213, 214, 218 legal basis and competence, European Commission 256–8 legal certainty principle 102, 109, 193, 194, 195, 197–8, 201, 203 legal effects, soft law definition 43–4, 45–6, 47–8, 50 legal obligation 10–11 civil society 273–4 climate change law 326–7, 330 financial crisis effects, EU 417, 418 public law, EU 104, 109 soft law definition 46, 47 legal oversight, Economic and Monetary Union (EMU) 214, 215–16, 217, 218 legal pluralism 10, 11, 14–15, 19 and anthropology of soft law 74–5, 76, 81 legalism, and global governance 163–4, 165 legality principle, courts and rule of law, EU 193, 194 legitimate expectations principle, courts and rule of law, EU 195, 201 legitimate power and accountability 9, 91, 92, 93–4, 96 Levi-Faur, D. 13, 92, 224 Levin, R. 237, 239, 240, 241, 243, 244 liberal democracy 88, 89, 90–92, 93, 94, 212 liberal institutionalists 163 López-Santana, M. 295, 305 Lorenzo, R. 77, 80–81, 82 Luhmann, N. 132, 136, 137, 139, 183 McAdam, D. 11, 13 MacDonald, T. 94, 95 Maher, I. 214, 254, 358 Majone, G. 18, 232 management system standards 177, 179, 180–81 Marburger, P. 149, 151 March, J. 90, 139, 141, 183 Marchant, G. 338, 424 Marjosola, H. 200, 263 market failures, remedying, economic analysis of soft law 56, 57, 64, 65 markets of verification, and international standards 179 Mattei, U. 14, 15, 19, 72, 75, 76, 77, 183

452  Research handbook on soft law Mattli, W. 10, 12, 87, 138, 149, 162 Mendelson, N. 237, 241, 242 Merry, S. 11–12, 14, 73, 75, 77, 81–2, 172 Metzger, G. 240, 241, 243 Meyer, J. 134, 139, 140, 185, 187 Michaels, R. 34, 38, 145, 272 Micklitz, H.-W. 4, 19, 145–61, 180, 186, 427 migration asylum seekers and public law 104–5 European Commission 254, 262 immigrant integration, citizenship regimes 360–61 refugees 81, 361 US administrative guidance 243–4, 245, 246–7 Mitchell, R. 63, 69, 371–2 Moloney, N. 407, 408, 410, 415 Moore, S. 11, 71, 74, 76, 78, 80, 81 Mörth, U. 3, 7, 12, 87–99, 108, 183, 212, 360 multicultural citizenship 354, 356–7, 358–9, 360 multiple accountability framework, economic governance, EU 212–14 Murphy, C. 11, 177, 178, 179 mutual recognition principle, European Commission 253, 255

EU Open Method of Coordination (OMC) 293 networked governance, and democracy 94 New Approach development international standards and the dilution of responsibility 180–82 technical standards 146, 147, 150, 151, 152, 153, 156 New Framework to Strengthen the Rule of Law 399–400 new governance context, courts and rule of law, EU 202, 204 NGOs 74, 197, 198 business and social norms 132–3, 136, 137, 138, 139, 140 civil society 276, 278, 279, 283 Niglia, L. 11, 17, 18, 19 Nijman, J. 15, 346 non-law, soft law as 46–8, 77–8 non-tariff barriers to trade, and technical standards 146, 148 North, D. 56, 65, 279 notice-and-comment rulemaking process, US 239–40, 241, 243 Nott, S. 43, 305

Nader, L. 15, 71, 72, 73, 74, 75 Nanz, P. 289, 297 national competent authorities (NCAs), and financial crisis effects, EU 413–14, 418 national equality bodies, public law 107–8 national and private sector initiatives, AI governance, EU 427 national regulations 13–14 AI governance, EU 429 and anthropology of soft law 76 better regulation agenda 122–3 and citizenship regimes 353 courts and rule of law, EU 195–7, 198, 199, 200 domestic factors and soft law cycle see domestic factors and soft law cycle economic analysis of soft law 60, 62 and EU Open Method of Coordination (OMC) 292–3 national regulatory authorities (NRAs) 226–7 public law, hard/soft law distinction, mitigation of 110 nationality right, citizenship regimes 354, 355–6 Nationally Determined Contributions (NDCs), climate change law 329–30 Nelken, D. 12, 361 neoliberal principles AI governance, EU 433, 434–5 anthropology of soft law 71, 72, 75, 77

Oberthür, S. 291, 325, 326–7, 329 ‘one person one vote’ principle, democracy 90 Oomen, B. 337, 338, 340, 346 organization theory, international standards and the dilution of responsibility 178, 182–4, 185 output legitimacy, democracy 90, 91, 93–4 para-law function, gender equality, EU 370–71, 372, 376, 378–9, 380, 381, 382 Paris Agreement, climate change law 63, 320, 325, 326–7, 328–32 Paris Club 50 Parisi, F. 6, 56, 57, 58, 62, 66 Parrillo, N. 237, 242–3 Pauwelyn, J. 44, 272, 320, 323, 326 Pech, L. 391, 393, 400, 401, 402 peer review 120, 182, 202, 290, 294 Peerenboom, R. 11, 15 Permanent Court of International Justice, Nationality Decrees Issued in Tunis and Morocco 355 Peters, A. 43, 50, 208 Petri, M. 227, 232, 233 Petropoulou Ionescu, D. 44, 256, 258, 263, 264 Philipsen, N. 2–3, 56–70 Pochet, P. 107, 292 policy coordination, EU Open Method of Coordination (OMC) 289–90, 291, 297

Index  453 policy guidelines, economic analysis of soft law as regulatory tool 57, 64, 66 policy process and politics, better regulation agenda 118, 120, 121, 122 political decision-making, and democracy 96 political impact, US administrative guidance 247–9 political leverage tool, EU Open Method of Coordination (OMC) 295–8 political oversight, and economic governance, EU, Economic and Monetary Union (EMU) 214–17 political priorities better regulation agenda, EU 122, 128 and global governance 168–9 politics and law, relationship between, global governance 166–7 Pollack, M. 34, 133, 320, 322, 337 populism 7 Posner, E. 57, 59–60, 62, 63, 68, 164, 237, 274 Powell, W. 139, 183 preliminary reference proceedings, economic governance, EU 213, 215, 217, 218–19 privacy issues AI governance, EU 426, 429, 432 smart cities 338, 342–3, 347 private governance, and democracy 92–4, 95 private law, and technical standards see technical standards and European private law private regulation, economic analysis of soft law, self-regulation comparison 61–2, 66–8 private sector anthropology of soft law 80–81 and decentralized agencies, EU 227 and economic analysis of soft law 60–61 gender equality, EU 368, 378, 379, 380 product safety and liability, technical standards 151–2, 153, 154, 156 profit, pursuit of, business and social norms in transnational governance 133, 134, 135 property rights in close-knit societies, economic analysis of soft law 56, 58–9 public interest 15 economic analysis of soft law, self-regulation 60, 61, 66 international standards and the dilution of responsibility 178, 181 public law, EU 101–15 annulment cases 109–10 and asylum seekers 104–5 bindingness 101, 102–3, 105, 106, 107, 108–9, 110, 111–12 and Commission 104–5, 106, 110, 111 competition law 108, 110 complementary role of soft law 107–8

conferral principle 12, 102–3, 110 Country-Specific Recommendations (CSRs) and budgetary surveillance 107 and COVID-19 pandemic 106, 108 economic governance 107, 111 and emergency measures 108 employment policy 106–7 enforceability 108, 111, 112 and European Council 104, 105, 106, 111–12 and European Parliament 106 implementation of EU legislation and institutional balance 105 institutional balance principle 103–6 legal certainty principle 102, 109 legal obligation 104, 109 legal system 106–8 national equality bodies 107–8 and Open Method of Coordination (OMC) 107 post-legislative rule-making 107–8 precursor role of soft law 107–8 and recommendation adoption 103, 104–5, 109–10 sincere cooperation principle 111 soft law concept 101–2 state aid law 102–3, 108, 110 supplementary EU competences 103 and text of Treaties 103–4 public law, EU, hard/soft law distinction, mitigation of 102, 107, 108–12 comply-or-explain mechanisms 111 direct effect exclusion 111–12 free trade agreements 111–12 institutional practice 111–12 judicial review 109–10 national courts 110 public participation, US administrative guidance 238–9, 244–5 public sector purchase (PSPP), Economic and Monetary Union (EMU) 215–16 racial discrimination 80–81, 356 Radaelli, C. 3, 72, 106, 116–31, 297 Rajamani, L. 324, 325, 326, 329 Ranchordás, S. 5–6, 337–52 rational choice theory, international soft law 57, 63 realism, and global governance 163, 164, 165 Recio, M. 5, 320–36, 354 recommendations, role of EU, Economic and Monetary Union (EMU) 214–16 European Commission and soft law making 260–62

454  Research handbook on soft law gender equality, EU 370–71, 372, 375–7, 378–9, 380 and global governance 168 and public law 103, 104–5, 109–10 rule of law crisis 397 rule of law crisis, remedies and effectiveness analysis 396 soft law definition 49 refugees see under migration regional aid maps 311–12 regulation and AI governance, EU 424, 426–30 better regulation agenda see better regulation agenda, EU by information, European Commission 253, 255–6, 259 city see city regulation and governance regulatory system reform, financial crisis effects, EU 409–16 remedies, and rule of law see rule of law crisis, remedies and effectiveness analysis rent-seeking behaviour, self-regulation 61, 68 republican and communitarian view, democracy 89, 90, 91, 92 responsibility corporate 136–8 and international standards see international standards and the dilution of responsibility Reuter, T. 340, 342 Riles, A. 77, 78, 83 Rio Declaration on Environment and Development 276, 277 risk, and AI governance, EU 424, 427, 428, 429–30, 431–2, 435 Rocca, P. 225, 226, 232 Rose-Ackerman, S. 19, 239 Rosenau, J. 12, 138, 278 Rowan, B. 139, 185, 187 Rubio, N. 198, 306 rule of law and courts see courts and rule of law, EU and democracy 91 and European Commission and soft law making 258, 261 and global governance 167 and governance through soft law 15 threat 44 rule of law crisis 391–406 autocracies and one-party states, rise of 393 backsliding 261, 391, 392, 393 compliance and sanctions 392, 393, 398 country-specific recommendations (CSRs) 397–8, 399 enforceability 395, 397, 398, 401, 402

EU Council’s Annual Rule of Law Dialogue 401 EU Justice Scoreboard 398–9 EU Lisbon Treaty 392 EU Rule of Law Review Cycle 401–2 and European Semester 395, 396, 397–8, 399 future research 403 and judicial review 391–2, 393 Macroeconomic Imbalance Procedure (MIP) 397 New Framework to Strengthen the Rule of Law 399–400 Poland and Hungary 393, 400, 401 recommendations 397 rule of law concept and development in EU 391–2 Stability and Growth Pact (SGP) 397 Tavares Report 393 threats caused by 392–3 rule of law crisis, remedies and effectiveness analysis 394–7 hard law relationship 396 institutional adoption 396 and policy changes 396–7 recommendations and national implementation 396 soft law definition and rules of conduct 394 steering instruments 394–5, 397 Sabel, C. 13, 203, 245 safe harbours sanctuary cities 340 US administrative guidance 242 Sahlin-Andersson, K. 12, 87, 92, 183, 425 sanctions and corporate responsibility 136 economic governance, peer pressure as sanction mechanism 210–11 rule of law crisis 392, 393, 398 sanctuary cities 340 Sandvik, K. 77, 81, 82 Sassen, S. 74–5, 338, 339 Saurugger, S. 49, 126, 209, 210, 305, 307, 308, 396 Schade, F. 6, 7, 423–37 Schammo, P. 217, 410 Scharpf, F. 15, 17, 18, 87, 96, 293, 306, 312, 371 Schauer, F. 39, 164 Schemmel, J. 217, 410, 413 Schepel, H. 12, 17, 145, 146, 150, 151, 152, 156 Scheppele, K. 392, 393 Schmidt, V. 293, 371 Schoenefeld, J. 299, 308 Scott, J. 13, 191, 200, 202, 203, 225, 226, 228, 232, 233, 263, 358

Index  455 self-interest, enlightened, and corporate responsibility 134–5 self-regulation anthropology of soft law 81 city regulation and governance 340 economic analysis see economic analysis of soft law as regulatory tool, self-regulation technical standards and European private law 153 voluntary, and corporate responsibility 135, 136 Senden, L. 6, 12, 15, 43, 46, 48, 49, 50, 68, 91, 101, 104, 107, 191, 225, 232, 253, 255, 305, 337, 368–90, 394, 395 sex discrimination directives, gender equality, EU 373–4 Shaffer, G. 10, 16, 34, 133, 320, 322, 337 Shaw, J. 6, 339, 353–67 Sikkink, K. 135, 272 sincere cooperation principle 48, 111, 195, 258, 265 see also collaborations and cooperation Slaughter, A.-M. 326, 329 Slominski, P. 105, 128, 254, 262 smart cities see city regulation and governance, smart cities Snidal, D. 43, 45, 77, 133, 138, 184, 209, 210, 320, 322, 435 Snyder, F. 2, 6, 10–30, 43, 46, 68, 71, 72, 75, 87, 88, 89, 101, 184, 201, 205, 225, 253, 305, 337, 371, 372, 395 social and cultural self-understanding, democracy 91 social inclusion, EU Open Method of Coordination (OMC) 295 social norms economic analysis of soft law as regulatory tool 57–60 transnational governance see business and social norms in transnational governance social ordering in small-scale societies 73 soft law definition and concept 1–2, 43–55 anthropology of soft law 71–3 and court decisions 48 decisional instruments 49 economic governance, EU 209–10 effectiveness assessment 47–8, 50 enforcement versus delegation 45–6, 48 formalization and rules of conduct 46–7 future research 43–4, 51 gender equality, EU 369–71, 374–5 governance through soft law 10–14 and guiding principles 49–50

and hard law interactions 48–50 hard law/soft law distinction see hard law/ soft law distinction and hybridity 50 and international relations 43 interpretative instruments 49 and judicial review 48 legal effects 43–4, 45–6, 47–8, 50 legal obligation 46, 47 multidimensional approach and weaknesses 45–6 and new modes of governance 43 policy change effects 47 public law, EU 101–2 recommendations, adopting 49 rule of law crisis, remedies and effectiveness analysis 394 as rule of law threat 44 sincere cooperation principle 48 soft law as distinct from non-law 46–8 Solanes Mullor, J. 4, 96, 107, 208–22 Spasova, S. 291, 295 Spiro, P. 355, 362 stakeholder involvement AI governance, EU 430, 431 better regulation agenda 123–4 business and social norms in transnational governance, corporate responsibility 135–6 city regulation and governance 346–7 domestic factors and soft law cycle, EU 309 economic analysis of soft law as regulatory tool 57, 64, 65–7 European Commission and soft law making 263–4 see also interest group involvement standardization, European Chemicals Agency (ECHA) and REACH 231–2 standards city regulation and governance, smart cities 346–7 international see international standards and the dilution of responsibility technical see technical standards and European private law state aid courts and rule of law, EU 194–5, 196 European Commission and soft law making 262, 264 and federalism and 311–12 public law 102–3, 108, 110 state regulation see ‘national’ headings statelessness, citizenship regimes 353, 355, 356, 359, 360, 361 steering instruments 50

456  Research handbook on soft law AI governance, EU 425 EU Open Method of Coordination (OMC) 292, 295, 297, 298 gender equality, EU 370–71, 372 rule of law crisis 394–5, 397 Ştefan, O. 5, 10, 13, 44, 47, 48, 106, 108, 191, 194–5, 196, 200, 209, 227, 232, 233, 253–71, 305, 306, 372, 394 Stiller, S. 5, 72, 106, 107, 191, 210, 253, 288–303, 305, 358 Strelkov, A. 394, 396, 398, 399, 402 Sturm, S. 200, 202, 203 Sunstein, C. 168, 239 supply chains, and corporate responsibility 132, 136–7, 138 supranational capitalism, anthropology of soft law 75 sustainable governance 18–20 Swiney, C. 338, 339–40, 346 Sykes, A. 62, 63, 274 Taffoni, G. 3, 106, 116–31 Tamanaha, B. 11, 14, 15 Tamayo-Álvarez, R. 5, 192, 272–87 Tamm Hallström, K. 4, 88, 147, 177–89, 427 Tavares Report, rule of law crisis 393 technical standards and European private law 145–61 bindingness 146–7 Common European Sales Law (CESL) 149 contract law 148–51, 156 contract law, and national courts 150 copyright protection 153 digital economy strategy and AI 147, 155–6 EU Consumer Sales Directive 150 EU Services Directive 147, 154 EU Unfair Terms Directive 153–4, 155 and European Civil Code 145 European Standardisation System (ESS) 146 judicial review 153–5 linkages between 147–53 New Approach development 146, 147, 150, 151, 152, 153, 156 New Legislative Framework (NLF) 146, 153 and non-tariff barriers to trade 146, 148 process standards 149 product safety and product liability 151–2, 153, 154, 156 self-certification and CE label 147 self-regulation 153 and Single European Act 146 standard harmonization 146–7 and standardization 146–7, 148, 153–6 tort law 151–2

and UN Convention on Contracts for the International Sale of Goods (CISG) 149–50 technology AI governance see AI governance, EU algorithmic governance 19 city regulation and technological innovation challenges 338, 339, 340–41 digital rights, smart cities 342–3, 344 draft technical standards, financial crisis effects, EU 409–12 gender-based cyber violence 381 and smart cities 341–2 telecommunications sector 172, 265, 345–6 Terpan, F. 2, 43–55, 87, 126, 201, 209, 210, 305, 307, 308, 383, 396 Teubner, G. 74, 75, 83, 132 Tholoniat, L. 120, 288, 292 Toope, S. 47, 273, 322–3, 324 transnational city networks (TCNs) 337–8, 339, 340, 345, 346 transnational corporations, and democracy 92–4 transnational corporations (TNCs) and supply chains 132 transnational governance 11–12, 15–16 transnational governance see business and social norms in transnational governance transnational legal processes 74–6, 81 transparency AI governance 431, 435 economic governance, EU 212–14 iReg Transparency index 124 Transparency International (TI) 278–80 Trauner, F. 105, 254, 262 Trebilcock, M. 72, 76, 77 Trubek, D. 11, 13, 14, 50, 72, 77, 128, 305 Trubek, L. 13, 14, 18, 50, 72, 128, 305 Tsing, A. 74, 77 Tuck, R. 33, 39 Türk, A. 6, 107, 197, 217, 218, 407–22 Twining, W. 75, 170 Uganda, Hussein v Attorney General 356 UK, Northern Ireland Bill of Rights Forum 80 Ukraine war effects, European Commission 262–3 Ulen, T. 58, 68 Ulrich, P. 134, 135, 138 UN Climate Change Conference and Copenhagen Accord 324, 326, 328 UN Committee on the Elimination of Racial Discrimination 356 UN Conference on Environment and Development (UNCED) 276

Index  457 UN Convention on Contracts for the International Sale of Goods (CISG) 149–50 UN Declaration on the Rights of Indigenous Peoples (UNDRIP) 47–8, 358–9, 360, 361 UN Framework Convention on Climate Change (UNFCCC) 63, 277, 291, 324, 326 see also international climate change law UN Global Compact 135 UN High Commissioner of Human Rights 82 UN High Commissioner for Refugees 49, 81 UN Standard Rules on the Equalization of Opportunities for Persons with Disabilities 49 UN Sustainable Development Goals (SDGs) 170, 345, 359, 360 UNIDROIT Principles of International Commercial Contracts (PICC) 37–8 United for Smart Sustainable Cities (U4SSC) 346 Universal Declaration of Human Rights (UDHR) 355, 356 Universal Declaration of the Rights of the Child 49 urban regulation see city regulation Urueña, R. 5, 94, 192, 272–87 US, American restatements 34, 36–7 US administrative guidance 237–52 Administrative Conference of the United States (ACUS) 242–3 Administrative Procedure Act (APA) 238–40, 241, 244 administrative regulations as hard law 239 agencies 238–9, 240, 241, 242–3 bindingness 240, 241, 242, 244, 246 Centers for Disease Control and Prevention (CDC) and COVID-19 237, 245, 248 Deferred Action for Childhood Arrivals policy 243–4 democratic accountability and guidance 243–4 discretionary power 241–2 enforceability 239, 240, 241, 242, 243 Environmental Protection Agency 240 Equal Employment Opportunity Commission (EEOC) 242 equal treatment of regulated parties 241, 242 fair hearing and impartial process focus 245 Food and Drug Administration’s (FDA) 242 guidance controversy 241–4 immigration 243–4, 245, 246–7 internal appeal mechanism 242–3 interpretive and procedural rules 240, 243 judicial review 241 legal framework 238–40 liberty and guidance 241–3

mediating role in coercive relations among private parties 242 moral impact 245–7 moral and political contribution 244–9 notice-and-comment rulemaking process 239–40, 241, 243 organic goods and labelling 245–6 political impact 247–9 private liberty and governmental discretion, conflict between 241–3 procedural rule 240 public participation in administrative rulemaking 238–9, 244–5 safe harbours 242 transgender rights and Dear Colleague Letter 246–7 US Department of Agriculture (USDA) 242, 245–6 US, cases Department of Homeland Security v Regents of theUniversity of California 244 Franklin v Massachusetts 238 Hoctor v U.S. Department of Agriculture 240 James V. Hurston Associates v Glickman 240 National Mining Association v McCarthy 240 Paralyzed Veterans of America v D.C. Arena L.P. 240 United States v Nixon 240 Vallejo, R. 145, 146, 153 Van Asselt, H. 320, 327, 328, 330 Van Dam, C. 6, 194, 197, 254, 259, 262, 370, 391–406 Van den Brink, T. 225, 232, 395 Van Gerven, M. 5, 72, 106, 107, 191, 210, 253, 288–303, 305, 358 Van Gestel, R. 145, 153 Van Leeuwen, B. 147, 186 Vanhercke, B. 107, 290, 291, 292, 293, 294, 295, 296, 297, 397 Vaughan, S. 4, 19, 211, 212, 223–36, 261 Verdun, A. 293, 294, 295, 296, 297, 397 verification markets, and international standards 179 Verschraegen, G. 295, 297 violence, gender-based 379–82 Vogenauer, S. 34, 37, 39 Voigt, C. 326, 329 Voigt, S. 56, 60, 65, 66, 68 voluntary self-regulation, corporate responsibility 135, 136 voluntary status, international standards 178, 179, 180, 183, 184, 185–6 Von Bar, C. 38, 147, 153, 157

458  Research handbook on soft law Von Bogdandy, A. 170, 284, 305, 308 Voorwinden, A. 5–6, 337–52 Wagner, W. 199, 202, 239 Wang, H. 12, 278 Weber, M. 14, 248 Weil, P. 44, 237, 320, 322 Weiler, J. 10, 15, 19 Weimer, M. 12, 17 Wellens, K. 10, 43, 101, 165, 201 Werksman, J. 328, 330 Wessel, R. 44, 105 Whish, R. 47, 254 Wolff, G. 398, 403 World Bank, Identification for Development Initiative (ID4D) 359 World Conservation Strategy (WCS) 277 World Economic Forum 94, 345, 368

WTO Appellate Body EC-Hormones 12 US-Gambling 12 Wu, Q. 10, 16 Xanthoulis, N. 197, 227, 305, 416 Yamin, F. 325, 328 Yates, J. 11, 177, 178, 179 Zeitlin, J. 13, 120, 293, 296, 297, 305, 397 Zelli, F. 88, 92, 94, 327, 328, 330 Zerilli, F. 3, 71–86 Zhelyazkova, A. 254, 305, 308 Zimmermann, R. 34, 35, 38 Zumbansen, P. 10, 16