Private Health Insurance and the European Union [1st ed.] 9783030543549, 9783030543556

Citation preview

Private Health Insurance and the European Union Edited by Cyril Benoît · Marion Del Sol Philippe Martin

Private Health Insurance and the European Union “This book offers a sharp and comprehensive overview of the ongoing creeping privatization of healthcare and how it is shaped by the European Union through the making of a European insurance market. The chapters investigate both the top down and bottom up dynamics of marketization and are thus of interest not only for students of healthcare but also of the Europeanisation of public policy more broadly. In times of global pandemics where the healthcare sector is facing major challenges, this book could not be more timely”. —Amandine Crespy, Associate Professor of Politics Science, Université Libre de Bruxelles, Belgium “This comparative study of private health insurance in Europe shines a welcome light on an intricate and poorly understood topic of great consequence. Private insurance schemes vary widely, with distinctive roles and regulatory frameworks in the member states and at the EU level. In some systems, their operation belies egalitarian promises of the broader health system. Member state and European reforms in the sector rightly awaken fears that private insurance will undermine social systems. This comparative analysis demystifies the complex world of private health insurance regulation while making clear the ways in which private insurance, and even well-intentioned policy reforms, can create inequality and inefficiency in health care”. —Scott Greer, Professor of Management and Health Policy, University of Michigan, USA “Private Health Insurance and the European Union explores the complex dialogue at European and national level between market and public health contrasting logics. This interdisciplinary book, which is a unique collection of theoretical and empirical research papers, shows that choices made about private health insurance greatly influence the nature of the coverage. The book underlines the ambiguity of the interactions between public and private stakeholders and at the same time questions the role of the latter. Researchers, practitioners and students in the fields of social protection, insurance, healthcare and EU-policy making will find invaluable information and elements of analysis in this research project”. —Jean-Philippe Lhernould, Professor of Law, Université de Poitiers, France

Cyril Benoît  •  Marion Del Sol Philippe Martin Editors

Private Health Insurance and the European Union

Editors Cyril Benoît Centre for European Studies and Comparative Politics, Sciences Po (Paris), CNRS Paris, France

Marion Del Sol Intitut de l’Ouest : Droit et Europe Université de Rennes 1 Rennes, France

Philippe Martin Centre for Comparative Labour Law and Social Security Université de Bordeaux, CNRS Pessac, France

ISBN 978-3-030-54354-9    ISBN 978-3-030-54355-6 (eBook) https://doi.org/10.1007/978-3-030-54355-6 © The Editor(s) (if applicable) and The Author(s), under exclusive licence to Springer Nature Switzerland AG 2021 Chapter 1 is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/). For further details see licence information in the chapter. This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Palgrave Macmillan imprint is published by the registered company Springer Nature Switzerland AG. The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Acknowledgements

Funding for this work was provided by the French National Research Agency (Agence Nationale de la Recherche—ANR) as part of the MaRiSa research project (Marché du Risque Santé), Grant No. ANR-17-CE260018. For Chaps. 2 and 7, Coron received additional support from the Erasmus + Programme of the European Union.

v

Contents

1 Introduction: The European Union, the Insurance Industry and the Public-Private Mix in Healthcare  1 Cyril Benoît, Marion Del Sol, and Philippe Martin Part I The Political Roots of EU Insurance Legislation  27 2 Insurance Directives and the Single Market: Towards a Trivialisation of Private Health Insurance? 29 Gaël Coron and Marion Del Sol 3 Solvency II, the European Government of Insurance Industry and Private Health Insurance 55 Cyril Benoît Part II The Impact of the European Union Private Health Insurance  83 4 The Uncertain and Differentiated Impact of EU Law on National (Private) Health Insurance Regulations 85 Philippe Martin and Marion Del Sol

vii

viii 

Contents

5 An Increasing Homogenisation of Private Health Insurers Under Solvency II?129 Philippe Abecassis and Nathalie Coutinet Part III Shifting the Public-Private Mix in Healthcare? Multifaceted Paths Towards Europeanization 161 6 Private Health Insurance in Belgium: Marketization Crowded Out?163 Cyril Benoît and Marion Del Sol 7 Europeanized, Marketized but Still Governed by the State? Private Health Insurance in France191 Gaël Coron, Thomas Houssoy, and Cyril Benoît 8 Ireland: The Ambiguous Role of the Health Insurance Market219 Philippe Martin and Pascale Turquet 9 The Dutch Way: Experimenting with Competition in the Healthcare System247 Pascale Turquet and Philippe Martin 10 In Between the Market and Public Health Insurance: A Place for Occupational Welfare in Europe?281 Thomas Houssoy, Marion Del Sol, and Philippe Martin

List of Contributors

Philippe  Abecassis  CEPN (UMR CNRS 7234), Université Sorbonne Paris Nord, Villetaneuse, France Cyril  Benoît Centre for European Studies and Comparative Politics, Sciences Po (Paris), CNRS, Paris, France Gaël Coron  Arènes (UMR CNRS 6051), EHESP, Rennes, France Nathalie  Coutinet  CEPN (UMR CNRS 7234), Université Sorbonne Paris Nord, Villetaneuse, France Thomas  Houssoy Clersé (UMR CNRS 8019), Université de Lille, Lille, France Philippe  Martin Centre for Comparative Labour Law and Social Security, Université de Bordeaux, CNRS, Pessac, France Marion  Del Sol  Intitut de l’Ouest : Droit et Europe, Université de Rennes 1, Rennes, France Pascale Turquet  LiRIS, Université de Rennes 2, Rennes, France

ix

List of Figures

Fig. 4.1 Fig. 4.2 Fig. 4.3 Fig. 5.1 Fig. 5.2 Fig. 5.3 Fig. 5.4 Fig. 5.5 Fig. 5.6 Fig. 5.7 Fig. 5.8 Fig. 7.1

Overview of the different functions of private health insurance (PHI)86 Organic scope of EU economic law 91 The three legal spaces resulting from the EU competition law 96 OOP payments evolution, 1991–2018 133 Level and evolution of share of public spending on health (1991–2017)135 Relationship between private health insurance coverage and voluntary health insurance (VHI) spending. Data 2011—(e) estimate137 Change in the number of M&A transactions inside and outside the EU (1997–2018) 144 The top three insurers’ market shares by country (CR3) in 2006 and 2011 145 Pattern of evolution of the M&A count in the five most active countries (1997–2018) 151 Deals by type of concentration, in %. Number of transactions in which the country is the acquiror 153 Opening and vulnerability of health insurance markets in Europe 154 The “distribution rule of cost” in the French health insurance system, its functioning and aggregate final distribution between financers 195

xi

xii 

List of Figures

Fig. 10.1 Fig. 10.2

Changes in distribution by level of coverage of beneficiaries between 2006 and 2013 Rate of coverage by complementary health insurance, based on income per consumption unit and the type of coverage in 2012

305 306

List of Tables

Table 2.1 Table 3.1

The “three generations” of EU Insurance directives 30 Weighted average solvency coverage ratios in health by category in four EU countries 75 Table 5.1 The five types of private health insurance markets in Europe 138 Table 5.2 Nature of voluntary health insurance contracts in countries where supplementary voluntary insurance prevails 140 Table 5.3 Number of private health insurers by status (2011) 147 Table 5.4 M&A deals by country (number and percentage: 1997 –2018) 150 Table 5.5 Type of concentration according to the role of private health insurance 156 Table 6.1 Number of affiliates per Sickness Fund in Belgium (2013) 169 Table 6.2 Number of affiliates per Sickness Funds in Belgium (2013–2017)184 Table 7.1 Number of independent firms and groups operating on French PHI market (2017) 206 Table 8.1 Risk equalization premium credits 228 Table 8.2 Community rating stamp duties 228 Table 8.3 Profitability of insurers on the Irish Private Medical Insurance (PMI) market in 2014 229 Table 10.1 Percentage of the population who have reported to have foregone care in 2014, in accordance with their insurance situation304

xiii

CHAPTER 1

Introduction: The European Union, the Insurance Industry and the Public-Private Mix in Healthcare Cyril Benoît, Marion Del Sol, and Philippe Martin

1   Introduction Over the last 20 years, European Union (EU) healthcare policies and their effect on healthcare systems and politics at national level have attracted significant attention from social, legal and political scientists (Anderson 2015, chapter 7; Greer and Kurzer 2016, Mossialos et al. 2010; Steffen 2005; see also Coron 2018). In this domain, treaties are categorical as to

C. Benoît (*) Centre for European Studies and Comparative Politics, Sciences Po (Paris), CNRS, Paris, France e-mail: [email protected] M. Del Sol Intitut de l’Ouest: Droit et Europe, Université de Rennes 1, Rennes, France e-mail: [email protected] P. Martin Centre for Comparative Labour Law and Social Security, Université de Bordeaux, CNRS, Pessac, France e-mail: [email protected] © The Author(s) 2021 C. Benoît et al. (eds.), Private Health Insurance and the European Union, https://doi.org/10.1007/978-3-030-54355-6_1

1

2 

C. BENOÎT ET AL.

the limited competencies of supranational provisions, and Member States ostensibly retain power over a number of crucial aspects of their health policies (Vollaard et  al. 2016). Unsurprisingly, what research has essentially reported in this broad context is a rather gradual development of EU healthcare policies, with more or less perceptible (and often indirect) effects at national level. Due to its frequently unanticipated nature, such development has typically taken circuitous routes and mainly manifested in two ways: first, by the dissemination of standards and guidelines or through the regulation of certain goods or commodities—knowing that the EU has a number of regulatory prerogatives in relation to vital health products, such as pharmaceuticals (Permanand 2006); and, second, through the court’s application of internal market law which, in turn, may affect several segments of health services delivery, policies or rules governing healthcare professionals (Duncan 2002). EU fiscal governance was recently recognized as a “third face” of EU health policy (Greer 2014). Indeed, and after the financial crisis, the EU has gained new powers to enforce budgetary austerity. Through a series of coercive policy recommendations to Member States about the governance of their healthcare systems, fiscal policy became in turn more “rigorous and intimate”, provoking significant shifts in healthcare decision-making at national level (Greer et al. 2016). There is an important segment of the healthcare sector that arguably lies at the crossroads of these three dimensions—in the sense that it is influenced by supranational standards and regulatory provisions, marked by tensions between EU and domestic law and affected by post-crisis regulation—yet it has received very limited attention to date (though see Thomson and Mossialos 2007). It is the private (usually voluntary) health insurance (PHI) industry, a term that refers to the variety of firms offering either or both types of substitutive health coverage (that would otherwise be provided by a national health insurance or system); complementary coverage (for services excluded from or not fully covered by the public purse); or supplementary coverage, which essentially supplies consumers with greater freedom of choice and faster access to care (Thomson and Mossialos 2007). In several EU countries, PHI accounts for a significant share of health expenditure and is even the main provider of care for some benefits. Crucially, it has faced major transformations over the last 30 years, and this as a result of changes in EU law and regulations, often in combination with more or less explicit forms of budgetary constraints or retrenchment efforts at the domestic level.

1  INTRODUCTION: THE EUROPEAN UNION, THE INSURANCE INDUSTRY… 

3

There is an obvious reason why PHI has largely remained under the radar of scholarship on EU healthcare politics. In effect, it is essentially through a series of directives aimed at governing the insurance industry as a whole that EU law has remodelled PHI.  A similar reason arguably accounts for the rather limited interest of students of the public-private mix in healthcare in the study of this issue—a term that we use in the following pages both to describe the combination of public and private actors in health coverage (see Ebbinghaus 2011), but also in health benefits and services delivery (see also Benoît and Coron 2019). As amply documented in the following pages, the potential implications of insurance regulation for the health sector were indeed regularly underestimated or simply disregarded, a statement that applies both to academic research and, critically, to policy-making at domestic level—with policymakers often unaware of the prior effects of EU law and regulation on private health insurers. In turn, this relative ignorance was a source of recurrent mismatches and unexpected effects of policy choices, with a number of implications for the PHI industry. Crucially, this eventually contributed to changing the public-­private mix in healthcare. In an attempt at filling this gap, this collective book gathers a multidisciplinary team of specialists in social policy and the insurance industry. On this basis, our ambition is to provide a broader account of the diverse long-term effects EU provisions had on the private health insurance industry—and their important implications for the political economy of contemporary Welfare States. In this book, we are more formally motivated by three series of research questions. The first relates to the influence of EU law and regulation on the nature and the place of PHI, which contributors intend to prove and to characterize. We are also seeking to determine if and how the effects of these directives and regulations interacted with both the outputs and the outcomes of health policy in several countries (Belgium, France, Ireland and the Netherlands), particularly in a well-documented context of “permanent austerity” (Pierson 1998) and where retrenchment efforts are now firmly entrenched in Welfare State reform (Pierson 2001). Thirdly (and ultimately) our goal is to evaluate if, how and to what extent the interactions between EU law and regulations and health policy at domestic level affected the nature and scope of health coverage in the countries under study. Overall, the story that we narrate here is that of a growing “decoupling” (Trein 2017) between insurance regulation and health policy, entailing a number of side effects for the private health insurance industry,

4 

C. BENOÎT ET AL.

for the structure of the public-private mix in healthcare and for the nature of health coverage, yet with contrasting effects from one country to another. At a more conceptual level, such a broad finding might seem quite trivial to readers familiar with existing scholarship on Europeanization (see Graziano and Vink 2007). What we describe is indeed an umpteenth story in which a series of directives are adopted with the explicit aim of harmonizing domestic rules applying to an industry that eventually results, after a conflictual implementation process, in multiple unintended effects due to its interactions with a variety of national institutional dependencies. There are, however, a number of lessons to draw from this seemingly classical conclusion due to the compound nature of the private health insurance industry. Indeed, our findings echo a number of burgeoning debates related to EU influence on health policy on the one hand and to the role played by private providers in social policy at national level on the other. Together, they delineate future research agendas discussed in greater length below. The rest of this Introduction is organized as follows. In Sect. 2, we start by positioning our research questions and our contribution in the wider literature. We then introduce the main analytical structure permeating the various contributions of the book. Section 3 provides an overview of our research design and case selection process. To understand how the European “matrix” affected PHI and the public-private mix in healthcare, we have undertaken an analysis of its genesis and of its effects at national level, through a comparative analysis of four countries. Section 4 present the contributions to the book, grouped into three parts—the first on the political economy of EU law and regulation related to PHI, the second based on large-n comparisons of the effect of these provisions on PHI and the third part dedicated to in-depth, country case studies. We also reflect in this section on how our findings echo wider debates in the literature.

2   Private Health Insurance in the European Union: Marketization Embraced? 2.1  EU Insurance Law and Regulation as a Potential Vehicle for the Marketization of PHI This book maps the transformation of the private health insurance industry in the EU over a period of around 30 years, starting with the

1  INTRODUCTION: THE EUROPEAN UNION, THE INSURANCE INDUSTRY… 

5

debates that led to the passage of “Insurance” directives in 1992. The choice of this point of departure is justified by the ambition of these texts, with the explicit aim of creating a unified legal framework for insurance activities in Europe, in order to facilitate competition between insurers in an increasingly unified Single Market. European integration is pushed a step further some five years later, with the opening-up of a 12-year process of fierce political battles around the reform of solvency rules governing insurance companies, resulting first in the adoption of two directives in 2002 (known as the “Solvency I” system) and, more significantly, of the Solvency II directive in 2009—which finally came into effect in January 2016. Presented as a response by EU authorities to the financial crisis, this text’s most explicit goal was to set up a riskbased approach to insurance regulation, involving a number of changes in terms of capital requirements, risk management and governance structures of insurance activities. As already suggested by this very brief outline of the European policies considered in greater detail throughout the book, PHI was not the principal target of these different texts. Intriguingly, healthcare as a policy matter was not a salient issue in the debates surrounding the building of such a “European government” (Jullien and Smith 2014) of the insurance industry, especially as compared with other life and nonlife insurance activities. Part of the reason for that situation lies in the relatively marginal status of PHI in the wider insurance industry in Europe—where in most Member States, a significant share of health expenditure is covered by the public purse. Private health insurance companies are thus typically small or medium sized firms. They are also less financialized than most of the insurance industry. Moreover, they also tend to cover a much more limited array of risks than the dominant insurance companies, essentially circumscribed to health coverage and occupational welfare. In turn, and when the largest insurance firms operating across the continent do offer health-­related products of some kind, they rarely constitute the core of these companies’ growth strategy. There is another set of peculiarities of PHI in Europe that explains its rather peripheral position in the industry and, by extension, in regulatory policy agenda. Most private health insurers across the continent are indeed non-profit companies organized along solidarity-based or

6 

C. BENOÎT ET AL.

democratic principles—meaning that they are usually run by a democratically elected board or by employer and employee representatives, and that they generally use their gains to increase the benefits or the coverage provided to their members. There are historical reasons for that: in many Western European countries (particularly in continental Europe), PHI has developed before the formation of contemporary Welfare States, often as emanations of churches, charities, trade unions or employers’ associations. In several countries, various compromises were adopted to integrate these entities after the formation of modern Welfare State institutions, sometimes through the delegation to these entities of all or part of the management of healthcare systems or services, sometimes by providing them with a more minor role. This means that PHI has almost always kept strong ties with or within the Welfare State and by the same token, is usually heavily regulated or even governed at domestic level. As a consequence, the public-­private mix in healthcare in Western Europe tends to be integrated into fixed and rigid regulatory frameworks. EU insurance law and regulation hits this balance. As of 1992, Insurance directives opened up the PHI market to competition, notably to for-profit insurance companies. More importantly, a number of provisions maintaining PHI close to public healthcare systems were considered selective advantages and gradually removed. With Solvency II, private health insurers now have to comply with stricter solvency requirements. They also need to make additional financial provisions and find ways to increase the value of their funds—which, in a broader context of low interest rates, often means searching for further (potentially riskier) diversification benefits (Standard and Poor’s 2016). Additionally, this text conveys a more explicit alignment of their governance with forprofit insurance companies and, more broadly, financial firms. In summary, EU laws and regulations, at least formally, seem to induce a growing marketization of PHI—with the openness of the health insurance market to new actors, posing a number of threats for the business model of non-profit entities, with an intensification of (possibly pricebased) competition and by potentially paving the way for the increased financialization of the sector. In this context, tracing the political sources of these formal provisions and determining whether they effectively translated into an actual marketization of private health insurers is the guiding thread of the different

1  INTRODUCTION: THE EUROPEAN UNION, THE INSURANCE INDUSTRY… 

7

contributions of this volume. Symmetrically, the notion of marketization permeates the three research questions addressed in the following chapters. In determining whether EU law and regulation changed the nature of PHI and if so, how, we are primarily interested in elucidating whether it was in the sense of increased marketization and, critically, with which measurable effects on the industry. By looking at whether and how these same provisions interacted with health policy at domestic level, our motivation is predominantly to evaluate whether it introduced more market logics, behaviours and strategies in the public-private mix in healthcare. The answer to our third question, on how these interactions eventually (re)shaped the nature and scope of health coverage is also driven by a similar concern—do the principles that originally governed the private side of health coverage now resemble more those of a market, as is the case in numerous countries outside Europe? 2.2  Marketization: A Multidisciplinary Approach for a Multifaceted Concept The growth of market forces in the Welfare State or social welfare institutions has been the subject of endless amounts of research over the last three decades. More recently, this topic was examined in closer connection to European integration (Crespy 2017). This book largely builds upon these various contributions. In particular, we draw from these the statement according to which marketization might come with a diversity of institutional features and that it can perform widely different (and sometimes contradictory) functions. This particularly applies to our case, where the provisions of EU insurance law and regulation are capable of generating different marketization paths as they interact with institutional orders stabilized at domestic levels. This claim also rests on well-established findings reported by the literature on the forging of markets in the Welfare State per se. In an authoritative contribution in the field, Gingrich (2011) identified six ideal types of markets in the Welfare State. Her approach combines the allocation dimension (is the responsibility for access collective or individual?) and the production dimension (who—namely the state, users or producers—has effective control over the market?). The multiple configurations between these two dimensions are capable of generating significant variations in market types, ranging from “austerity markets” (where there is individual

8 

C. BENOÎT ET AL.

responsibility for access and State control, as in Dutch healthcare markets) to “pork barrel markets” with collective responsibility for access and where producers are in capacity to extract rents (typified in Gingrich’s study by the elderly care market in England in the 1980s). A related feature of marketization is that it is fundamentally multifaceted. It can operate at the level of formal provisions while not necessarily translating into effective practices of firms and organizations. Similarly, there can be marketization when some actors endorse market logics in their strategies and behaviours or when these same logics penetrate their cognitive frames—yet, and crucially, this can be a result (or not) of legal or more explicit political impulses. There may be obvious attempts from political actors to marketize public or private entities’ way of operating by changing the institutional rules governing their activities; but for various reasons, an actual market might not emerge at the end as older practices prove enduring (see also Schelkle 2019). These rather familiar statements call for an analytical disambiguation of the different stages and the different processes through which marketization is liable to occur and to draw appropriate inferences from these various possible marketization processes. This motivates the multidisciplinary approach retained in this book, which brings together economists, legal and political scientists. Accordingly, each of our four case studies are drafted by at least two authors from different disciplinary backgrounds, who are addressing our research questions by considering the plural forms marketization often takes in the country under study. This includes its most obvious, formal and policy appearances—typically when there is a number of legal provisions that generate the conditions for the increasing marketization of the private health insurance industry. But marketization is also measured through the lens of the economic strategies and behaviours deployed by the actors exposed to it, principally private health insurers in different countries. Marketization is also envisioned as a possible unexpected outcome of the interaction between EU law and regulation and health policy at domestic level. As such, we also assume that it can be an outcome without necessarily having been a purpose of public policy—as we will see, this particularly applies to significant portions of the interaction between the legal matrix established through EU insurance law and regulation and health policy at the domestic level. While most studies on marketization have focused on the transformation of Welfare State institutions, this book uses this concept to study a more atypical configuration, since we are essentially considering broad

1  INTRODUCTION: THE EUROPEAN UNION, THE INSURANCE INDUSTRY… 

9

patterns of marketization through the lens of what happens within the private health insurance industry. In the literature, marketization has indeed often been discussed by authors primarily concerned with the privatization of the Welfare State—with the former phenomenon often seen as eventually leading to the latter, typically through an increase in public spending paralleled with discrete support for market solutions (Jensen 2011). In effect, marketization was often regarded as a particular case of “hidden politics of [Welfare State] retrenchment” (Hacker 2004). Here, we report patterns that differ from these classical accounts for obvious reasons, as the health insurance industry considered throughout the book is private. Yet, and as already suggested, it is not necessarily marketized due to its non-profit commitment, the institutional setting in which it operates and the principles governing its activities. This does not mean that the transformations of this industry never interacted with policies explicitly aimed at privatizing or introducing market mechanisms within the Welfare State. Moreover, and as repeatedly shown in the book, the effects of the marketizations of the private health insurance industry are often amplified by retrenchment efforts at domestic level. Nevertheless, by retaining this case, our ambition with this book is also to depart from the canonical debates around privatization that have dominated a significant body of scholarship on the Welfare State. In so doing, we intend to reflect more explicitly upon the actual practices and logics that lie behind public and private providers in order to appreciate their overall impact on the organization and division of solidarities and the associated power relations within the public-private mix in healthcare. 2.3  Making the Case for the Contingency of Europeanization and its Usages If EU insurance law and regulation is likely to marketize the private health insurance industry, and possibly a share of the public-private mix in healthcare, by which mechanisms is it likely to do so? In brief, what this book describes is the development of a sizeable legal and regulatory structure for the insurance sector that gradually institutionalizes at domestic level, generating a set of transformative effects affecting the private health insurance industry. In this way, this overall process entails numerous unprecedented interactions with domestic healthcare reform—which, in turn, affects both the public-private mix in healthcare and the nature and scope of health coverage.

10 

C. BENOÎT ET AL.

We were not, though, expecting to observe automatic effects at play or institutions mechanically colliding with each other here, for two main series of reasons. One is associated with the case under scrutiny. Despite its magnitude and density, EU law and regulation in this domain strikes a richly textured environment, moulded by several decades of reforms that have given the public-private mix in healthcare its current shape in the different countries analysed in this volume. Stated differently, there are a number of interest groups, veto points and more broadly, beneficiaries of past policy choices or compromises that are more than likely to intervene in the course of this implementation process (including private health insurers themselves) and that are more than likely, too, to deflect EU policies from their initial purpose. A second, deeper reason relates to the very process by which EU law and regulation is transposed and implemented. It is indeed widely acknowledged that implementation always involves different sets of actors which, through their mobilizations, interpretations and usages, shape the contours of this overall process—meaning that the same policies or legal provisions can be constructed and used in many different ways (Hay 2007) from one country to another, or even within a single country. This fundamentally implies that transposition always involves “translation” in the broadest sense of the term (Smith 1997). For our research design, a wider implication of embracing such conception is that we expect the outcomes of these transformations under study to be highly contingent, a statement that equally applies to our three research questions. Stated differently and more generally, we assume here that EU insurance law and regulation should be regarded as creating a number of constraints and opportunities for a large range of actors that are in a position to use tools and resources (possibly offered by European integration itself) to politically shape both the implementation and the likely effects of EU policies—in summary, there is always a range of possible (and possibly multiple) “usages” of Europe in each situation under study (Graziano et al. 2011). Accordingly, it is through a focus on (individual and collective) actors’ political work that the different contributors to this volume grasp and draw inferences from the transformations of the private health industry and the interactions between EU and domestic levels—in particular by paying special attention to the legal, financial, cognitive, political or institutional resources these same actors may possess and actually use to shape this overall process (Jacquot 2008). This is not to suggest that actors and

1  INTRODUCTION: THE EUROPEAN UNION, THE INSURANCE INDUSTRY… 

11

actions are favoured over institutional factors by the different contributors (see also Exadaktylos and Radaelli 2012 on this distinction). On the contrary, actor-related and institution-related variables are studied in their intimate connection, according to the basic assumption that institutional rules and norms are effective when they actually appear in practices. Institutional or legal factors are thus considered and reported when they are observable and are actual prescriptions for behaviour, and their relative effect is appreciated through the lens of their practices (Itçaina et al. 2016).

3   Research Design and Case Selection 3.1  The Structure of the Book To approach the transformation of the private health insurance industry under the effect of EU law and regulation—and ultimately, its implications for the public-private mix in healthcare and health coverage—we have organized this book in three parts. Each part corresponds to a specific analytical level and addresses our research questions from a particular angle. In Part I, two chapters narrate the political genesis of the two sets of European directives that directly affected PHI, namely “Insurance” directives in 1992, then “Solvency I” and “Solvency II” directives in 2009. These chapters focus on the participants and the various motivations of the individual and collective actors involved in the development of this far-reaching European architecture. As such, the two chapters describe the landscape of the government of the insurance industry in Europe. They do so through the lens of the political work undertaken by the various actors that shaped it, with special reference to the regulatory politics at play, and coupled with broadest reflections on the governance of financial services in the EU (see Quaglia 2010). Part II is also formed of two chapters that consist of large-n studies of the impact of this matrix on private health insurers per se. In 1992, Insurance directives essentially consisted of changes in legal and formal provisions that modified the status and the governance of private health insurers, particularly vis-à-vis the rest of (essentially public) health coverage providers. Thus, the first chapter of this part uses legal methods to appreciate the differentiated impact of EU law in a large set of European countries and healthcare systems, describing a sizeable array of possible (and actual) combinations of public and private provisions. By contrast,

12 

C. BENOÎT ET AL.

Solvency II mostly concerned a change of the prudential regime governing the daily activities of private health insurers, with likely impacts on the scale of their business model, their strategies and corporate governance. The second chapter of this part thus involves quantitative analyses of the transformations that have affected the private health insurance industry over the recent period. Part III consists of four in-depth country case studies allowing for the comparison of the different trajectories followed by Belgium, France, Ireland and the Netherlands in this domain. These chapters more frontally address the question of the collision of the EU framework with national (mostly health or social policy-related) political agendas. As such, these chapters not only examine the potential marketization of the private health insurance industry from the perspective of firms, but also from the vantage point of the public-private mix in healthcare. The fifth and last chapter of Part III adopts a more prospective stance, reflecting on the relation between private health insurance, occupational welfare and the distribution of solidarities in the public-private mix in healthcare, also on the basis of national experiences in these domains. 3.2  A Multimethod Perspective As suggested by this brief outline, the different chapters use a variety of quantitative and qualitative methodologies drawn from the three disciplines participating in the book (economics, law and political sciences). All chapters are formed of a similar empirical bedrock obtained through documentary analysis, systematic content analysis of press releases (including the professional press of the insurance industry), an extensive review of public and organized interests’ positions in each policy debate under consideration and when applicable, preliminary expert interviews. This first empirical stage is completed with additional evidence generated through more specific methodologies, depending on the scope and the specific questions posed in each chapter. Chapters 2 and 3 (first part) mapping the development of EU law and regulation add to this basis a range of semi-structured interviews with key participants in these processes, notably civil servants, staff of European administrative services and directorates, as well as ministerial and European Commission officials. They also rely on second-hand academic sources, mostly on the basis of an extensive review of articles published in actuarial

1  INTRODUCTION: THE EUROPEAN UNION, THE INSURANCE INDUSTRY… 

13

journals where a variety of stakeholders expressed opinions and shared alternative propositions as these processes were unfolding. In Part II, Chap. 4 assesses the differentiated impact of EU law on private health insurance and its regulation on the basis of a systematic and exhaustive analysis of EU and national legal databases, which roughly consisted of an examination of the national transposition of European directives and the application of European Court of Justice (ECJ) rulings. Chapter 5, on the impact of Solvency II on the private health insurance industry, mobilizes a range of time series analysis techniques based on an original dataset including 23 countries, for a timeframe between 1997 and 2018. It is completed with the identification of specific patterns for the more active countries of the sample, estimated through regression models. Part III (Chaps. 6, 7, 8, 9 and 10) combines different methodologies, in line with our ambition to identify the diverse forms that marketization might (or might not) take in different countries, as well as to grasp from various angles the broader consequences of the coevolution of insurance law and regulation and health policy. As such, all chapters are informed by an exhaustive review of the legislation produced in the country in question, descriptive statistics and semi-structured interviews with a diverse range of actors, including civil servants and administrative officials, private health insurers and their representatives as well as healthcare professionals. Using data from interviews with representatives of key organizations or players has become quite common in social policy analysis and in political economy, notably for case-oriented studies involving several countries (Ebbinghaus and Naumann 2018). In this part, the combination of different methodologies and data sources more fundamentally allows for a better description of the sequencing of policy decisions and policy change, a crucial point when it comes to evaluating the effects of EU law and regulation on an industry heavily institutionalized at national level. 3.3  Case Selection While the first part of the book deals with the different pieces of the European regulatory architecture, the second part involves large-n comparisons of its legal and economic impacts on PHI. By contrast, in Part III we address our research questions (especially related to the public-private mix in healthcare and the nature and scope of health coverage) through a limited number of in-depth country case studies. The ultimate goal of these chapters is to characterize causal mechanisms, yet not to identify a

14 

C. BENOÎT ET AL.

single scheme of linear causality. Instead, our ambition is to uncover traces of causal mechanisms within the context of the cases under study (Bennett and Elman 2006), by providing for each case “an accurate picture of causal constellations” (Roger 2013) between a range of intervening variables (such as the prior effects of past healthcare reforms, the mobilization of key veto players or interest groups and the ties that private health insurers might possess in political parties—each providing grounds for various possible “usages of Europe”). As discussed before, we are expecting the outcomes for each case to be highly contingent upon the political work of a number of actors evolving in compact institutional settings. We are also expecting the EU matrix to be subjected to a variety of potential interpretations and framings. As such, we are adopting throughout the book a rather classical approach in healthcare Europeanization, where the goal of the analysis is to understand “how, why and to what extent” this process has taken a given form in the cases under study, assuming that a lot of national specificities are at play from one country case to another (Martinsen 2012). Our three research questions involve a number of underlying assumptions that do matter when it comes to case selection. Determining whether and how EU law and regulation changed the private health insurance industry obviously requires retaining cases where this industry exists and where it accounts for a significant (which does not necessarily mean large) share of health expenditures. In addition, measuring how the outcomes of this matrix interacted with prior or simultaneous effects of past policy choices is better achieved, we think, through selecting cases where there are some explicit and institutionalized linkages between the public and the private sides of health coverage. Arguably, this same statement equally applies to our third research question, namely whether and how health coverage changed in nature or scope as a result of the interactions between the European framework and health policy at domestic level. Moreover, an overall expectation underlies these three research questions. We are indeed hypothesising that the institutionalization of the EU framework results in a growing marketization of private health insurers, of the public-private mix in healthcare and, eventually, health coverage. This requires selecting cases where the private health insurance industry is potentially exposed to such a transformation, at least formally—and symmetrically, not selecting cases where it is already marketized. It is noticeable that cases where there exists a fully marketized private health insurance industry are quite rare in Europe. Granted, there are some countries where

1  INTRODUCTION: THE EUROPEAN UNION, THE INSURANCE INDUSTRY… 

15

there exists a (rather small) market for private health insurance. Yet it generally has no particular institutional linkages with public coverage (this is, e.g., the case in Spain). In accordance with our research questions and hypotheses, we have therefore decided not to include such cases in our country case analysis (see further). On the basis of these statements, we have retained for our analysis most-similar cases (Seawright and Gerring 2008), to compare the relative effects of factors that we expect to be present in the different cases under study, but crucially, interacting and being at play differently. We combine this approach with a most-likely case selection strategy (Rohlfing 2012), namely by selecting national cases which are most likely to be impacted by EU law and regulation. Here, our ambition is to maximize our chances of observing the PHI marketization process in order to be better able to identify a wide set of different mechanisms—knowing that we expect this process to be an input of EU law and regulation, despite its interaction with other institutional features of healthcare systems that are not directly affected by this initial source of change. It is on this broad basis that we have retained the four countries under study (Belgium, France, Ireland and the Netherlands), which all comply with this twofold requirement (see Chap. 5 for an extensive discussion). In each of these countries, the share of health expenditure covered by PHI is significant (above 12%). These are also countries where “governments recognize that [PHI] can contribute significantly to social protection” (Mossialos and Thomson 2004) and thus where there exist strong ties between PHI and the Welfare State. Lastly, the private health insurance industry is marked by the presence of a large number of non-profit firms in these four countries, and thus particularly exposed to different forms of marketization. Crucially, there are also for-profit insurance companies offering health coverage in each of these cases, thus allowing an appreciation of the effect of EU law and regulation on these entities, and their potential conflict-provoking effects on “institutionalized relationships” (Jullien and Smith 2008) within the industry. It is also worth mentioning that if these four cases are both most-similar and most-likely from the angle of our research questions and assumptions, they are not usually considered as such by comparative research on Welfare States or health policy. If Belgium, France and the Netherlands are all Bismarckian systems to a certain extent, this is not the case of Ireland, classified as a Beveridgian (tax-financed) system. In terms of health expenses, one can also note a significant decrease in the share of health expenditure

16 

C. BENOÎT ET AL.

covered by the public purse in Ireland and the Netherlands during the period examined, while this share remained relatively stable in Belgium and France. There are, in addition, a number of differences in terms of the status, governance and organization of non-profit health insurers per se from one selected country to another. We expect these second-order differences to provide meaningful insights into the mechanisms accounting for the penetration and effects of EU law and regulation in this domain.

4   The Book in a Nutshell There are two ways of seeing the story narrated in this book and to envision the answers it provides to our research questions. Readers primarily interested in health policy will find in the subsequent chapters a discussion of the interactions of two trends (marketization and privatization) that arguably stand at the forefront of the debates related to this field, yet that have been subjected to only scant discussion in conjunction with EU influence. Those who are mostly interested in the insurance industry and more broadly, financial services regulation, might also find an interest in reading this volume. Scholars are now paying increasing attention to this sector of prime importance for our understanding of the political economy of contemporary capitalism. Still, the literature has to date mainly focused on large transnational firms and on life insurance, arguably at the forefront of the changes affecting the wider industry (see Graz 2019). Here they will find an analysis of its transformations from a different standpoint, allowing for an appreciation of some of the many side effects associated with the regulatory and industrial changes it has experienced in Europe over the last 30 years. After introducing the different contributions to this book, we return to these questions in a discussion of the wider implications of our findings for current debates on these two broad classes of issues. 4.1  Contributions to the Book The book opens with Gaël Coron and Marion Del Sol’s analysis of the long march towards a unified European government for insurance activities, starting off with the adoption of “first generation” (1973) and culminating with “third generation” Insurance directives in 1992—the latter having constituted a critical juncture for the private health insurance industry in most countries (Chap. 2). Coron and Del Sol particularly insist in this chapter on the proactive role played in this domain by European

1  INTRODUCTION: THE EUROPEAN UNION, THE INSURANCE INDUSTRY… 

17

Commission services in charge of financial institutions and corporate law, who have made an intensive use of ECJ rulings to advance their own visions and agendas for the sector. More generally, what he describes is a process leading to a twofold trivialization. On the one hand, and as a result of the Commission officials’ political work, insurance activities were growingly assimilated to financial services and treated as such, leading to various implications for their regulation in the Single Market. This overall trend was paralleled with a more discrete yet significant trivialization of private (mostly non-profit) health insurance companies, gradually associated with the rest of the insurance activities—notably through Insurance directives in 1992. In Chap. 3, Cyril Benoît examines the fierce political battles that led to the far-reaching reform of solvency rules governing insurance activities in the EU between 1994 and 2016, with special reference to the private health insurance industry. Focusing on the prudential regime laid down in EU law and regulation, he shows that the adoption of Solvency II in 2009 amplified the movement described by Coron, albeit through a renewed approach. While previous directives were essentially preoccupied with finding common rules to organize competition within the Single Market, Solvency II developed as a more conceptual architecture, with rules, requirements and standards aimed at becoming ingrained in insurers’ daily activities. This shift was essentially legitimized as providing greater transparency and safety to financial investors and policyholders. Benoît shows that this transformation is fundamentally more likely to induce major changes for private health insurers (notably non-profit and smaller firms), as their activities have developed using different principles, values and ruling structures. After a focus in Part I on the sizeable European regulatory infrastructure that developed over the last three decades, Part II starts with an extensive, large-n study of the differentiated implications it may have on PHI in various Member States (Chap. 4). In this regard, Marion Del Sol and Philippe Martin, with a particular focus on the competitive framework created by Insurance directives, arrive at a twofold conclusion. First, EU law and regulation might effectively be considered as a potential vehicle for the increasing marketization of the sector, particularly in those countries where private health insurers are non-profit firms and closely integrated into Welfare State institutions. Yet, and crucially, it should be symmetrically regarded as providing a matrix rather than being a source of vertical integration—as it is actually flexible and open to interpretation,

18 

C. BENOÎT ET AL.

and as it leaves a wide latitude for national regulation of the PHI industry, including its relation with public providers. As such, one can expect a high degree of contingency and many possible usages of this legal framework in its transposition at domestic level. In their study of the impact of Solvency II, Philippe Abecassis and Nathalie Coutinet draw a similar conclusion from the vantage point of the insurance industry per se (Chap. 5). It is right, they assert, to expect that EU law and regulation result in an increasing homogenization of private health insurers, in the sense of their growing alignment with the behaviours, strategies and logics already in play in significant segments of the sector. Yet, the pattern suggested by their data is more complex than what this broad-brush picture suggests. Indeed, the financialization of PHI (and notably, mergers and takeovers) does not coincide with the business cycle of the rest of the industry. This does not mean that the impact of EU law and regulation was neutral. Insurance directives provided strong incentives for PHI to develop transnational activities; Solvency II, as it involves stricter solvency requirements, incentivized larger dominant firms to diversify their risk portfolios, leading to alliances with PHI or to the development of health-related products—as the latter are less demanding in terms of solvency requirements. However, this matrix, Abecassis and Coutinet argue, is better understood as providing a basis for various strategies than as an exogenous shock implying a standardised response. For the time period considered, it even appears that national reforms played a more decisive role in providing constraints and opportunities to PHI. These converging statements are followed in Part III by in-depth country case studies. Cyril Benoît and Marion Del Sol (Chap. 6) first consider the case of Belgium, where the “public” side of health coverage is delegated and organized around non-profit private health insurers, namely mutual benefit societies—and as such is excluded from the perimeter of EU insurance law and regulation. Over the last two decades, and as a result of various governmental attempts at reducing health expenditure, mutual benefit societies nevertheless developed and managed on their own a variety of complementary coverage, initially without any formalized legal boundaries. This situation was challenged during the 2000s by for-­ profit insurance companies seeking to penetrate the market. In this context, they used both Insurance and Solvency II directives in their search for supranational support, in order ultimately to challenge the position of mutual benefit societies. In turn, the latter responded by working politically to secure their position at domestic level. As a result of these political

1  INTRODUCTION: THE EUROPEAN UNION, THE INSURANCE INDUSTRY… 

19

struggles, a reform adopted in 2010 reinforced several features of the Belgian public-private mix by safeguarding the position of mutual benefit societies for complementary coverage. But this same reform also opened the supplementary side of health coverage to competition and aligned it with EU provisions, thus marketizing a share of the public-­private mix in Belgium—with recent figures suggesting that this new pillar is now rapidly expanding. Mutual benefit societies might also be found in France, as explained by Gaël Coron, Thomas Houssoy and Cyril Benoît in their contribution (Chap. 7). Historically, however, their activities were circumscribed to the complementary share of health coverage, where they have to compete with other non-profit entities and (increasingly) for-profit insurance companies. What has happened in the country, Coron, Houssoy and Benoît argue, is an early and manifold marketization process of private health insurers as a result of the application of EU law and regulation—but decisively, this trend was markedly reinforced and shaped by a series of policies adopted at national level. Indeed, over the last 20 years, successive French governments have tried to increase health coverage without expending the share already covered by the public purse. This strategy ostensibly involved private health insurers in achieving several governmental objectives, yet the prior effects of Europeanization on these entities were poorly acknowledged by policymakers. As such, the many consequences associated with the rise of a “European-driven” market now increasingly conflict with a “State-driven” market. While Europeanization was associated with greater marketization of the public-private mix in healthcare in Belgium and France, this has initially not been the case in Ireland, as argued by Pascale Turquet and Philippe Martin in their contribution (Chap. 8). In this “two-tiered” system, private health insurance duplicates public coverage by offering additional benefits and services such as access to private hospitals. While there is competition on the PHI market in Ireland in application of EU legal provisions, the Irish government succeeded in maintaining strong regulation of the sector due to the important role it plays in the provision of care. This was notably achieved through an exemption of (semi-public) non-­ profit firms from compliance with Insurance directives and through a risk-­ equalization scheme. However, this initial compromise was increasingly contested in the 2000s, with new entrants on the market trying to use European provisions to challenge the position of their competitors. This also coincided with debates on whether Voluntary Health Insurance

20 

C. BENOÎT ET AL.

(VHI) Healthcare, the non-profit leading firm of Irish PHI market, should comply with solvency requirements. These turbulences were amplified by the consequences of the financial crisis, which significantly raised the cost of insurance policies over the last few years. Turquet and Martin then turn to an examination of the case of The Netherlands, a country where health policy has been market-oriented since the beginning of the 1990s. The system is now characterized by managed competition between health insurers, and the government only retains regulatory and supervisory prerogatives (Chap. 9). In this country, EU law and regulation was thus not the only—or the main—source of marketization. In addition, there seems to be no contradiction between the latter’s most market-like dimensions since similar features are arguably in play in the Dutch healthcare system. However, and crucially, Turquet and Martin show that it does not necessarily mean that Europeanization has had no effect on the country. Solvency rules on insurance activities as well as the degree of openness of the market (in close relation to the policies pursued by the Dutch Central Bank) notably constituted major issues, with collateral implications for the public-private mix in the course of recent years. In the last contribution to the book, Thomas Houssoy, Marion Del Sol and Philippe Martin adopt a more prospective stance by reflecting on some additional repercussions of the transformations observed in the previous chapters (Chap. 10). In the four countries under study, PHI is now more marketized than it was at the beginning of the 1990s, even if the shape and the very reasons for the development of health insurance markets differ from one country to another. What are the specific implications of such changes when PHI turns out to be acquired at corporate level, and thus becomes part of occupational welfare? More precisely, how do firms and trade unions behave in such marketized environments when they become purchasers of private health coverage? What are the related implications of their choices for the distribution of solidarities within the public-­ private mix? After a discussion of the many ramifications of these questions (which ostensibly resonate with the literature on pension reform and the pillarization of social protection), Del Sol, Martin and Houssoy consider the French case, viewing it as a quasi-natural experiment. Indeed, while once limited in scope, corporate PHI grew significantly in this country after a reform in 2013, which introduced an obligation for employers to provide their workers with a PHI scheme. Crucially, it came into force in

1  INTRODUCTION: THE EUROPEAN UNION, THE INSURANCE INDUSTRY… 

21

an environment already reshaped in the sense of a greater marketization of PHI by prior reforms. Each of these chapters comes with its own conclusions and series of responses to our research questions. Additionally, we think that the book in itself generates further insights on two broader debates in the literature that are worth briefly introducing before allowing the reader to enter into the many complexities of the relationship between private health insurance and the European Union. 4.2  Understanding the Presence of Financial Firms in Social Policy The various conclusions raised in the aforementioned chapters explicitly echo a wider controversy in the political economy literature. Political economists have indeed long debated whether and why capital—often equated with employers—tended to express support for or oppose social protection (see Hall and Soskice 2001). More recently, scholars have argued that our understanding of “capital” should also include financial firms, since as potential competitors of social insurance, they can reasonably “be expected to be key proponents of retrenchment” to become integral participants of social coverage (Naczyk 2013). The book provides this burgeoning literature with additional accounts on why financial firms— here insurance companies—might have an interest in prospering in social protection or policy, particularly in a more protected and less lucrative market, as health insurance undoubtedly is. Admittedly, and by paving the way for a removal of some of the barriers that integrated PHI into the Welfare State, EU provisions considerably eased access to this market to other (essentially for-profit) insurance companies. One can note that their presence increased in the four countries for the period under study, but crucially, this was not as a result of an explicit (or an implicit) European political agenda—as health insurance, as shown in Chap. 2, was a theme of particularly little importance during the debates surrounding Insurance directives. In a similar vein (and more fundamentally) several chapters of the book argue that this also has to do with the incentives created by Solvency II, while the goal of this text was initially to strengthen regulation and improve regulatory standards. What is described in the following pages is that by imposing stricter requirements, EU provisions created a very powerful incentive for for-profit insurance companies to diversify, notably on more secured markets such as healthcare (see

22 

C. BENOÎT ET AL.

Chaps. 2, 3 and 4). As such, the presence of insurance companies (and their demands, as exemplified by the case of Belgium, for government to facilitate their access to the health insurance market—see Chap. 6) also appear to be a side effect of an overall increase in regulatory requirements. Put differently, an apparent consolidation of regulatory demands for the wider industry here resulted, at least in some of its segments, in increasing pressures and financialization. This does not necessarily contradict the finding reported by studies identifying more explicit attempts by financial companies to gain access to social protection, and country case studies in Part III provide evidence of more direct and more classical forms of interest group politics. But this arguably provides a set of additional factors of the formation of the structure of opportunity that shape financial firms’ behaviours and motivations in this domain—as we highlight how social protection might be unintentionally affected by wider transformations of financial services regulation. A similar statement arguably applies to related debates on the financialization of social protection. Here also, we report, especially in Chaps. 3, 4 and 7, an increasing financialization of health insurers, notably non-profit insurance companies. And here again, we show how it is an outcome (and to some extent a side effect) of the stricter solvency requirements demanded by Solvency II. In our view, such findings should invite political economists to pay greater attention to multilevel changes and unexpected outcomes of past regulatory choices when they seek to understand why and how there is a rise of financial firms or financialization in social protection. 4.3  The EU and the Shifts in the Public-Private Mix in Healthcare As already suggested at the beginning of this Introduction, the literature on EU healthcare politics has essentially focused on the propensity of Europeanization to diffuse a number of (notably regulatory) standards, to alter health services delivery and policies through the application of internal law and to stimulate budgetary austerity—three features largely exhibited in the case of PHI. This book hopefully helps to make another aspect more explicit, which complements rather than disputes the already known faces of EU healthcare politics as documented in the literature. This aspect relates to how Europeanization contributes to the reshaping of the public-­ private mix in healthcare in two ways, amply discussed in Chaps. 6, 7, 8 and 9. Firstly, it can introduce competition within the sector (or facilitate

1  INTRODUCTION: THE EUROPEAN UNION, THE INSURANCE INDUSTRY… 

23

competition where it already exists), together with the alignment of non-­ profit and for-profit providers. Secondly (and less noticeably), because Europeanization might result in an institutional decoupling between PHI and Welfare State institutions. This, in turn, renders public and private coverage more sealed off from each other. In addition to modifying the nature of health coverage, such reconfiguring might also affect the scope of health policy (as illustrated by the case of France studied in Chap. 7). As a result, Europeanization here is not necessarily “shifting the public-­ private mix” (Seeleib-Kaiser et al. 2012), but rather reinforcing its institutional (and possibly financial) sedimentation. This latter statement will certainly sound familiar to those who know the literature dedicated to the political economy of pension reform and policies, where scholars are accustomed to speaking in terms of pillarization and multi-pillar structures—and where they are accustomed, too, to studying marketization in close connection with privatization in broad “multi-pillar” settings (see Ebbinghaus 2015). There are a number of issues involved when the public-­private mix in healthcare is analysed in the language of pillars, as discussed in Chap. 10. But we strongly think that there is something to be gained from keeping a watchful eye over some similar transformations arguably at play between the two sectors, in close connection with the role of the EU and, more broadly, Europeanization. We also believe that this holds true even if, in stark contrast with pension reform, the changes that are described here are mostly unintended outcomes of the interactions between various EU and national policy choices, and are not (or not yet) part of an explicit political agenda to combine pillarization with various forms of privatization and marketization of the public-private mix in healthcare. Yet some critical outcomes (that prominently include marketization, a segmentation of coverage and growing presence of financial firms) are now changing both sectors, which are arguably at the crossroads of national “growth regimes” (Hassel and Palier 2020).

References Anderson, K. (2015). Social Policy in the European Union. Basingstoke: Palgrave. Bennett, A., & Elman, C. (2006). Qualitative Research: Recent Developments in Case Study Methods. Annual Review of Political Science, 9, 455–476. Benoît, C., & Coron, G. (2019). Private Health Insurance in France: Between Europeanization and Collectivization. The Milbank Quarterly, 97(4), 1108–1150.

24 

C. BENOÎT ET AL.

Coron, G. (Ed.). (2018). L’Europe de la santé: enjeux et pratiques des politiques publiques. Rennes: Presses de l’EHESP. Crespy, A. (2017). Welfare Markets in Europe: The Democratic Challenge of European Integration. Basingstoke: Palgrave. Duncan, B. (2002). Health Policy in the European Union: How It’s Made and How to Influence It. British Medical Journal, 324(7344), 1027–1030. Ebbinghaus, B. (Ed.). (2011). The Varieties of Pension Governance: Pension Privatization in Europe. Oxford: Oxford University Press. Ebbinghaus, B. (2015). The Privatization and Marketization of Pensions in Europe: A Double Transformation Facing the Crisis. European Policy Analysis, 1(1), 56–73. Ebbinghaus, B., & Naumann, E. (2018). Introduction: Analysing Organised Interests and Public Opinion Towards Welfare Reforms. In B.  Ebbinghaus & E. Naumann (Eds.), Welfare State Reforms Seen from Below: Comparing Public Attitudes and Organized Interests in Britain and Germany. Basingstoke: Palgrave. Exadaktylos, T., & Radaelli, C. (Eds.). (2012). Research Design in European Studies: Establishing Causality in Europeanization. Basingstoke: Palgrave. Gingrich, J. (2011). Making Markets in the Welfare State: The Politics of Varying Market Reforms. New York: Cambridge University Press. Graz, J.-C. (2019). The Power of Standards: Hybrid Authority and the Globalisation of Services. Cambridge: Cambridge University Press. Graziano, P., Jacquot, S., & Palier, B. (Eds.). (2011). The EU and the Domestic Politics of Welfare State Reforms. Basingstoke: Palgrave. Graziano, P., & Vink, M. (Eds.). (2007). Europeanization: New Research Agendas. Basingstoke: Palgrave. Greer, S. (2014). The Three Faces of European Union Health Policy: Policy, Markets and Austerity. Policy and Society, 33(1), 13–24. Greer, S., Jarman, H., & Baeten, R. (2016). The New Political Economy of Healthcare in the European Union: The Impact of Fiscal Governance. International Journal of Health Services, 46(2), 262–282. Greer, S., & Kurzer, P. (2016). European Union Public Health Policy: Regional and Global Trends. New York: Routledge. Hacker, J. (2004). Privatizing Risk without Privatizing the Welfare State: The Hidden Politics of Social Policy Retrenchment in the United States. American Political Science Review, 98(2), 243–260. Hall, P., & Soskice, D. (Eds.). (2001). Varieties of Capitalism: The Institutional Foundation of Comparative Advantage. Oxford University Press: Oxford. Hassel, A., & Palier, B. (Eds.). (2020). Growth and Welfare Reforms in Global Capitalism: How Growth Regimes Evolve. Oxford University Press: Oxford. Hay, C. (2007). Constructivist Institutionalism. In R.  A. Rhodes, S.  Binder, & S.  Rockman (Eds.), The Oxford Handbook of Political Institutions. Oxford: Oxford University Press.

1  INTRODUCTION: THE EUROPEAN UNION, THE INSURANCE INDUSTRY… 

25

Itçaina, X., Roger, A., & Smith, A. (2016). Varietals of Capitalism: A Political Economy of the Changing Wine Industry. Ithaca: Cornell University Press. Jacquot, S. (2008). National Welfare State Reforms and the Question of Europeanization: From Impact to Usages. REC-WP 01/2008 Working Papers, Edinburgh. Jensen, C. (2011). Marketization via Compensation: Healthcare and the Politics of the Right in Advanced Industrialized Nations. British Journal of Political Science, 41(4), 907–926. Jullien, B., & Smith, A. (Eds.). (2008). Industries and Globalization: The Political Causality of Difference. Basingstoke: Palgrave. Jullien, B., & Smith, A. (Eds.). (2014). The EU Government of Industries: Markets, Institutions and Politics. London: Routledge. Martinsen, D. (2012). The Europeanization of Healthcare: Processes and Factors. In T. Exadaktylos & C. Radaelli (Eds.), Research Design in European Studies: Establishing Causality in Europeanization. Basingstoke: Palgrave. Mossialos, E., Permanand, G., Baeten, R., & Hervey, T. (Eds.). (2010). Health Systems Governance in Europe: The Role of EU Law and Policy. Cambridge University Press: Cambridge. Mossialos, E., & Thomson, S. (Eds.). (2004). Voluntary Health Insurance in the European Union. Brussels: World Health Organization and European Observatory on Health Systems and Policies. Naczyk, M. (2013). Agents of Privatization? Business Group and the Rise of Pension Funds in Continental Europe. Socio-Economic Review, 11(3), 441–469. Permanand, G. (2006). EU Pharmaceutical Regulation: The Politics of Policy-­ Making. Manchester: Manchester University Press. Pierson, P. (1998). Irresistible Forces, Immovable Objects: Post-industrial Welfare States Confront Permanent Austerity. Journal of European public Policy, 5(4), 539–560. Pierson, P. (2001). Coping with Permanent Austerity: Welfare State Restructuring in Affluent Democracies. In P.  Pierson (Ed.), The New Politics of the Welfare State. Oxford: Oxford University Press. Quaglia, L. (2010). Governing Financial Services in the European Union. London: Routledge. Roger, A. (2013). What’s in a Comparison? Some Remarks about the Analysis of Recurrent Processes. In J. Iglesias, N. Stojanovic, & S. Weinblum (Eds.), New Nation-States and National Minorities in the 21st Century. Essex: ECPR Press. Rohlfing, I. (2012). Case Studies and Causal Inference: An Integrative Framework. Basingstoke: Palgrave. Schelkle, W. (2019). EU Pension Policy and Financialization: Purpose without Power? Journal of European Public Policy, 26(4), 599–616.

26 

C. BENOÎT ET AL.

Seawright, J., & Gerring, J. (2008). Case Selection Techniques in Case Study Research: A Menu for Qualitative and Quantitative Options. Political Research Quarterly, 61(2), 294–308. Seeleib-Kaiser, M., Saunders, A., & Naczyk, M. (2012). Shifting the Public-Private Mix. In P. Emmenegger, S. Hausermann, B. Palier, & M. Seeleb-Kaiser (Eds.), The Age of Dualization: The Changing Face of Inequality in Deindustrializing Societies. Oxford: Oxford University Press. Smith, A. (1997). Studying Multi-Level Governance: Examples from French Translations of the Structural Funds. Public Administration, 75(4), 711–729. Standard and Poor’s. (2016, January 5). French Mutual Insurers are Feeling the Many Forces of Consolidation. S&P Ratings Direct. Steffen, M. (Ed.). (2005). The Governance of Health in Europe. London: Routledge. Thomson, S., & Mossialos, E. (2007). EU Law and Regulation of Private Health Insurance. Health Economics, Policy and Law, 2(2), 117–124. Trein, P. (2017). A New Way to Compare Horizontal Connections of Policy Sectors: ‘Coupling’ of Actors, Institutions and Policies. Journal of Comparative Policy Analysis, 19(5), 419–434. Vollaard, H., van de Bovenkamp, H., & Martinsen, S. (2016). The Making of a European Healthcare Union: A Federalist Perspective. Journal of European Public Policy, 23(2), 157–176.

Open Access  This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/ by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made. The images or other third party material in this chapter are included in the chapter’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

PART I

The Political Roots of EU Insurance Legislation

CHAPTER 2

Insurance Directives and the Single Market: Towards a Trivialisation of Private Health Insurance? Gaël Coron and Marion Del Sol

1   Introduction The health insurance market has an economic reality in all Member States of the European Union (EU), though the place and role of the private insurance sector in covering health care expenditure varies, sometimes considerably, from one country to another (Sagan and Thomson 2016). The health insurance market is also a legal reality: the various private health insurance actors pursue their activity within a common legal environment that was forged, from 1973 onwards, by the “insurance” directives adopted by the EU.

G. Coron (*) Arènes (UMR CNRS 6051), EHESP, Rennes, France e-mail: [email protected] M. Del Sol Intitut de l’Ouest : Droit et Europe, Université de Rennes 1, Rennes, France e-mail: [email protected] © The Author(s) 2021 C. Benoît et al. (eds.), Private Health Insurance and the European Union, https://doi.org/10.1007/978-3-030-54355-6_2

29

30 

G. CORON AND M. DEL SOL

This normative environment is the fruit of a long political process seeking to create a single insurance market. The common framework was constructed gradually via the adoption of three generations of directives. The first two sets of directives sought to allow insurance actors to exercise two fundamental economic freedoms laid down by the Treaty: freedom of establishment and freedom to provide services. The third generation of directives adopted in 1992 brought the process to completion and, in a way, formally established the internal insurance market (Table 2.1). To a great extent, these texts were due to the activism of the European Commission, more specifically, to that of the Directorate-General in charge, at the time, of the portfolio “Financial Institutions and Company Law” (hereinafter DG 15).1 From the outset, it should be highlighted that this Directorate-General held significant power and wielded considerable political sway, which are both essential levers for moving European projects forward (Smith 2016; Georgakakis and Rowell 2013; Greer et  al. 2019). For this particular project, DG 15 placed its actions within the European Commission’s transverse single market project. With regard to services, the three generations of directives contributed towards this goal. Consequently, the political horizon of the “insurance” directives was not sector-specific; it was part of the project to ensure the free movement of financial services within the European area. This would prove to be decisive in the progress of the political process; as it would be for the provisions adopted, these giving little consideration to the specificities of Table 2.1  The “three generations” of EU Insurance directives

Non-life insurance (including health insurance) Life insurance

First generation— freedom of establishment

Second generation— freedom to provide services

Third generation— completion of the internal market

Directive 73/239/ EEC of 24 July 1973 Directive 79/267/ EEC of 5 March 1979

Directive 88/357/ EEC of 22 June 1988 Directive 90/619/ EEC of 8 November 1990

Directive 92/49/EEC of 18 June 1992 Directive 92/96/EEC of 10 November 1992

Source: author(s) 1  In later reforms, this Directorate-General would be included within the scope of the DG Market, in charge of the internal market.

2  INSURANCE DIRECTIVES AND THE SINGLE MARKET… 

31

the insurance sector. In other words, this initial context configured to a great extent the course of the European framework: a kind of trivialisation of the activity and the institutions of the insurance sector, which was subsumed into the far greater financial services sector (2). When examining the insurance sector and, notably, the third generation of directives (1992), the matter of taking into consideration intrasectoral specificities inevitably arises. These directives had a blanket scope of application that only distinguished between two large classes of insurance on the basis of a technical criterion:2 life insurance, which includes all types of insurance for which risk is linked to the duration of the insured person’s life (insurance payable on death or survival); and non-life insurance, which includes all types of insurance that are not linked to the person’s life (property insurance, liability insurance, health insurance, etc.). Given that these texts set out to forge a common framework of practice for all insurers, the matter of intrasectoral adaptations arose in the course of the political process. This was the case for health insurance, notably owing to national specificities (e.g. the existence of mutual benefit societies in France);

Method

The results presented in this chapter are derived from four sources: –– a review of the legal literature concerning the three generations of directives; –– a series of semi-structured interviews conducted in 2018 with former members of the European Commission who were involved in drawing up the directives and with former managers from the private health insurance sector; –– private archives collected by a former civil servant of the Commission, which allowed us to access the speeches presenting the second- and third-generation texts to various bodies and also the Commission’s preparatory documents; –– concerning the French situation, the minutes of executive committee meetings and the general assembly of the Fédération Nationale de la Mutualité Française (the leading federation of French health mutual benefit societies, hereinafter FNMF).

2  Several classifications exist for insurance. The directives adopted the classification based on the modalities of managing premiums, by capitalisation (life insurance) or by-distribution (non-life insurance).

32 

G. CORON AND M. DEL SOL

nevertheless, the issue never assumed a central role, to the extent that one can once again speak of trivialisation in another form, that of the actors and activities of the health insurance sector (3).

2   The Political Horizon of the “Insurance” Directives: Realisation of the Internal Market One might reasonably expect that the “insurance” directives were drawn up in response to the specific concerns of the insurance sector and the needs expressed by representatives of this industry and/or end consumers. The reality is however very different. The normative framework created was a component of the transverse project to realise the single market. Furthermore, these directives were the outcome of a historical process in which the availability of technical solutions to institutional actors was primarily determined by their political feasibility. The process is detailed below. The directives also demonstrated the adaptability of the European Commission (Jabko 2006), a strategic actor, and also the decisive role played by the Court of Justice in constructing the cognitive frameworks, which underpin political projects (Weiler 1994). 2.1  First Attempts to Construct a European Insurance Market on the Basis of Freedom of Establishment For the construction of the European insurance market, three legal bases could be mobilised. They corresponded to three fundamental economic freedoms recognised by the Treaties: –– freedom of establishment, by virtue of which an economic actor (here, an insurer) registered in one Member State can access national (insurance) markets via the creation of a durable establishment in the territory of one or more other Member States. Non-national actors must not be discriminated against; they must be able to access national markets under the same conditions as local economic actors. –– freedom to provide services, by virtue of which an economic actor can pursue cross-border activities without needing to have a durable or permanent establishment in the territory of another State. In other words, the freedom to provide services allows an insurer registered in one Member State to access the insurance market of another

2  INSURANCE DIRECTIVES AND THE SINGLE MARKET… 

33

State without being required to have a local establishment. The State regulating the market cannot impose as a prerequisite the existence of an establishment in its territory. –– free movement of capital, which enables the implementation of an integrated financial system. This freedom permits the movement of funds necessary for insurance financial products and links policies in the insurance field with the legislation applicable to financial services and instruments (e.g. mutual funds). The diversity of these legal bases explains a large part of the political game that preceded the adoption of legislation for insurance at the European level (Dehousse 1998). The legal techniques for implementing a unified, then single, market, changed over time as a result of a complex game between States and European institutions. The life insurance/non-­ life insurance dichotomy adopted by the first-generation directives is a perfect illustration of this political game. The first generation of directives (1973 and 1979) employed freedom of establishment as their legal basis. They sought to coordinate the conditions of access and of exercise that insurers had to fulfil in order to establish themselves in another Member State, whether as a primary undertaking (registration of headquarters) or a secondary undertaking (creation of agencies or branches). Supervisory mechanisms and conditions varied, sometimes quite substantially, between Member States. For example, authorisation was required in some States but not in others; some States issued broad authorisations, whereas others issued special authorisations (for specific types of insurance activity) (De Fruto Gomez 1991). The notion of supervision could also vary between countries, notably in terms of financial requirements. With the stated aim of facilitating freedom of establishment, the directives sought to eliminate areas of divergence between national legislations with regard to supervision. Particular attention was given to coordinating the provisions concerning financial guarantees; securing the undertakings made by insurers is an essential means of protecting insured parties. The question was even more essential in view of the fact that the first directives systematised administrative authorisation as a prerequisite and entrusted Member States with the competence of granting authorisation to insurers who wanted to become established in their territory whether as a primary or secondary undertaking.

34 

G. CORON AND M. DEL SOL

From this period, the texts incorporated requirements related to technical provisions, that is, reserves that had to be sufficiently high to fully settle all contractual undertakings towards insured parties. They corresponded to financial assets appearing in the liabilities balance sheet and that allowed the insurer to cover the cost of claims. Hence they represent a debt to insured parties. The first-generation directives made a sufficient level of technical provisions a prerequisite for obtaining administrative authorisation in all Member States. This represented a process of coordination but also of convergence of national legislations. However, convergence was minimal because the texts granted States the power to determine the amount of technical provisions required for obtaining authorisation. The importance that the first directives attached to financial guarantees had another concrete manifestation. In addition to having sufficient technical provisions, the insurer was required to have a sufficient solvency margin, that is, holdings free of predictable undertakings, with incorporeal (own funds) elements deducted. The sufficient solvency margin covered all of the insurer’s activities and had to be determined in accordance with methods laid down by the directives. In this regard, the distinction between non-life insurance (1973 directive) and life insurance (1979 directive) raised the issue of simultaneous pursuit (accumulation) of activities in both categories. This led the European legislator to establish, in the 1979 directive, the specialisation principle which, in essence, is a principle of non-accumulation of a life insurance activity with a non-life insurance activity. The specialisation principle “gate-crashed” the political process. As Guy Levie, principal administrator of the “Insurance” division of the European Commission, explained in an article published in 1980, the continuation of “multi-activity” businesses was an important issue in the discussions that culminated in the 1979 directive. The United Kingdom, Belgium and Luxembourg lobbied the Commission to permit multi-activity, whereas the other countries wanted pluriactivity abolished within fifteen to twenty years. (Levie 1980)

The provisions of the final text represented a kind of compromise between these two positions: on the one hand, article 13(3) allowed insurers who accumulated both categories before the announcement of the directive to continue to pursue both activities but on the condition that they adopted independent management (with separate technical provisions and solvency margins for each activity); on the other hand, article

2  INSURANCE DIRECTIVES AND THE SINGLE MARKET… 

35

13(6) gave States the option to end accumulation situations that predated the directive over a period that they determined. Application of the first-generation directives had contrasting effects. Officials of the European Commission have frequently considered that national authorities did not fully apply the principle of equal treatment to national and non-national insurers. It seemed that the authorisation procedures also provided a mechanism for States to control access to their insurance market by granting no or few authorisations to non-national insurers to establish themselves in their territory. In this respect, the weakness of the chosen legal basis (freedom of establishment) for constructing a European direct insurance market was revealed: it left open the possibility of national resistance.3 2.2  Revival of the Political Process via Mobilisation of Case Law Two events contributed to revival of the process for integration of the insurance sector. The first was not specific to the field and occurred somewhat in the background: publication of the Commission’s White Paper Completing the Internal Market in 1985. The second event did concern the insurance sector, more specifically the activity of brokers; it consisted of a series of rulings by the Court of Justice in 1986. These two events acted upon the consequences of another essential event in the construction of Europe that occurred in 1979: the Rewe-Zentral ruling also known as the “Cassis de Dijon” ruling.4 The window of opportunity opened by the Court with the Cassis de Dijon ruling. The case concerned the marketing of alcohol produced in one Member State in the territory of another Member State, which therefore engages the principle of free movement of goods. The Court considered that “in the absence of common rules relating to the production and marketing of alcohol […], it is for the Member States to regulate all matters relating to the production and marketing of alcohol in their territory” (point 8). In other words, with exceptions limited to those enumerated, if 3  In contrast, it is noteworthy that, in the field of reinsurance, the principle of freedom to provide services was established in 1964 (Council Directive 64/225/EEC of 25 February 1964 to eliminate, in reinsurance and retrocession, restrictions on the freedom of establishment and the freedom to provide services). 4  ECJ 20 February 1979, Rewe-Zentral AG, Case 120/78, ECR 1979, p. 649.

36 

G. CORON AND M. DEL SOL

a good is produced in compliance with the legislation of one country, it should be able to move throughout the territory of the single market. The only exceptions were “mandatory requirements relating in particular to the effectiveness of fiscal supervision, the protection of public health, the fairness of commercial transactions and the defence of the consumer” (point 8). The Court also—and this is the crucial point—dismissed the argument advanced by the German government that the European alcohol market did not exist in the absence of European regulations; at the time of the ruling, the regulation intended to create and regulate this market was blocked at the Council. The following year, the Commission issued a Communication interpreting the ruling in order to extend its effects. It considered that “any product imported from another Member State must in principle be admitted to the territory of the importing Member State if it has been lawfully produced, that is, conforms to rules and processes of manufacture that are customarily and traditionally accepted in the exporting country” (Commission 1980). The Commission thereby reinforced the “mutual recognition” principle and extended it beyond the market implicated by the Cassis de Dijon ruling. In doing so, it implemented a method for overcoming national resistance. Incidentally, the Commission announced its programme of action at the end of the Communication: “The Commission will therefore have to tackle a whole body of commercial rules which lay down that products manufactured and marketed in one Member State must fulfil technical or qualitative conditions in order to be admitted to the market of another […]”. The “mutual recognition” principle also holds the promise of limiting the proliferation of supranational regulations, while preventing the creation of vast quantities of national regulations. In effect, mutual recognition makes economic integration possible without the need to replace national regulations with transnational regulations and without the need to set up supranational political institutions (Streeck 1995). The background goal: an internal market for financial services. From the 1980s, this case law would be deployed by the European Commission and the Court of Justice to remove obstacles to integration. And it would be deployed by these actors in a different field: services. As Molinier wrote: “the Court of Justice transposed the Cassis de Dijon case law, concerning the free movement of goods, to people and services” [author’s translation from French] (Molinier 1995, p. 142).

2  INSURANCE DIRECTIVES AND THE SINGLE MARKET… 

37

In its 1985 White Paper, the Commission outlined its strategy in favour of completing the internal market in general and for financial services in particular (COM/85/0130 final). This document is considered to be representative of the Delors method—named after the President of the European Commission at the time—both in the way it was drawn up and in its content (Drake 2002). In the White Paper, the Commission explained the approach it intended to adopt with regard to financial services and clearly established a link with the Cassis de Dijon ruling: The liberalisation of financial services, linked to that of capital movements, will represent a major step towards Community financial integration and the widening of the Internal Market. The accent is now put increasingly on the free circulation of “financial products”, made ever easier by developments of technology. Some comparison can be made between the approach followed by the Commission after the “Cassis de Dijon” judgements with regard to industrial and agricultural products and what now has to be done for insurance policies […]. The Commission considers that it should be possible to facilitate the exchange of such “financial products” at a Community level, using a minimal coordination of rules (especially on such matters as authorisation, financial supervision and reorganisation, winding up, etc.) as the basis for mutual recognition by Member States of what each does to safeguard the interests of the public.

What is interesting to note is that the argument advanced about safeguarding insured people (one that would be deployed in official speeches as the primary means of legitimising future texts) is paradoxically linked to reciprocal (or mutual) recognition, the express aim of which is to reduce state supervision, which could potentially reduce consumer protection. Turning point: Commission versus Federal Republic of Germany ruling concerning the insurance sector. While the Cassis de Dijon case law and the White Paper formed the background to the revival of the process to create an internal insurance market, a ruling within the insurance field would have a major impact; specifically, the Commission/Federal Republic of Germany ruling by the Court of Justice.5 The matter originated in infringement proceedings brought by the Commission against the German State for failure to fulfil its obligations under the Treaties. Specifically—and this is a vital point—the infringement proceedings were brought by DG 15. It 5  ECJ 4 December 1986, Commission of the European Communities versus Federal Republic of Germany, Case 205/84, ECR 1986, p. 3755.

38 

G. CORON AND M. DEL SOL

was not a decision endorsed by the College of Commissioners. In other words, DG 15 did not have to negotiate with the other Directorates-­ General, who could have opposed it (Scharpf 2000). In this matter, the Commission built its case on a broad interpretation of the Cassis de Dijon case law to request that the Court sentence the German State for maintaining in its national law a provision that: –– required that insurance undertakings of the Community wanting to provide direct insurance services in Germany, through intermediaries, representatives or agents, be established and have authorisation; –– prohibited insurance intermediaries established in the Federal Republic of Germany from offering residents insurance contracts with insurers established in another Member State. In parallel, the Commission was confronting France, Denmark and Ireland on similar grounds. These countries intervened during the proceedings, as did the Belgian and Italian governments, in support of the German position. The Commission received the support of Great Britain and the Netherlands and its position was reinforced by the Court’s ruling in 1986 that the requirement for establishment was in breach of Community law: It must be stated that the requirements in question in these proceedings, namely that an insurer who is established in another Member State, authorised by the supervisory authority of that State and subject to the supervision of that authority, must have a permanent establishment within the territory of the State in which the service is provided and that he must obtain a separate authorisation from the supervisory authority of that State, constitute restrictions on the freedom to provide services inasmuch as they increase the cost of such services in the State in which they are provided, in particular where the insurer conducts business in that State only occasionally. It follows that those requirements may be regarded as compatible with Articles 59 and 60 of the EEC Treaty. [Concerning the freedom to provide services and that became article 56 et seq of TFEU]6 6  Nevertheless and in compliance with previous case law, the Court acknowledges infringements to the freedom to provide services on condition that tests of legitimacy and proportionality are satisfied: on the one hand, restrictions on the freedom to provide services must be on the grounds of a public interest that is not safeguarded by the rules of the State in which the establishment is located; on the other hand, the limitations on the freedom to

2  INSURANCE DIRECTIVES AND THE SINGLE MARKET… 

39

The remarkable aspect of the decision is the substitution of the legal basis by the Court of Justice. This would enable the Commission to revive the political project aiming to create a single insurance market by transforming the political game around the directives. The decision’s grounding in the freedom to provide services rather than the freedom of establishment paved the way for a European insurer to prospect directly in any Member State. In parallel, the authorisation procedures that States had been using to discriminate between national and European insurers became irrelevant and hence ineffective. The political game upended. This case law changed the terms of the political game. In the process leading to adoption of the 1988 and 1990 directives, the Commission exploited synergies with the case law of the Court of Justice. In using this phrase rather than “transposed case law”, we want to emphasise that the Commission ably extracted from the Court’s rulings the tools and political strategies necessary to achieve its goals. It maximised the effects of the Cassis de Dijon case law. More specifically for insurance, it systematised the solutions of the ruling of 4 December 1986 for the case concerning insurance brokers in which it confronted Germany. Its political strategy consisted of confronting States with the negative consequences of the construction of the single market in the absence of harmonisation by the political route. The States would have to endure the activities of insurers of other Member States on their territory, while their own insurers would not be able to export their services; they would not have participated in the construction of the regulatory framework because the case law route was favoured. Incidentally, the strategic use of case law in this manner by the Commission corresponds to the process that Fritz Scharpf refers to when he speaks about the “constitutionalisation” of competition law (i.e. the legal value of competition rules being upgraded by the Court of Justice) (Scharpf 2000). Representatives of States negotiated the future texts with a sword of Damocles hanging over their head: if they did not reach agreement on market conditions via the legislative route, these would be created by the Court’s “pretorian” case law (Dehousse 1998).7 The legal pressure explains to a large extent why States stopped provide services must be strictly limited to what is necessary to protect the public interest evoked. Moreover, the limitations have to apply to all operators. 7  Nevertheless, in view of the volume of extra-territorial transactions at the time, a “Nimzowitsch effect” is apparent (the threat is stronger than the execution) because the volume of transnational insurance services remains very low.

40 

G. CORON AND M. DEL SOL

blocking the adoption of the second- and third-generation texts at the Council. A technically complex realisation. Drawing support from the Treaty articles proclaiming the freedom to provide services, the Commission launched the second series of directives that came to fruition in 1988 (non-life insurance) and 1990 (life insurance). The stated goal was to allow an undertaking established in the territory of one Member State (State of origin) to exercise its activities in the territory of another Member State without having an establishment there, and without facing discrimination favouring established undertakings (active freedom to provide services). And reciprocally, every insurance-seeker would be allowed to freely choose an insurer in the common market (passive freedom to provide services). The two texts, of 1988 and 1990, were very careful with regard to the powers of States. So, of the 33 articles making up the 1988 directive, 21 include phrases such as “a Member State may, by way of derogation [to the present directive]”, “on the conditions laid down by national law”, “this provision shall not affect the right of Member States” and “each Member State may” (Berr 1988). Given that the texts were laying the foundations for the freedom to provide services, they had to provide guidance on the level of supervision and on the division of competences between the State of the insurer (home country control) and the State of the buyer (purchaser’s country control). Furthermore, the directives made several distinctions: between large risks and mass risks;8 between active freedom to provide services and passive freedom to provide services; between home country control and host country control. In fine, this gave rise to a complex legal landscape for the freedom to provide services. In addition, the entire landscape became fragmented between the scheme by way of establishment (first generation) and the scheme by way of freedom to provide services (second generation). The announcement, from the outset, that this complex situation was transient—the third series of directives having already been launched—certainly rendered it more acceptable.

8  Large risks correspond to situations where the insured party exercises in a professional capacity, an industrial or commercial activity or as a self-employed professional, and the risks associated with this activity. These most often concern businesses. Mass risks are those covering non-professional situations and exclusively concern private individuals.

2  INSURANCE DIRECTIVES AND THE SINGLE MARKET… 

41

2.3  Finalisation of Construction of the European Insurance Market by Adoption of the “Passport” Technique Shifting alliances. The path to the third-generation directives engendered a specific configuration for reforms characterised by an alliance between the European Commission and several liberal States and by incorporation into the overall project to complete the single market. Despite this, the first two generations led to complexification of the rules, to the extent that, to paraphrase Paul Pierson, we could speak of dependence on detours (Pierson 2000). This describes the fact that the circumstantial solutions followed in the first- and second-generation directives had a bearing on the content of the new texts. Hence, the goal of the new texts was to a great extent to introduce some simplicity by bringing the different legal schemes closer together. Although the second-generation directives did not have a significant legal effect, they did greatly transform the alliances and the balance of power between States in favour of liberalisation and those against it. Two blocs formed from among the 12 (at the time) Member States: “liberal” States, primarily the UK and the Netherlands, supported the Commission’s projects whereas other States, notably at the beginning France and Germany, were clearly opposed to deregulation; the remaining States were less involved in the negotiations. During the process that culminated in the second generation of directives, the position of the representatives of France evolved: France switched sides, making adoption of the 2nd and 3rd directives rather easy. The French government did not have any problems with the 3rd directives. The French government was pro-3rd directives. […]. The problem […] those who truly resisted right to the end, was Germany. [author’s translation from French] (interview with a former Seconded National Expert (SNE), DG 15, European Commission)

Afterwards, the main actor to slow down the adoption of the texts was actually Parliament. The blockages were less to do with principled opposition to the idea of a single market than to the reinforcement of Parliament’s power by the treaties and Parliament’s desire to brandish its new competence: And there were never-ending debates on the 3rd directives but because Parliament wanted to affirm its role. But it was a debate of no substance.

42 

G. CORON AND M. DEL SOL

[author’s translation from French] (interview with former SNE, DG 15, European Commission)

The European passport, a common model of regulation for the financial services sector. In substance, the third-generation directives (1992) simplified and unified the schemes via a simple principle: a single authorisation issued by the home country on the basis of criteria laid down by the European texts. This authorisation would serve as a European passport. It would allow any insurance undertaking established or having obtained authorisation in one Member State to market its insurance products and services—via agencies or branches or by way of freedom to provide services—in the national territory of all other Member States without any prerequisites. Henceforth, when a new branch or subsidiary was established in a Member State, the only obligation was a simple formal notification to the competent authority of the home country. In other words, the passport obtained from the competent national authority of the Member State where the head office of the organisation was located was valid before other Member States (mutual recognition); and for two modalities: in the context of freedom of establishment (whereby an authorised insurance undertaking can open a branch in any other member country without having to apply to that country’s authority for authorisation); and in the context of freedom to provide services. The single-passport scheme ended the system of authorisation for establishment given that there was no longer a need to apply for authorisation in the host country before creating an establishment or providing services there. It also ended the distinction between active freedom to provide services and passive freedom to provide services, because insurance-­ seekers could choose an insurer from any Member State. The European passport provided a relatively simple mechanism enabling European insurers to pursue their activities within the single market. The regulatory model adopted was light and its characteristics were borrowed from techniques deployed in the field of financial services (e.g. banking licence). As noted by the Secretary-General of the European Committee on insurance in 1991, “the distinction between freedom of establishment and freedom to provide services, which persists in law, is likely however to diminish in practice if agencies and branches can be used as commercial sites for activities pursued by way of freedom to provide services; undertakings with a large network of sites would find themselves at a considerable relative advantage by being able to provide a local service, which is

2  INSURANCE DIRECTIVES AND THE SINGLE MARKET… 

43

essential in insurance, especially non-life insurance” [authors’ translation from French] (Lohéak 1991). In other words, the scheme put in place would favour the most powerful actors and, in the long term, consolidation of actors (see Chap. 5). Groups with several subsidiaries would have the means to take advantage of the “cultural aspect” of insurance contracts by adapting the content of the contracts to the customs of the buyer’s country. The corollary of liberalisation. The corollary of the European passport was the harmonisation of rules used by the home country to grant authorisation. These rules were essentially a (re)affirmation of prudential and technical obligations enacted at the European level. With regard to the calculation of technical provisions, the texts were based on directive 91/674/EEC that harmonised national provisions relating to the constitution of technical provisions imposed on insurers, which made mutual recognition of provisions possible. The directives of 1992 supplemented the text of 1991 with some general actuarial principles: fixing of a prudent interest rate; compatibility of the method of evaluation of mathematical provisions with those for the corresponding assets; and coherence when taking into account shares in future profits. In short, the third generation of directives created a harmonised legal environment for the pursuit of insurance activities. It represented fulfilment of the initial goal of constructing a single insurance market, realised after a long and sometimes chaotic process characterised by a high level of activism by the European Commission and its strategic use of the Court of Justice’s rulings. With the directives of 1992, national markets were opened up to all authorised operators via use of a legal technique (single passport) employed in the financial services sector. This represents the trivialisation of the specificities of the insurance sector.

3   Trivialisation of Health Insurance and its Actors in the “Insurance” Directives For the purposes of realising the internal insurance market, legislation governing the conditions of exercise of insurance activities was highly integrated. Conversely, insurance contract law essentially remained untouched by the harmonisation process. Similarly, the “insurance” directives devoted very little to the specific problems of some branches of insurance. Such is the case for health insurance, which received little debate despite the fact

44 

G. CORON AND M. DEL SOL

that national specificities and differences are numerous and sometimes substantial. It is worth highlighting an attempt by DG 15 to challenge the obligation to join a basic scheme and two initiatives by Member States (one by Germany and one by France). 3.1  The Failed Attempt to End the Monopoly Held by Social Security A maximalist project emerged during the debates around the third-­ generation “insurance” directives. Led by the European Commissioner responsible for the issue, Leon Brittan, it aimed to take advantage of the new texts to end the monopoly held by compulsory health care insurance organisations. In order to understand the foundations of this project, it is useful to consider its instigator. In 1989, Leon Brittan, a former minister of Margaret Thatcher, was appointed European Commissioner in charge of two important Directorates-General: DG 4 “Competition policy” and DG 15 “Financial institutions and company law”. Brittan had considerable political power and levers in influential administrations. It is also worth highlighting Brittan’s political ambition to give full rein to European competition laws, notably against monopoly situations (Smith 2015). Finally, it is necessary to mention the method used. In his endeavours, Brittan systematically drew support from the Court of Justice’s case law; in a way, this case law served as a legal anchor to ground the political project he led. Invoking competition law in the process of realising the single market became a legal mechanism enabling DG 15 to politically justify its incursion in domains within the competence of Member States. Social security is one such domain. In this field, national systems can impose compulsory affiliation to a public scheme having a monopoly. Leon Brittan’s plan to end the Social Security monopoly in health care coverage was part of a broader plan to realise the European insurance market of which private health insurance was but one branch. In passing, it is of interest to note that Brittan’s team was pursuing the same goal in the field of pensions. In this respect, mention should be made of a preparatory document drawn up by DG 15 about realising the internal market in the field of private pensions (European Commission 1990, document no. 15/224/90). And also particularly noteworthy are some of the leanings that emerge from the first draft of the directive concerning the freedom of management and investment of funds collected by pension

2  INSURANCE DIRECTIVES AND THE SINGLE MARKET… 

45

institutions (COM(91)301 final). This draft adopted a wide view of the freedom to provide services and could have led to a challenge to the obligation to affiliate in certain specific cases, such as the by-distribution supplementary schemes that existed in France (Coron 2006). But both these projects were abandoned: Brittan, who therefore is very liberal, wanted to insert an opt-in/out clause in the directives, in relation to social security. […]. Such a clause is—well, he realised that it wouldn’t get very far. [Author’s translation from French] (interview, former SNE, DG 15)

We do not have full data on the reasons for this abandonment. It may be assumed that the political acceptability of the text necessitated courting France, which had made consensus for the third-generation directives possible when it switched sides (see above). One may also mention the fact that similar proceedings were brought before the Court of Justice in 1991 (Coron 2006) and gave rise to the Poucet and Pistre rulings of 17 February 1993.9 This case concerned two artisans who were obliged by French regulations to affiliate with an insurer Assurances Générales de France (AGF) in the context of a compulsory health care insurance scheme set up for self-employed people. Without bringing into question the principle of compulsory affiliation, the two demanded the right to be able to approach any private insurer established in the European territory; implicitly, they were claiming that AGF enjoyed a dominant position contrary to competition law. The decision was issued two years later, that is, after Leon Brittan’s project had been abandoned, but in the meantime the Commission’s legal department had issued its conclusions on the matter, dismissing the plaintiffs’ argument due to the lack of autonomy of the managing insurer owing to the legal framework of the scheme (Coron 2006). The members of DG 15 could have believed that the legal context was not conducive to their initiative and that the Court of Justice would decide in the same direction as the legal department. In this regard, statistical data constructed by Sweet and Caporaso between 1970 and 1992 showed that the Court followed the Commission’s legal department in 88% of cases (Stone Sweet and Caporaso 1998).

9  ECJ joined cases C-159/91 and C-160/91, Poucet and Pistre, 17 February 1993, ECR 1993, p. I-637.

46 

G. CORON AND M. DEL SOL

3.2  Refusal to Adopt the Exclusivity Principle for the Health Insurance Activity Another bone of contention concerning health insurance emerged during the process of adopting the directives. It came not from the European Commission but from a Member State: Germany. The root of the conflict was the specific configuration of the German health insurance system, specifically the existence of a mechanism for substitute health insurance by virtue of which a segment of the population (those who have revenues above a certain threshold) has the option of choosing private health insurance to substitute for the “compulsory” public cover (Busse et al. 2017). The conflict was centred on the provision of German legislation prohibiting insurers in the substitute health insurance market from pursuing any other activity (principle of exclusive activity). For example, the health activity is the only one pursued by the group Deutsche Krankenversicherung (DKV)  (translated as “German health insurance”), which maintained its “specialisation” principle when it expanded abroad, notably by founding its subsidiary DKV Belgium in Brussels in 1964.10 In 1973, the first directive focused on non-life insurance highlighted this particular situation. Its article 7 ((2) c) granted, in principle, temporary permission to Germany to continue to prohibit in its territory accumulation of health insurance with another branch of insurance. During the gestation of the directive of 1992, Germany endeavoured to have the exclusivity (or health specialisation) rule incorporated into the “insurance” directives with general application. From the perspective of realising the internal market, such a rule would have favoured the development of German groups in countries in which part of the health coverage was set apart for insurance undertakings. However, Germany did not obtain this concession. The non-accumulation (or exclusivity) rule that Germany could impose up to that point was eliminated by directive 92/49 on the grounds that it could no longer be justified. In the miscellaneous provisions at the end of the directive, the health specialisation rule was abolished. Because up to the 3rd directives, a Member State could prohibit a country from dealing in health and something else. This was notably the German case. But it’s basically an insurance activity. It was done more by

 Conversely, the attempt to become established in France failed.

10

2  INSURANCE DIRECTIVES AND THE SINGLE MARKET… 

47

Social Security, but massively regulated. [author’s translation from French] (interview with former SNE, DG 15)

However, it should be underlined that the directive of 1992 made other concessions for substitute private insurance. These were for the purpose of protecting the general public for this branch of insurance. Specifically, the provisions of article 54 of this directive allow a Member State to require, in its territory, compliance with a certain number of specific rules for the contract and activity of substitute health insurance. As indicated by Thomson and Mossialos, in this case, the obligations imposed on insurers were considerably reinforced in comparison to other insurance contracts (Mossialos and Thomson 2004). Although article 54 is worded in general terms and is not explicitly directed at a particular national context, Germany, by the particular configuration of its system, is the most affected. Hence, the directive provided German law (as the host country) the possibility of imposing specific conditions on any insurer of another Member State wanting to access the German substitute health insurance market (e.g. the annuity nature of the insurer’s undertaking or technical management on a similar basis as life insurance) (see Chap. 4).

The Principle of Specialisation in Insurance Activities

As mentioned above, in relation to health insurance, directive 92/49/EEC did not establish the principle of specialisation. Thus Member States were prohibited from requiring that health insurers exclusively pursue health insurance activities (single branch). A health insurer could therefore pursue multi-branch insurance activities. However, from the first directives, the principle of specialisation (or exclusivity) of insurance activities was laid down. Article 8b of directives 73/239 and 79/26 stipulates that insurance undertakings must limit “their objects to the business of insurance and operations arising directly there from, to the exclusion of all other commercial business”. The principle of specialisation therefore imposes monoactivity on insurance undertakings. It prevents insurers from extricating themselves from established prudential rules to pursue non-­insurance (continued)

48 

G. CORON AND M. DEL SOL

(continued)

activities. The goal of this principle is to protect the rights of insured parties by preventing financial resources gained from insurance activities from being used for other purposes—at stake is the effectiveness of the system of financial security underpinning contracts entered into with insured parties. This principle encountered particular difficulties in France when directive 92/49 was being transposed on mutual benefit societies. In France, many of the mutual benefit societies found themselves in a situation of pluriactivity: along with their health insurance activity, they managed health and social activities (clinics, establishments and services for elderly and disabled people, optical and dental centres, pharmacies, leisure centres). These social endeavours were a strong distinctive feature of mutual benefit societies and historically were at the heart of their values. As one author wrote, “the surplus from provident activities, rather than creating profits in the framework of commercial undertakings, were reused in social economy activities. which were not financially “profitable” but effective on a human level for all members” [authors’ translation from French] (Charvin 1996). On several occasions, the French government, impelled by the mutualist movement, proposed several solutions to the European Commission that would have prevented the break-up of mutualist activities—none swayed the conviction of the Commission (Del Sol 2003). Consequently, and inexorably, with the (late) transposition in 2001, France introduced the principle of specialisation in insurance activities into its mutual benefit societies Code: a given mutual could no longer pursue an insurance activity alongside the management of health, social or cultural activities. The latter activities were limited to mutual benefit societies pursuing them in an exclusive manner.

3.3  Failure of the Project for an Alternative Legal Framework for Mutual Benefit Societies Another front of resistance to implementation of the “insurance” directives came from non-profit-making actors within the health sector, specifically French mutual benefit societies.

2  INSURANCE DIRECTIVES AND THE SINGLE MARKET… 

49

The representatives of French mutual benefit societies wanted health mutual benefit societies to be outside of the scope of the “insurance” directives and instead to be governed by a specific directive. Initially, they supported a joint initiative by actors of the social economy, namely associations, cooperatives and mutual benefit societies. The goal of this initiative was to obtain an ad hoc European legal category, a kind of equivalent to the European “joint-stock company” category for the social economy sector. A favourable political and institutional environment made several significant advances possible. In 1989, during the creation of the second European Commission presided by Jacques Delors, a new unit “Social economy” was created within Directorate-General 23 “Enterprise Policy, Commerce, Tourism and Social Economy”. A Frenchman, Paul Ramadier, was appointed to head the unit. The year 1989 was also favourable for two more reasons. First, France held the Council Presidency in the second half. Second, in this context, it organised, in November, the first European Conference on Social Economy, which supported the creation of a European legal category for cooperatives, mutual benefit societies  and associations (Duverger 2016). This is the context in which the social economy cause was placed on the European agenda following a Communication by the Commission to the Council in December 1989, on “Businesses in the Social Economy sector: Europe’s frontier-free market”. At this stage, the goal of the project was a single text for all actors falling within the definition “entities that fall within the scope of the economic field by virtue of the fact that they pursue productive activities, which allocate resources in order to satisfy needs; these entities produce goods or services sold for commercial gain (…) or not (provided at no cost or at a price unrelated to its cost, with the difference provided by funding external to the market: contributions, subsidies, donations)” [authors translation from French] (De Combrugghe 1993, p.  131). However, in 1991, on the request of the European Parliament, the prospect of a single category vanished. After this date, three texts were envisaged and mutual benefit societies henceforth backed a draft text exclusively concerning them and that would create a “mutual passport” allowing them to operate throughout Europe. A note dated 21 June 1991 by the Executive Committee of FNMF presents as imminent the adoption by the European College of Commissioners of the draft text creating the category “European mutual benefit society”. The note indicates that representatives of FNMF were involved in drafting the text.

50 

G. CORON AND M. DEL SOL

The note makes it clear however that the content fell very short of the expectations of mutual benefit societies. Under pressure from DG 15 “Financial Institutions and Company Law”, the text would contain only provisions of an institutional nature (modalities of creation and operation, administrative bodies, etc.). Everything concerning the activity of mutual benefit societies was subsumed into the corresponding sectoral directives, such that the entire spectrum of mutual benefit societies’ activities could be covered: insurance directives, banking directives for mutual credit societies, tourism directives for mutual societies pursuing social tourism activities, and so on. In addition, the national context was also evolving. The French Ministry for Solidarity and Integration was insisting very vigorously that mutual benefit societies and provident institutions (the other major non-profit-making actors providing supplementary health insurance in France) eliminate their risk of infringing competition rules by requesting to enter the scope of the “insurance” directives. The abandonment of the project for a single European legal category for all actors of the social economy meant that mutual benefit societies could no longer rely on the support of other actors of this field, notably cooperatives and associations. On a national level, they could no longer present a united front with provident institutions, whom their representatives were urging to enter the “insurance” directives. Finally, during its general assembly dated 19 October 1991, FNMF reaffirmed its desire for an ad hoc text on the European level; but, at the same time, it was preparing mutual benefit societies to accept their entry into the “insurance” directives; this is clear from the minutes issued by the general assembly [personal archives, FNMF]: The General Assembly demands that efforts continue to be pursued energetically for the purpose of achieving: – a European legal category for mutual benefit societies introducing the possibility of creating, in partnership with Social Economy organisations of other member countries, transnational mutuals, – access, by mutual benefit societies, to the freedom to provide services. The mutual insurance sector demands that, from 1 January 1993, it has the same capacity for action as limited companies and commercial insurance companies, but it requests that European texts:

2  INSURANCE DIRECTIVES AND THE SINGLE MARKET… 

51

– recognise its specificity – impose on all, including European insurance companies, compliance with the basic code of ethics covering risks concerning human persons. […] It requests that all components of the movement redouble their efforts to improve their technical and management performance, in view of increased international competition. [Authors’ translation from French]

From the moment the request for entry into the directives was legally manifested, negotiations would occur directly between representatives of the French state and the European Commission. Thus, the question of what the mutual benefit societies would receive in exchange for joining the “insurance” directives was discarded: France makes its request, the Commission replies: “Yes”, but these are not matters that were escalated for arbitration, when we were trying to have the directives adopted at the level of, […] the Commissioner. There was arbitration before. In short, it wasn’t a source of conflict between France and the Commission. [Authors’ translation from French] (interview, former SNE, DG 15, European Commission)

Thus there was no longer the possibility that the field of private health insurance would take into consideration the specificities of mutual benefit societies. After their entry into the “insurance” directives, health mutual benefit societies became just like any other insurance undertaking—they were trivialised. Given the failure to bring into existence a specific text covering non-profit-making insurance, French mutual benefit societies resigned themselves, after their mobilisation had failed, to being considered common-law insurance undertakings (Benoît and Coron 2019).

4   Conclusion Contrary to a legal and functionalist approach that would have conceived European insurance texts as a unique and rational project meeting the “needs” of insured parties, we have seen that the first two generations of directives can be explained above all by the political configurations of the time (relations between States and institutions), a transverse political project (the single market) and the legal opportunity afforded by the impetus of the Court of Justice. The third generation of texts took into account

52 

G. CORON AND M. DEL SOL

the status quo but the actors also sought to simplify the applicable legal framework, then characterised by fragmentation. The result was a text that sought to implement the single market but also led to transformation of national actors. In the field of health insurance in which the markets of several countries retain a strong national dimension, the directives have not, at least in the first few decades since their entry into force, led to insurers from other countries of the Community becoming a dominant force in national markets. For example, as shown by Pascale Turquet and Philippe Martin in this book (see Chap. 8), VHI Healthcare, the original non profit insurer, has remained a leading actor in Ireland. The transformations induced by the directives did not lead to straightforward substitution of insurers of one country by those of another country that had achieved a dominant position in the continental market. In a certain way, this is a logical observation because Member States sought, during the political process, to protect the strong position of their insurers. The directives did pave the way for internal transformation of health insurers. Systematisation of prudential rules changed the mode of operation, a change that would be reinforced by the Solvency 2 directive (see Chap. 3). Furthermore, the “insurance” directives would reinforce the Ministries with responsibility for the insurance sector; these ministries which, to use Pierre Bourdieu’s phrase, belong to the right hand of the State (Bourdieu 2018), would have a stronger role in managing social matters. The balance of power between insurers could also be modified. An analysis of the “insurance” directives is therefore impossible without consideration of how national actors appropriated them and how they deployed European devices.

References Benoît, C., & Coron, G. (2019). Private Health Insurance in France: Between Europeanization and Collectivization. The Milbank Quarterly, 98, 1108–1150. Berr, C. J. (1988). Droit européen des assurances: la directive du 12 juin 1988 sur la libre prestation de services. Revue trimestrielle de droit européen, 4, 655–688. Bourdieu, P. (2018). On the State: Lectures at the Collège de France, 1989–1992. Cambridge: Polity Press. Busse, R., Blümel, M., Knieps, F., & Bärnighausen, T. (2017). Statutory Health Insurance in Germany: A Health System Shaped by 135 Years of Solidarity, Self-governance, and Competition. Lancet, 390, 882–897.

2  INSURANCE DIRECTIVES AND THE SINGLE MARKET… 

53

Charvin, R. (1996). Le mouvement mutualiste français face aux directives « assurances » européennes. Droit ouvrier, 357. Coron, G. (2006). Union européenne et système de retraite français, une perspective de sociologie du droit. Thèse de sociologie, Université Paul Verlaine de Metz. De Combrugghe, D. (1993). La future réglementation communautaire sur le statut de l’association, de la coopérative et de la mutualité européennes. Revue du marché unique européen, 1, 127–150. De Fruto Gomez, M. (1991). Vers un marché intérieur des assurances. Revue du marché unique européen, 3, 45–100. Dehousse, R. (1998). The European Court of Justice—The Politics of Judicial Integration. London: Macmillan. Del Sol, M. (2003). Les mutuelles de santé: entre logique concurrentielle et éthique solidariste. Revue de droit sanitaire et social, 71–84. Drake, H. (2002). Jacques Delors et la commission européenne: Un leadership unique ? Politique européenne, 8, 131–145. Duverger, T. (2016). Les transformations institutionnelles de l’économie sociale et solidaire en France des années 1960 à nos jours. Revue Interventions économiques, 54, mis en ligne le 01 mars 2016, consulté le 13 décembre 2019. Retrieved from http://journals.openedition.org. Georgakakis, D., & Rowell, J. (2013). The Field of Eurocracy: Mapping EU Actors and Professionals (288 p). Basingstoke: Palgrave Springer. Greer, S., et al. (2019). Everything You Always Wanted to Know about European Union Health Policies But Were Afraid to Ask (202 p). World Health Organization. Jabko, N. (2006). Playing the Market: A Political Strategy for Uniting Europe, 1985–2005 (202 p). New York: Cornell University Press. Levie, G. (1980). La directive européenne sur l’assurance vie. Bull. Ass. De Verzekering, 257, 5–25. Lohéak, F. (1991). L’Europe de l’assurance, le cadre institutionnel. Risques, 8, 9–31. Molinier, J. (1995). Droit du marché intérieur européen (165 p). Paris: LGDJ. Mossialos, E., & Thomson, S. (2004). Voluntary Health Insurance in the European Union (206 p). World Health Organization. Pierson, P. (2000). Increasing Returns, Path Dependence, and the Study of Politics. American Political Science Review, 94(2), 251–267. Sagan, A., & Thomson, S. (2016). Voluntary Health Insurance in Europe. Role and Regulation. World Health Organization, European Observatory on Health Systems and Policies, n°43, 50. Scharpf, F. (2000). Gouverner l’Europe. Paris: Presses de Sciences Po. Smith, A. (2015). Leon Brittan: A Commissioner with a Neo-liberal Vision for Europe. Retrieved from http://theconversation.com/leon-brittan-acommissioner-with-a-neo-liberal-vision-for-europe-36639.

54 

G. CORON AND M. DEL SOL

Smith, A. (2016). The Politics of Economic Activity (208 p). Oxford: Oxford University Press. Stone Sweet, A., & Caporaso, J.  A. (1998). From Free Trade to Supranational Polity: The European Court and Integration. In W.  Sandholtz & A.  Stone Sweet (Eds.), European Integration and Supranational Governance (pp. 92–113). Oxford: Oxford University Press. Streeck, W. (1995). From Market-Making to State-Building? Reflections on the Political Economy of European Social Policy. In S.  Leibfried & P.  Pierson (Eds.), European Social Policy: Between Fragmentation and Integration (pp. 389–431). Washington, DC: The Brookings Institution. Weiler, J. H. H. (1994). A Quiet Revolution: The European Court of Justice and his Interlocutors. Comparative politics, 26(4), 510–534.

CHAPTER 3

Solvency II, the European Government of Insurance Industry and Private Health Insurance Cyril Benoît

1   Introduction On January 1, 2016, a new set of rules came into effect for the European Union (EU) insurance sector, under the umbrella of the Solvency II Directive. The new regulatory regime laid out in the directive now governs the operations of insurance businesses in the EU. Adopted after more than ten years of negotiations and fierce political battles, the text is over 3200 pages long. The financial press hailed it as a “monumental” piece (Ralph 2016) ushering a regulatory “revolution” in the sector (see Schneider et al. 2009). While Solvency II includes a wide range of technical provisions that this chapter will only briefly touch upon, its underlying philosophy has primarily spurred these reactions amongst observers and actors in the insurance industry. Indeed, Solvency II introduces a completely different conception of regulation. The text’s most explicit goal was to set up a “risk-based” approach to insurance regulation, in line with C. Benoît (*) Centre for European Studies and Comparative Politics, Sciences Po (Paris), CNRS, Paris, France e-mail: [email protected] © The Author(s) 2021 C. Benoît et al. (eds.), Private Health Insurance and the European Union, https://doi.org/10.1007/978-3-030-54355-6_3

55

56 

C. BENOÎT

the very basic idea that insurers and reinsurers must maintain capital reserves proportionate to their overall solvency risk. Before it was implemented, a major criticism raised by the European Commission was that the previous regulatory framework did not take into account a number of risks, including market, credit and operational risks with regard to capital requirements. Another set of criticisms pertained to the disparity among EU countries under Insurance Directives: supervisors at the national level had far too much discretion to use their own methodologies when implementing the general provisions established by European directives. Solvency II’s unified and so-called risk-based regulation therefore operates on the basis of a mutually interactive, three-pillared structure, similar to the Basel II and III agreements in the banking sector (Tsingou 2008; Gatzer and Wesker 2012, see however Laas and Siegel 2016). The first pillar sets the quantitative financial requirements on the basis of two different concepts: the Solvency Capital Requirement (SCR) refers to the capital an entity should possess to “absorb an exceptional shock”, and the Minimal Capital Requirement (MCR) is the threshold below which an entity is considered unsustainable and subject to withdrawal of its insurance authorization (see Sandström 2006). A second pillar covers supervisory assessment and governance. It notably defines the concept of Own Risk and Solvency Assessment (ORSA), a process for insurers to identify the risks to which they are exposed and to justify their risk management strategies and controls. The third pillar mostly addresses transparency and reporting, both to the regulator and to the general public. With very few exceptions, all insurance companies covered by the 1992 Life and non-Life Insurance Directives are affected by this change. This includes a wide range of firms, from transnational companies covering a large portfolio of diversified risks to (relatively) small and mid-sized entities focused on a much more limited number of risks or products. As we will see, private health insurance in continental Europe typically falls into the second, very broad and internally fragmented category. This chapter has two main purposes. In Sect. 2, we provide a brief historical account of solvency rules in the EU. Section 3 examines the development of Solvency II, in order to explain the institutional genesis of this particular regulatory framework as well as the rationale for its operational tools. Section 4 examines Solvency II’s distinctive implications for private health insurers in general and for non-profit firms in particular. Interestingly, healthcare as a policy matter was rarely considered an issue requiring specific provisions during the process that led to the adoption of

3  SOLVENCY II, THE EUROPEAN GOVERNMENT OF INSURANCE INDUSTRY… 

57

Solvency II.  When given special attention, it was mostly from a highly technical perspective, such as the calibration of capital requirements and volatility adjustments. Crucially, this also applied to non-profit firms, which many health insurers in continental Europe were. While the smallest entities secured exemptions from Solvency II, most of the health insurance industry now fully operates within its framework. This situation is attributable to the relatively marginal position of several sectoral issues in the development of Solvency II and to the lack of cognitive and strategic resources that non-profit firms and health insurers had to effectively intervene in the EU policy-making process. The chapter draws on political science and international political economy literature. In comparing the previous European prudential framework with Solvency II, we argue that the latter’s main change relates to the regulatory “conceptual power” (Carpenter 2010) it exercises on the insurance industry. As a regulatory framework, Solvency II should indeed not be considered a mere attempt to direct the behaviour of insurance companies in an appropriate manner, as was the case for previous regulations. Like many other directives of this type in contemporary regulation, Solvency II is best described as an “order of concepts” that establishes common definitions and new categories (see also King and Stears 2011). It legitimizes certain ideas—such as the very definition of a “risk-based” regulation—as well as certain metrics and calibrations that provide a conception of what should count, what should be viewed as a risky behaviour or risky product and, conversely, what should not (see Benoît 2019). Concepts such as minimal or solvency capital requirements, the appropriate measures that accompany them and the statistical tools that firms are required to use—all come with their own meanings and hierarchies that are expected to become ingrained in everyday business practice and shape the routines of private organizations. This general process of standardization, which is arguably still unfolding, has neither been a one-way nor a linear process. A diverse cast of firms and rival institutional actors all helped shape the rules and norms set by Solvency II.  As such, the directive did not proceed without creating a number of uncertainties and ambiguities—as similarily argued by Gratz in his own study of Solvency II (Gratz 2019). As this chapter will show, the case of private health insurance under Solvency II largely extends this claim. The rest of the chapter is organized as follows: on the basis of documentary analysis and prior historical work, it retraces the process that led to the adoption of Solvency II. Then, it reviews the genesis of Solvency II

58 

C. BENOÎT

and explains the directive’s basic rationale. This second section considers the case of private health insurers on the basis of qualitative interviews with health insurers and public officials both at the EU and at the national levels and a systematic analysis of 1680 articles drawn from the professional press of the insurance sector.

2   Solvency Rules in the EU before Solvency II: A Very Short Introduction 2.1  From a Fire-Alarm System… The genesis of Solvency II is inseparable from the long and sinuous path that the development of a European regulatory framework in this sector has followed since the early 1960s. Before Solvency II, and as Coron and Del Sol show in their own contribution to this book, this process was marked by the adoption of several directives for non-life (1973, 1988 and 1992) and life insurance activities (1979, 1990 and 1992). With the explicit aim of unifying the insurance market at the European level, each of these was accompanied by specific rules, norms and conventions defining what should be included and excluded from the scope of insurance activities and how to translate security and freedom principles in corporate governance and regulation (Pool 1990). In this sense, the process was very much akin to the formation of a European “government” for the insurance industry (Jullien and Smith 2014). Throughout the period, this government extended its reach by requiring private insurance organizations to comply with rules increasingly defined at the European level. However, the proliferation of directives between 1973 and 1992 does not reflect any institutional upheaval during these two decades. On the contrary, the promulgation of new texts typically responded to needs to extend the principles established by a previous directive (usually from non-life to life activities) or to account for general changes in the European integration process. This was notably the case for the 1992 “third generation” directives adopted as a result of the creation of the Internal Market. However, the underlying conception of prudential regulation and more broadly, of insurance activities per se, has remained essentially the same from one text to the next, despite minor adjustments (see Pool 1990; Sandström 2006; Coron and Del Sol this book).

3  SOLVENCY II, THE EUROPEAN GOVERNMENT OF INSURANCE INDUSTRY… 

59

Post-war actuarial theories, sometimes dating back to the early 1940s, have served as the main notional framework underlying regulatory improvements and changes adopted over this long time period. In this respect, the framework was built around a small collection of basic principles. Until the 1990s, the overall EU regulatory regime’s purpose was essentially considered to be a “fire-alarm” system (François 2020b). Grounded in a stable ideational framework and common objectives, the approach was that European legislation should guarantee the freedom of insurance activities, while ensuring the security of both consumers and producers. This involved establishing a small set of flexible and readable indicators and ensuring their applicability. This conception has mostly been applied to solvency margins, that is, the regulatory capital that an insurance undertaking “is obliged to hold over and above its technical provisions that cover the liabilities” (Vandenabeele 2014). The margin is used as a protection mechanism for insured persons and policyholders “against adverse business fluctuations” and “unforeseen events” (ibid). More formally, it is defined by EU law as “the financial resources of an insurance undertaking, i.e. in essence the difference between the assets and the liabilities of the insurer” (Müller 1997). Following Pool (1990) and Sandström (2006), François (2020b) has extensively described the main features of this regulatory regime as well as its rather simple technical architecture. François argues that the main goal of the first, second and third Insurance directives was not to provide a detailed and sophisticated account of a firm’s solvency, contrary to Solvency II. Rather, the purpose of the regulation and the key motives for its gradual and slow transformations was to identify a range of ratios that could preventively detect potentially risky situations. A major issue was thus to ensure the comprehensibility of these ratios, both from the perspective of insurers and policyholders. This was the bedrock of most of the period’s technical debates. Thus, the question of what entities should be covered by these directives was a much greater matter of dispute than the validity of the solvency margins, as exemplified by the creation of the third-generation directives studied by Coron and Del Sol (Chap. 2). The gradual accumulation of policy tools from one directive to the next was subsequently defined as forming the “Solvency I” system of insurance regulation. Its most basic features, put in place just before the beginning of the Solvency II “process” (see below), have been summarized in Vandenabeele (2014):

60 

C. BENOÎT

Under Solvency I, the solvency requirements are calculated as a percentage of the technical provisions or the mathematical reserves […] The undertaking is deemed solvent if the solvency requirement is covered by sufficient available capital […]. The solvency of an insurance undertaking is measured as a fixed ratio (available capital over required capital), being transparent, easy to calculate and simple to audit. [Emphasis added]

In this particular setting, a highly readable framework was intended to fundamentally help define clear and explicit signals, enabling supervisory authorities—whose main work is to prevent the bankruptcy of insurance companies—to detect risky situations. As such, the solvency margins measured in percentages provided the fire-alarm function. This general conception exhibited two basic facets of regulatory power (Benoît 2019). Through a range of legal and regulatory provisions, European directives formed a “directive power” (Carpenter 2010), namely an attempt to direct the behaviour of some regulated entities. It was in turn legitimized by a range of values translated into specific mechanisms. These values mostly stemmed from a compromise between the security of policyholders and the liberty of producers, as is most often the case in European law and regulation (Smith 2016). Their translation proceeded through a range of technical yet basic actuarial theories (Sandström 2006). To some extent, this framework can also be seen as a “gatekeeping power” (Carpenter 2010), since the solvency margins as a fire-alarm system was in principle established to prevent any abuse or risky behaviour from a firm. In this respect, it was a power exercised prior to interactions between insurance companies and policyholders, since the former needed to define their strategies on the basis of these solvency requirements, and to avoid behaviours that could jeopardize an internal equilibrium between their activities. That being said, this regulatory framework was not likely to influence everyday business operations, since considerable discretion was left to firms and national regulators on many of their activities. The result was “coercive” regulation that directly prescribed a rather small number of set requirements (see Treib et al. 2007). 2.2  … To a Genuinely Risk-Based Regulation? In this context of apparent stability, how to explain the emergence of the ambitious project that Solvency II became between the end the 1990s and the beginning of the 2000s? Another look back at the context of the

3  SOLVENCY II, THE EUROPEAN GOVERNMENT OF INSURANCE INDUSTRY… 

61

development of the 1992 Insurance directive is needed to account for this transformation. While a reform of solvency rules was not at the heart of the agenda, several actors, mostly located within the Financial Services unit (Internal Market General Directorate) of the European Commission, suggested a review of the solvency rules. This suggestion was echoed in a similar position adopted by the European Council; however, it was simultaneously acknowledged that such a costly enterprise could substantially slow down the adoption of the third-generation directives. The discussion was therefore postponed (EC 1997). Following this request, the European Commission initiated two separate reports on the issue in 1994. A questionnaire was first developed under its auspices and submitted to three industry representative bodies: the Comité européen des assurances (CEA, “representing EU insurance undertakings limited by shares and certain mutual undertakings”); the Association des assureurs Coopératifs et Mutualistes (ACME, “representing other mutual and cooperative insurance undertakings”) and the Actuaries Advisory Group (ibid). At the same time, the Commission requested a report on solvency margins from the Conference of Insurance Supervisory Services of the Member States of the EU (hereinafter “Conference of supervisors”, a group consisting of national supervisors). Chaired by Helmut Müller, vice-president of the Bundesaufsichtsamt für das Versicherungswesen (German federal supervisory Authority), it collected similar information, but focused more on the expectations of national supervisory authorities (Müller 1997). Both reports shared a similar roadmap. Their common objective was to determine whether a change in the calculation of solvency margins was needed, and if so, whether greater harmonization was required. Both also came to very similar conclusions: Examination of insolvency cases and, more generally, financial difficulties of insurance undertakings within the EU in the past twenty years leads to the conclusion that the current solvency margin scheme not only has proved to be soundly based as to principle, but also has produced satisfactory results in practice. EC (1997) …the insurance system created by the community in 1973 and 1979 has proved itself and there [is] therefore no reason to totally revise it. On the contrary, the working group is convinced that the existing system adequately takes account of the insurance-specific risk situation if some amendments and additions, which will not affect the core of the system, are carried out. Müller (1997)

62 

C. BENOÎT

The two reports thus considered the overall framework established by the 1992 directives to be appropriate and reasonable. Some amendments were recommended, notably on long-term and investment risks (EC 1997, pp. 11–12; Müller 1997, pp. 18–19). It was also acknowledged that these adjustments would require the establishment of a new group of government experts chaired by the Commission (EC 1997, p. 15). Nevertheless, both reports explicitly agreed that a complete overhaul of the system would be very costly and hazardous. Crucially, the possibility of implementing a “risk-based” approach—as already existed in some countries, like the United States—was explicitly considered. Still, the Müller report suggested proceeding with great caution in this regard and recommended eschewing this approach, arguing that “sufficient experience with the risk-­ based capital system has not been gained to date” (1997, p. 12). Just two years later, however, the European Commission published a new report (EC 1999). It had been prepared by the “Insurance and Pension Funds” division (hereinafter Insurance division), an entity associated to the aforementioned Financial Services unit of the Commission. In this new report, the Commission distanced itself from the rather moderate position recommended by the Müller report, and by the same token, from the position it had adopted in 1997. Two major and—at least in appearance—paradoxical announcements were made at this time. First, and with the explicit goal of strengthening consumer solvency protection in the insurance market, the European Commission tackled coordination issues across supervisory bodies by providing an initial rules-based set of minimum capital requirements (Van Hulle 2005). At the same time, a more ambitious agenda was outlined. The Commission’s overall objective was to implement a holistic “risk management approach toward capital standards” in the next few years (Eling et al. 2007). The Commission simultaneously announced that it would update and unify the “outdated thresholds” of the existing regime without changing its overall structure, while fundamentally reforming the supervisory structure itself over the next few years (Doff 2008). The first proposal would be operationalized through the EU Directive 2002/13/EC for non-life insurers and EU Directive 2002/83/ EC for life insurers. This would come to be known as the Solvency I system—a term that was not commonly used until then. The second much more ambitious proposal became the Solvency II “process”. Interestingly, the broad set of values attached to it did not differ from previous EU-led reforms in this industry. Solvency II was presented as a

3  SOLVENCY II, THE EUROPEAN GOVERNMENT OF INSURANCE INDUSTRY… 

63

necessity to “deepen the integration of the EU insurance market, to enhance the protection of policyholders and beneficiaries, to improve the international competitiveness of EU insurers and reinsurers, and to promote better regulation” (Doff 2016). Yet the very definition of these same values and, conversely, the means to achieve the objectives attached to them, were entirely different. Indeed, the Commission suggested the need for a new approach to regulation that would account for a wider range of risks, within a much more unified framework, and in accordance with higher transparency principles. The Commission report argued that these three dimensions were missing from the Solvency I framework. According to its diagnosis, this could be remedied by introducing a risk-based system and stricter requirements on capital adequacy and risk management for insurers, on the basis of internal or external models that would explicitly consider diversification and risk-mitigation effects (see below; Der Meij 2011; Vandenabeele 2014). Several contextual explanations have been provided in the literature to account for this apparently sudden shift. Quaglia (2011) has for instance emphasized the “strong alliance between the Commission and the United Kingdom, backed up by the large member states, some old member states and industry, particularly large companies and transnational groups”. Taking a broader perspective, François (2020c) has argued that the very parameters of these debates and the role of the Commission are more attributable to long-term transformative changes that occurred in other areas and gradually permeated entire segments of the insurance industry. According to François, a first series of changes occurred in the academic field, specifically in the area of financial theory, where many concepts related to reporting in Solvency II originated (see also François and Frézal 2018). These developments gradually spread to regulatory bodies in the banking sector and to different communities of transnational experts. A second sphere was the world of accounting standards, where a number of initiatives were carried out by international bodies throughout the second half of the twentieth century—such as the International Accounting Standards Board (IASB)—to update accounting frameworks, sometimes without the explicit goal of changing the rules governing the insurance industry. Yet these standards were increasingly recognized and pre-emptively adopted by some large companies, as well as some national regulators (see also Doff 2008). This was notably the case for “fair value accounting”, namely the use of current market values to recognize certain assets and liabilities instead of the price at which they were initially

64 

C. BENOÎT

purchased (book values). The last—but certainly not the least—area of transformative changes occurred in banking regulation, unsurprisingly in close connexion with prior developments in financial theory. Widely acknowledged in the insurance industry, the journey from the Basel I (1988) to Basel II (1998) agreements (see Lall 2012) was a major vehicle for a significant part of Solvency II’s conceptual framework. This was the case for the concept of “internal models”—a model developed by a firm and approved by the regulator that determined the capital required to meet risks to which a firm was exposed—and, more broadly, a risk-based approach to minimum capital adequacy requirements (Gordy et al. 2014). François (2020c) argued that the coevolution of these three spaces of standards formation and the growing diffusion of these same standards ultimately paved the way for a change in the Commission’s approach to regulation. As aforementioned, “Bancassurances” (i.e. banks offering insurance products) and large insurance companies arguably played a major role in this process. Some regulators at the national level also contributed to it. This was notably the case in the UK, where additional rules on capital requirements were introduced (Van Hulle 2005), explaining Quaglia’s (2011) finding of UK influence in setting the agenda of EU authorities. The origins of Solvency II are thus manifold, as were the range of hierarchies, metrics and ideas that would form the basis of its conceptual structure. From an internal perspective, the diffusion of standards from various spaces increasingly came to shape the Insurance division’s own diagnosis on the existing regulatory framework. Our own interview data suggests that the 1992 Insurance Directives were indeed considered as an initial step by at least some key participants in the process and that they believed further improvements would be needed over the next years. The [1992] Directives opened the market. But on the operational rules, on the prudential rules, they did not fundamentally change anything. They did not change capital requirements […] Solvency I dates back to the first Insurance Directives in 1973. In the 3rd generation directives, we just added … well on capital requirements, we did not go too far. We basically said that they sufficed [laughs]. Interview, Former Seconded National Expert at the Insurance unit (European Commission) Author’s translation from French

3  SOLVENCY II, THE EUROPEAN GOVERNMENT OF INSURANCE INDUSTRY… 

65

However, at this initial stage members of the Insurance division were not in a position to use the aforementioned concepts, link them within a substantial programme, and more importantly, integrate them within a coherent “policy narrative” by referring to general principles (Schmidt and Radaelli 2004). A few years later, the adoption of Basel II agreements opened a window of opportunity, showing that “it was possible to move towards a finer appreciation of regulatory risks”, according to the former chief of the Insurance division himself (Van Hulle 2005). In this regard, the provisional updating of Solvency I thresholds announced in 1999 was an essentially “tactical change” that allowed these actors to engage into a more wide-ranging review of solvency rules (Thorburn 2004). It is indeed clear that rather than solely offer minor improvements to the existing framework, their goal was to explicitly push for a more radical and philosophical change. Accordingly, the Commission formally initiated a review of existing insurance industry regulations in May 2001 (EC 2003; see also Eling et al. 2007): In the first phase, May 2001 to April 2003, fundamental arrangements were specified, a general framework was defined and several studies were ordered by the European Commission. In the second phase (beginning December 2003 […]), these fundamentals are being developed into specific rules and guidelines, with frequent opportunities for input from member states and relevant stakeholders. Eling et al. (2007)

3   Solvency II: The Making Architecture

of a Conceptual

3.1  The Beginning of the Solvency II “Process” “Fair value accounting”, “Risk-based Regulation”, “Internal Models”… Solvency II developed as a series of regulatory concepts that have become so complex that it is hopeless to attempt a comprehensive review of each of its components in the space we have, and this chapter will not attempt such a review (for a more detailed account, see Doff 2016; European Commission 2015; François 2020a; Frézal 2018; Gatzer and Wesker 2012; IFA 2016; Marano and Siri 2017, esp. Chap. 1, 5 and 7; see also Zhang 2014). Instead, the following synthesis outlines key features of the Solvency II process and demonstrates why it—as well as the final version

66 

C. BENOÎT

of its text—can be described as the emergence of a particular form of regulatory power that is quite distinct from the 40-year-old framework that preceded it. The previous section showed that Solvency I merely consisted of a fire-­ alarm system, and that major criticism of the system was related to the fact that it was considered insufficiently risk-based. Two reports, drafted in the early phase of the Solvency II process, helped legitimize and entrench this diagnosis. Both were commissioned to set the stage for subsequent discussions, and they framed the Commission’s agenda for the next few years (Eling et al. 2007). A first report was commissioned from the KPMG consulting firm and was published in 2002. Closely aligned with the initial preferences of the Commission’s Insurance division, the KPMG report’s main conclusion was that Solvency II should adopt the same three-pillared structure that of the banking sector. As was the case in the Basel II agreements, these three pillars would respectively involve quantitative requirements, supervisory activities, and public disclosure (KPMG 2002, see below). The second report was commissioned from Paul Sharma, chairman of the aforementioned Conference of national supervisors, and it consisted of a survey conducted among all regulatory bodies within the EU. While it did not fundamentally challenge the Commission’s views, this report was much more interested defining risks and in detecting and correcting risky situations (Sharma 2002). The report recommended that these be appreciated in terms of “capital adequacy and solvency”, measured as a firm’s ability to financially cope with the effects of the risks to which it was exposed (Eling et  al. 2007). Contrary to the three-pillared structure mentioned in the KPMG report, this approach was not necessarily at odds with Solvency I.  However, the Sharma report also called for coverage of the “full risk causal chain”, requiring the creation of additional indicators and preventive tools in order to strengthen an ex-ante regulation of threats to an insurer’s solvency (Sharma 2002). In the same vein, the report recommended a greater assessment of “management quality and adequacy of internal systems”. More explicitly, it called for a regime focusing on internal factors like the quality and suitability of an insurer’s risk management system and the adequacy of corporate governance practices (Eling et al. 2007). On the basis of these two reports, a proposal was issued by the Commission and endorsed by Member States’ representatives in April 2003 (Quaglia 2011). Meanwhile, discussions unfolded on what

3  SOLVENCY II, THE EUROPEAN GOVERNMENT OF INSURANCE INDUSTRY… 

67

legislative process should be monitored to turn the proposal into a directive. Following recommendations made by the Parliament and Council, the Commission decided to implement the recently endorsed “Lamfalussy process”, which had initially been designed for the securities sector. Adopting a four-level approach, this system distinguished between the framework directive focused on general principles (level 1), delegating and implementing acts, that is, the more technical regulations and formulas (level 2), regulatory standards and non-binding guidance (level 3) and enforcement of EU legislation where the Commission services checked if the legislation had been properly followed (level 4) (Linder and Ronkainen 2004). In practice, each level operated on the basis of a different committee system. The European Commission, the European Parliament and the Council of the EU were involved at level 1. At level 2, the process was overseen by a new European Insurance and Occupational Pensions Committee (EIOPC)—an entity consisting of “high-level representatives of Member States” and established to advice the Commission. At the same time, the Conference of supervisors became the Committee of European Insurance and Occupational Pensions Supervisors (CEIOPS) and was established as a level 3 committee that would also provide the Commission with some advice at level 2 (Vandenabeele 2014). Several working groups were formed within this overall context to help the Commission issue a proposal for a directive of the European Parliament and the Council, ultimately submitted in July 2007 (Rae et al. 2017). Broadly speaking, the regulation proposed in Solvency II had two main features (Eling et al. 2007; Linder and Ronkainen 2004). As aforementioned, the first was a three-pillared structure, in line with the KPMG report’s recommendations and the Commission’s initial preferences. Pillar 1 established capital requirements, including technical provisions and rules for determining the minimal capital required and the target capital, which can be broadly defined as the amount of capital a company seeks to hold to allow for adverse future experiences. It also stated the conditions for using a standard or an internal (firm-based) formula. The four controlling and monitoring risk categories specified for consideration (EC 2004a, b) were underwriting risks (i.e. risk that may arise from an inaccurate assessment of the risks associated with an insurance policy or from uncontrollable factors), credit risks, market risks (i.e. resulting from the fluctuation of all relevant market prices such as stocks or bonds) and operational risks (“which result from inadequate or failed internal processes, people and systems or from external events”—see Eling et  al. 2007). Pillar 2

68 

C. BENOÎT

essentially reflected the proposals made in the Sharma report, since it established some principles for internal risk management and internal risk control along with associated supervisory interventions, in line with the risks recognized by quantitative models in Pillar 1 (Eling et  al. 2007; Linder and Ronkainen 2004). Pillar 3 was closer to suggestions made in the KPMG report, since it covered market transparency and disclosure requirements, in order to promote market discipline on the basis of internationally recognized financial reporting standards (Linder and Ronkainen 2004). The other main feature of the proposed framework was its approach to capital requirements, as defined in Pillar 1, since a two-level method was envisioned. The first level very much resembled the Solvency I framework, since it consisted of a standard rules-based minimum capital. However, the proposed version also suggested taking into account additional requirements for life insurers based on investment and actuarial risks, in “a foreshadowing of Solvency II intended risk-based approach” (Eling et al. 2007). Presented as the “crux” of the text, the second level consisted of a required solvency capital based on market value. It reflected “the economic capital an insurance company needs for running its business within a given safety level, and is determined as the probability of insolvency or by using the tail-value-at-risk or another similar measures” (see Eling et al. 2007, Sandström 2006 for a discussion). 3.2  When Solvency Rules become a Site of Political Struggles This basic proposal became a field of intense political struggles, particularly until the final version of Solvency II came into force in 2016, and even years later since the text is subject to periodic reviews. From a theoretical standpoint, a number of clues indicate that it is already a profoundly different approach to insurance regulation, and this beyond its technical features—or more precisely, as reflected by its technical features. While the Solvency I fire-alarm system allowed regulatory entities to detect and sanction illegitimate behaviours at a certain distance from the market, Solvency II developed an entire conceptual structure, wherein regulatory concepts are ingrained in everyday business practice. Contrary to Solvency I, quantitative requirements in Solvency II explicitly provide a conception of optimal capital allocation, conceptualized as an allocation that is assumed efficient in terms of both risk and return for shareholders (see EC

3  SOLVENCY II, THE EUROPEAN GOVERNMENT OF INSURANCE INDUSTRY… 

69

2015). It also includes a range of regulatory metrics and calibrations to guide private risk management strategies and business decisions. As such, this overall conceptual structure builds on a number of standards and definitions of “risky” and “safe” situations, with assets and behaviours themselves defined hierarchically into specific categories. The relationships between the three pillars, the two levels of capital requirements and even disclosure requirements and transparency are also similarly linked. In sum, Solvency II offers a new vocabulary, giving firms and regulators new guiding categories (King and Stears 2011), by contrast to Solvency I, which was both a more coercive and distant form of regulation. Formal provisions appear as technical expressions of inter-value hierarchies (“security”, “transparency”, “freedom”, “comparability” or “safety”, see Smith 2016) that now ought to govern the industry. This general statement does not mean that regulators necessarily secured more leverage to govern the behaviour of insurance companies under Solvency II. The profusion of rules and regulatory instruments ultimately became a source of regulatory failure in other areas, and Solvency II’s direct impact on firm behaviour is still a matter of debate (Rae et  al. 2017; see also Abecassis and Coutinet this book). Yet, and although it is a question of prime importance, the formal power that is exercised by firms and regulators does not constitute a criterion for gauging the intensity of such a conceptual power. Indeed, this notion refers more to the type of power exercised than to the extent to which different parties exercise it (see Benoît and Thiemann in press for a discussion). Accounts of the process that led to the final adoption of Solvency II describe intense debates on the scope, choice and very definition of particular concepts used in the framework. Most of them were somehow linked to accounting standards, the level of capital requirements and the role that internal models should play in this new regulatory environment (see in particular François 2020c, see also Rae et al. 2017). As such, this process can be described as highly political, because of both its conflictual and contingent nature. Three periods can be loosely distinguished. The first runs from the updating of Solvency I thresholds (around 2002) to the adoption of a Solvency II proposal for a “Directive of the European Parliament and of the Council on the taking-up and pursuit of the business of Insurance and Reinsurance” in July 2007, with a target date of October 31, 2012 (Rae et al. 2017; see also Steffen 2008). This period was notably marked by three waves of Quantitative Impact Studies (QSIs) launched by the CEIOPS on the practicability of the calculations

70 

C. BENOÎT

involved in Solvency II (QSI1), on the assessment of technical provisions with regard to minimal and solvency capital requirements (QSI2) and on standard formulas, including the assessment of eligible capital (QIS3). A QSI is a process whereby non-life and life insurance undertakings and reinsurers are invited to test some parameters of the future directive on their own activities (CEIOPS 2006a, b, 2007). Most QSI participants were large insurance and bancassurance companies, although smaller firms—particularly health insurers—also participated in these tests (see below). Of moderately low to moderately high intensity, the debates that occurred during this period essentially focused on risk weighting in the standard formula, which was of prime importance for larger groups (notably, market risk), and the calibration of penalization factors (see in particular FFSA 2006). Such debates were not driven by ideational factors so much as by country interests closely aligned with those of “national champions”—such as Axa for France and Allianz for Germany (see François 2020d; Gratz 2019). A major subject of inquiry was thus the fate of some national provisions under the new regulatory framework. For example, life insurance contracts in Germany and Austria may use collective bonus reserves—usually reserved for policyholders’ future surplus participation—to partly cover future losses (Burkhart et  al. 2017). Deciding whether to consider these surplus funds as basic own funds or as technical provisions greatly impacts a firm’s solvency ratios. The French and British representatives (both of national treasury institutions, industry federations and even individual firms such as Generalli, Axa and Aviva) were much more absorbed by debates around the notion of “group support”, a major issue in terms of capital requirements and reporting work (François 2020d). The discussions here focused on whether capital requirements and supervisory activities should be organized at a group level or at a subsidiary level. Unsurprisingly, major international companies promoted a framework that reflected their transnational dimension. Other debates related to the weighting of shares and bonds in terms of capital requirements—a concern raised by French insurance companies in particular, the general idea being that Solvency II should not penalize insurers’ investments in the “real economy”. A compromise was finally reached on these different points in 2007, especially between France, Germany and the UK and between the European Parliament and the Council. In this version, the text notably acknowledged the notion of group support by allowing subsidiaries to recognize spare capital within the group. In addition, SCR

3  SOLVENCY II, THE EUROPEAN GOVERNMENT OF INSURANCE INDUSTRY… 

71

in excess of MCR could also be covered by group support (McCulloch 2009). Germany also secured what it had sought on surplus funds. In 2008, the bankruptcy of Lehman Brothers called all these compromises into question, opening a second, brief but highly conflictual stage of negotiations. In the wake of the financial crisis, the different working groups reconsidered the current version of the text, as well as its potential implications in light of the financial crisis. Some actors, such as CEIOPS, problematized this overall situation as emphasizing the “even greater importance of moving to a system that encourages and enhances risk management”, notably through a better “understanding of credit and liquidity issues in extreme stress scenarios and over different time horizons”, as well as of “underlying risk exposures for complex instruments” (Kean 2009). The principle of Solvency II was thus re-legitimized—and an extended acceptation of the risk-based approach it entailed was presented as an appropriate response to the challenges raised by the ongoing crisis. While the overall conceptual structure remained, some major debates arose around the notion that the text placed excessive emphasis on the freedom of larger firms to the detriment of financial safety issues. The notion of group support in particular was seen as granting them too much leverage, especially if a group wanted to strip capital or assets from its subsidiaries in other countries. Moreover, additional guarantees would be needed to ensure that a subsidiary would be protected by its parent company in all cases. This point was notably raised by countries without national champions, and where the insurance industry mostly featured subsidiaries of foreign companies, such as Spain and Eastern European countries (Quaglia 2011). This intense process finally resulted in a text that maintained most of the features of Solvency II, except the notion of group support, which was dropped from the newer version and replaced by additional provisions that strengthened cross-border regulation (McCulloch 2009). A new agreement was thus reached on this basis, and the European Parliament approved the Solvency II Framework Directives (level 1) in 2009. While slated to take effect on October 31, 2012, several technical questions remained and conflicts at the levels 2–3 (delegating and implementing acts) forced Commissioner Michel Barnier (DG Internal Market, MARKT) to delay the process until January 2016. The industry and its representatives proved to be a major source of gridlock during this third stage for two main reasons. First, larger groups were facing the

72 

C. BENOÎT

consequences of the financial crisis; and underlying actuarial assumptions behind some technical instruments laid down by Solvency II were now more intensively questioned (Foroughi 2012). On the other hand, and more globally, several years of negotiations gave insurance companies the time to better grasp the multiple implications of this new framework on their daily activities as well as on their long-term strategies. The introduction of liquidity premiums and questions related to whether they should apply to insurance liabilities constituted an especially large stumbling block at this stage of the negotiation, since it could impact overall capital requirements (CEIOPS 2010). A number of issues related to long-term products, volatility and pro-cyclicality, proportionality and the reporting burden were also fiercely discussed (Rae et al. 2017). These debates lasted until 2014, when the European Parliament approved the Omnibus II Directive, which, by focusing discussions on the valuation of assets and liabilities and the risk-free interest rate, led to compromises on these issues—especially between French, German and British representatives. In its amendment to the Solvency II (2009) framework directive, Omnibus II addressed some practical matters, including the recognition of the newly created supervisory authority (the European Insurance and Occupational Pensions Authority, EIOPA), which replaced CEIOPS.  It also aligned the introduction date of Solvency II with the accounting year of most insurers (Doff 2016). The general principles and overall structure of the text did not change during these latter stages of the process. In its 2016 version, Solvency II thus comprised three pillars, wherein Pillar 1 “sets out the minimum capital requirements that firms are required to meet”, “specifies valuation methodologies for assets and liabilities based on market consistent principles”, using “two distinct capital requirements” (i.e. SCR and MCR)—the former can be calculated using a “prescribed standard formula approach, or by using a company-specific internal model, which has to be approved by the regulator” (Institute and Faculty of Actuaries 2016). Pillar 2 includes the “supervisory review process, systems of governance and risk management under which supervisors may decide that a firm should hold additional capital against risks that are either not covered or inadequately modelled under Pillar 1”. Each insurance company is also required to carry out an Own Risk and Solvency Assessment (ORSA) to identify “the risks to which it is exposed, including those not covered under Pillar 1, to

3  SOLVENCY II, THE EUROPEAN GOVERNMENT OF INSURANCE INDUSTRY… 

73

identify the risk management processes and controls in place, and to quantify its ongoing ability to continue to meet the MCR and SCR”. Pillar 3 mostly consists of the disclosure and supervisory reporting regime, under which defined reports to regulators and the public are required to be made (ibid). As of January 2016, Solvency II (Directive 2009/138/EC)—as amended by Directive 2014/51/EU (“Omnibus II”)—replaces 14 existing directives known as Solvency I (EC 2015). As suggested by this brief outline of the debates, however, the directive has been amended with an enormous volume of complex technical details. It is an understatement to say that for over 15 years, Solvency II has constituted a field of political struggles that our brief outline has barely scratched the surface of. Nevertheless, some findings are apparent. The deeply political process wherein all the pieces of this regulatory framework were intensively debated, negotiated and fought over reveals the conceptual nature of this emerging regulatory power, in the sense that it is essentially based on linked, hierarchical categories, forming a conceptual structure that is in turn expected to shape entire segments of insurance activities. As such, “the change in the game made by Solvency II is as much about the power of the regulators as it is about conferring authority on standards and internal models” (Gratz 2019). These symbolic and normative hierarchies qualify certain types of capital as superior to others, and many debates during the Solvency II process focused on matters of classification, as suggested by the discussions on what should count as “surplus” or “basic” funds—in other terms, how they should be hierarchized and valued accordingly. In this respect, formal provisions laid down by Solvency II reflect inter-values hierarchies, as illustrated by concerns raised after the bankruptcy of Lehman Brothers, or in relation to group support, or even the need for transparency that legitimized the provisions in Pillar 3. While Solvency I proceeded through more direct and tough punctual intervention, Solvency II defines and enforces rules, norms and an ultimately more sophisticated regulation by ensuring common standards affecting the daily lives of both private and public organizations. The conflictual nature of many disputes around the directive demonstrates that the different actors involved were at least at some point aware that they were fighting about the very standards (King and Stears 2011; Scott 1998) of their industry that will define many of their operations for years to come. It is certainly because of this nomothetic ambition that Solvency II has been described as “well-intentioned folly” (Swarup 2012).

74 

C. BENOÎT

4   Health Insurance under Solvency II 4.1  Health Insurance: A Peripheral Issue during the Solvency II Process Assessing the magnitude of the changes involved in such a conceptual overhaul is difficult in the short-term, especially because so many different organizational settings are involved (see, however, Gratz 2019 for a recent illustration). But it is important to note that Solvency II’s conceptual features ensure that while insurance companies are the main regulatory target, they will also serve as key players in shaping, diffusing and interpreting this order. Health insurance is a case in point. In Western Europe, where the bulk of health expenses are covered by public or quasi-public entities, it has mostly developed as a voluntary insurance scheme. It operates on a complementary basis by covering services that are not fully reimbursed by the public purse or on a supplementary basis—as in the UK—by guaranteeing “a wider choice of providers, faster access to treatment or […] superior accommodation in hospital” (Thomson and Mossialos 2002). Health insurance is particularly established in Ireland and in the Netherlands— two healthcare systems characterized by a mix of public and privately funded health services—and in France, where complementary health insurance accounts for around 13% of health spending. In the UK, “private” health insurance is typically purchased to avoid waiting lists and to gain access to a greater choice of providers and treatments (Clarke et al. 2017). Most of these firms are essentially focused on health insurance, even though some of them may have other activities such as pension-like or even life products, such as Achmea in the Netherlands and April in the UK. Most of them are also small or medium-sized companies, and even the bigger ones are much smaller than the dominant firms operating in Europe, where the 20 largest groups collect about half of the premium income across the continent. During the development of Solvency II, these companies with fewer strategic, institutional and cognitive resources focused most of their political efforts on capital requirements linked to health risk, especially after the second wave of QISs launched by CEIOPS (2006b). Depending on the category in which health is classified, it might be associated with longer or more or less volatile risks, which might significantly increase capital requirements (IFA 2016). Led by French mutual benefit societies and

3  SOLVENCY II, THE EUROPEAN GOVERNMENT OF INSURANCE INDUSTRY… 

75

supported by the French government, this political work has proven successful. [With the support of the French government], we explained that health was relatively non-volatile. This helped substantively modify its calibration […] [and] had a direct impact in terms of capital requirements for our organizations that total around 2.5 billion euros. Former president of the national federation of French Mutual Benefit Societies, quoted in L’Argus de l’Assurance, March 2014 Author’s translation from French

In the aforementioned countries, it particularly helped firms with a focus on health to maintain average solvency coverage ratios that were significantly higher than the mean of European non-life insurers, except for the UK (Table 3.1). 4.2   Changing Organizations … and Beliefs? However, and as previously shown, Solvency II covers much more than just capital requirements. Even if this framework considers health as a short-term, relatively simple and consequently less demanding activity, health insurers are very much exposed—more so than diversified and bigger companies—to regulatory necessities. Indeed, a number of health insurers in Europe are non-profit firms, such as cooperatives (e.g. VGZ, one of the largest health insurance companies in the Netherlands) and mutual benefit societies in Belgium and France. Accordingly, their operational rules and practices were quite different from their for-profit counterparts before Solvency II came into force. As a result, they have had to Table 3.1  Weighted average solvency coverage ratios in health by category in four EU countries Average solvency coverage ratio

SCR (%)

MCR (%)

Irish Health Insurers UK Health Insurers Dutch Health Insurers French Health Insurers European Non-Life Insurers

286 162 196 215 184

979 412 523 475 –

Based on Clarke et al. (2017), data for France based on 2017 DREES figures (Ministry of Health)

76 

C. BENOÎT

integrate a vocabulary and a range of actuarial techniques such as advanced calculation methodologies and the principle of narrative reporting as codified in ORSA. In order to meet regulatory requirements, these non-profit entities now “take a prospective stance, based on a full business plan” similar to the practices of capitalist organizations; moreover, they “adopt a risk-tolerance framework” and “carry out stress-testing”, in order to define and implement “a capital management strategy to ensure on-going compliance with [capital] requirements” (Masse 2016). Solvency II also has an impact on corporate governance, particularly for mutual benefit societies in France. They traditionally operated on the basis of “democratic” principles, since they were run by an executive board elected from their membership (see Benoît and Coron 2019). In accordance with “fit and proper” requirements established by Solvency II, these people now have to demonstrate professional qualifications, practical experience and analytical skills that most of them are unlikely to possess, requiring specific trainings and conversely, an intense familiarization with most features of Solvency II (EIOPA 2013). This conceptual order also involves a range of provisions and guidelines to ensure “good corporate governance”. For mutual benefit societies, this has notably resulted in an explicit separation and an increased formalization of the relationships between their democratically elected board and chief executive officers (CEOs). CEOs are not necessarily new actors within mutual benefit societies; however, Solvency II has given them a much more prominent role. Granted, these are essentially formal provisions, and organizations do not radically change in a few years, even under the pressure of vast and powerful standards, such as those shaping Solvency II’s conceptual structure. Old principles are not necessarily incompatible with the new framework gradually ingrained in the daily lives of these entities. Moreover our data does not suggest that Solvency II involves or possesses agency per se. What they reveal, as we would like to emphasize, is that it has defined and developed a series of contexts wherein agency and a set of behaviours are both authorized and enacted (Hay 2014). The much-needed adaptation to this directive and the diffusion of a dedicated vocabulary and associated categories need to be understood as such, because this conceptual order ultimately rests on stabilized acceptations of what should be legitimately considered as a safe, sound, robust and prudent organization. Again, non-profit health insurers constitute a meaningful case to further justify this claim. With Solvency II, they have indeed been exposed to less noticeable but enduring transformations that have changed inter-value

3  SOLVENCY II, THE EUROPEAN GOVERNMENT OF INSURANCE INDUSTRY… 

77

hierarchies within these organizations. They have traditionally reinvested their surpluses into benefits and services to their members. Even though the notion of capital requirements does not constitute a major strategic issue for them (as given in Table 3.1), it affects the underlying conceptions that mutual benefit societies have used to define their activities. Indeed, these entities, particularly in France and Belgium, developed against “capital-based solidarity”—a conception that is now jeopardized by risk-­based management techniques and capital requirements. I understand that [solvency requirements] also constitute a guarantee for the members [of a non-profit insurance company]. I understand. But this means that, and this is not necessarily a criticism, we are entering a new era […] It was through group solidarity that we envisioned facing future challenges […] Solidarity between today’s members address today’s challenges. Since the organization is human-centred […] tomorrow’s members will address tomorrow’s challenges and this will continue […] It is thus a human, collectively-based guarantee. [Under Solvency II] […] the guarantee is based on capital requirements […] We substituted a guarantee based on solidarity between human agents, today and tomorrow, with capital-based solidarity, which is no small matter. Interview, Former President of the biggest French Mutual Benefit Society Author’s translation from French

In several countries, Solvency II has also changed the relationship between private health insurers and their regulators in a way that largely exceeds the text’s formal provisions. Once again, France offers a relevant case to grasp some of the transformations arguably at play here. Before Solvency II, non-profit private health insurers had their own supervising authority to better address the distinctive features of their business model—the Autorité de contrôle des assurances et des mutuelles. This authority was then merged with that of bancassureurs and for-profit insurers—the Autorité de contrôle prudentiel et de résolution (ACPR), an entity overseen by the governor of the Bank of France. In this general context, our qualitative data suggest a growing integration of private health insurers with the demands of their regulator’s own monetary and policy agenda. We had informal meetings with Christian Noyer [former governor of the Bank of France], who brought together the so-called stakeholders of the financial area. Well, representatives of the bancassureurs and representatives of the insurers were there. The lunch was all about what policies needed to

78 

C. BENOÎT

be taken with our provisions and mandatory investments, essentially from insurers more than from the bankers, to participate in the stability of the Eurozone […] [The governor told us] that we [Mutual Benefit Societies] had significant funds, and that we were in a position, through our arbitrations, to have a significant influence… Interview, Former President of the National Federation of Mutual Benefit Societies Author’s translation from French

In a way, this latter example reveals what international political economists increasingly acknowledge in their study of standards. Their ambiguity constitutes their substantive attribute (Gratz 2019). It is, in other words, a form of power largely distinct from the more explicit and directive attributes exhibited by the insurance regulatory frameworks that preceded Solvency II.

5   Conclusion Developed as a complex and sizeable array of regulatory concepts, Solvency II marked a major turning point in the European regulatory framework for the insurance industry. It introduced a number of changes based on conceptions of regulation, policy instruments and techniques that radically differed from the logic underpinning four decades of EU policy-­ making in this area. Before that, three generations of life and non-life insurance directives developed a loosely structured “fire-alarm” system of regulation designed to preventively draw attention to potentially risky situations. Breaking with this essentially directive power, Solvency II attempted to introduce a “risk-based” approach that was not necessarily more coercive but rather ingrained in everyday business practice. As such, this framework introduced a whole structure to define “risky” and “safe” situations, to set a basis for risk management strategies and reporting, and ultimately to provide shared definitions of comparability, transparency and competitiveness within the sector. While the internal contradictions within the complex framework that emerged from the policy process (Frézal 2018) and the overall impact of Solvency II on firms’ strategies (Stoyanova and Gründl 2014) are still matters of debate, it is these reticular, standardizing and normative features that clearly distinguish Solvency II from Solvency I. As the chapter showed, this statement does not imply that this unifying force developed linearly or without disputes. On the contrary, the

3  SOLVENCY II, THE EUROPEAN GOVERNMENT OF INSURANCE INDUSTRY… 

79

process that led to the development of Solvency II was highly conflictual, undetermined and contingent. But the European Commission’s initial proposal, itself a product of long-term transformative changes, has proven to be resilient.

References Benoît, C. (2019). The New Political Economy of Regulation. French Politics, 17(4), 482–499. Benoît, C., & Coron, G. (2019). Private Health Insurance in France: Between Europeanization and Collectivization. The Milbank Quarterly, 97(4), 1108–1150. Benoît, C., & Thiemann, M. (in press). Regulation. In J. Pevehouse & L. Seabrooke (Eds.), Oxford Handbook of International Political Economy. Oxford: Oxford University Press. Burkhart, T., Reuß, A., & Zwiesler, H.-J. (2017). Allowance for Surplus Funds Under Solvency II: Adequate Reflection of Risk Sharing Between Policyholders and Shareholders in a Risk-based Solvency Framework? European Actuarial Journal, 7(1), 51–88. Carpenter, D. (2010). Reputation and Power: Organizational Image and Pharmaceutical Regulation at the FDA. Princeton University Press: Princeton. CEIOPS. (2006a). QIS1—Summary Report. Brussels, CEIOPS-FS-01/06. CEIOPS. (2006b). QIS2—Summary Report. Brussels, CEIOPS-SEC-71/06S. CEIOPS. (2007). Report on QIS3 for Solvency II. Brussels, CEIOPS-DOC-19/07. CEIOPS. (2010). Task Force Report on the Liquidity Premium. Brussels, CEIOPS-SEC-34/10. Clarke, S., Hayward, T., Houtepen, J., & Serre, D. (2017). Analysis of Insurers’ First Set of Solvency and Financial Condition Reports. Milliman Research Report. Der Meij, L. (2011). Solvency II: Friend or Foe? Pensions, 16, 137–139. Doff, R. (2008). A Critical Analysis of the Solvency II Proposals. The Geneva Papers, 33, 193–206. Doff, R. (2016). The Final Solvency II Framework: Will it Be Effective? The Geneva Papers, 41, 587–607. EIOPA. (2013). Guidelines System of Governance. Brussels, EIOPA-CP-13/08. Eling, M., Schmeiser, H., & Schmitt, J. (2007). The Solvency II Process: Overview and Critical Analysis. Risk: Management and Insurance Review, 10(1), 69–84. European Commission. (1997). Report to the Insurance Committee on the Need for Further Harmonisation of the Solvency Margin. Brussels, COM (97) 398 Final. European Commission. (1999). Révision de la position financière globale d’une entreprise d’assurance—Exercice Solvabilité II. Brussels, MARKT/2095/99.

80 

C. BENOÎT

European Commission. (2003). Design of a Future Prudential Supervisory System in the EU—Recommendations by the Commission Services. Working Paper, Brussels, MARKT/2509/03. European Commission. (2004a). Solvency II: Organization of Work, Discussion on Pillar I Work Areas and Suggestions of Further Work on Pillar II for CEIOPS. Working Paper, Brussels, MARKT/2543/04. European Commission. (2004b). The Draft Second Wave Calls for Advice from CEIOPS and Stakeholder Consultation on Solvency II. Working Paper, Brussels, MARKT/2515/04. European Commission. (2015). Solvency II Overview: Frequently Asked Questions. EC Fact Sheet. Retrieved April 30, 2019, from http://europa.eu/ rapid/press-release_MEMO-15-3120_en.htm. FFSA. (2006, July). Letter to CEIOPS Chair on QIS2. Paris: FFSA. Retrieved May 1, 2019, from https://eiopa.europa.eu/Publications/QIS/LettreHenrikBjerreNielsen21juillet06.pdf. Foroughi, K. (2012). Market-consistent Valuations and Solvency II: Implications of the Recent Financial Crisis. British Actuarial Journal, 17(1), 66–67. François, P. (Ed.). (2020a). Solvabilité 2: Nature, genèse, consequences. Paris: Presses de Sciences Po. François, P. (2020b). La longue marche de Solvabilité I.  In P.  François (Ed.), Solvabilité 2: Nature, genèse, consequences. Paris: Presses de Sciences Po. François, P. (2020c). Tabula Rasa. In P.  François (Ed.), Solvabilité 2: Nature, genèse, consequences. Paris: Presses de Sciences Po. François, P. (2020d). Négocier dans le Maelström. In P. François (Ed.), Solvabilité 2: Nature, genèse, consequences. Paris: Presses de Sciences Po. François, P., & Frézal, S. (2018). Instituer l’incohérence. Aléa et hétérogénéité au sein du secteur assurantiel. Sociologie du travail, 60(1), 1–23. Frézal, S. (2018). Solvency II is Not Risk-based: Could It Be? Evidence from Non-Life Calibrations. North American Actuarial Journal, 22(3), 365–379. Gatzer, N., & Wesker, H. (2012). A Comparative Assessment of Basel II/III and Solvency II. The Geneva Papers, 37, 539–570. Gordy, M., Heitfield, E., & Wu, J. (2014). Risk-Based Regulatory Capital and the Basel Accords. In A.  Berger, P.  Molyneux, & J.  Wilson (Eds.), The Oxford Handbook of Banking. Oxford: Oxford University Press. Gratz, J.-C. (2019). The Power of Standards. Transnational Hybrid Authority and the Globalization of Services. Cambridge: Cambridge University Press. Hay, C. (2014). Neither Real Nor Fictitious But ‘As If Real’? A Political Ontology of the State. British Journal of Sociology, 65(3), 459–480. IFA. (2016). Solvency 2—Health Insurance. London: Institute and Faculty of Actuaries. Jullien, B., & Smith, A. (Eds.). (2014). The EU Government of Industries: Markets, Institutions and Politics. London: Routledge.

3  SOLVENCY II, THE EUROPEAN GOVERNMENT OF INSURANCE INDUSTRY… 

81

Kean, S. (2009). How has Solvency II been Affected by the Financial Crisis and How Will it Affect Risk Management Among Insurers? Journal of Risk Management and Financial Institutions., 3(1), 7–10. King, D., & Stears, M. (2011). How the U.S State Works: A Theory of Standardization. Perspective on Politics, 9(3), 505–518. KPMG. (2002). Study into the Methodologies to Assess the Overall Financial Positions of an Insurance Undertaking from the book Perspective of Financial Supervision. Brussels, ETD/ BS-3001/C/45. Laas, D., & Siegel, C. (2016). Basel III versus Solvency II: An Analysis of Regulatory Consistency under the New Capital Standards. The Journal of Risk and Insurance, 84(4), 1231–1267. Lall, R. (2012). From Failure to Failure: The Politics of International Banking Regulation. Review of International Political Economy, 19(4), 609–638. Linder, U., & Ronkainen, V. (2004). Solvency II: Towards a New Insurance Supervisory System in the EU. Scandinavian Actuarial Journal, 6, 462–474. Marano, P., & Siri, M. (Eds.). (2017). Insurance Regulation in the European Union. New York: Springer. Masse, E. (2016, November 25). Solvency II’s impact on French Healthcare Mutual. Paris: Paper Presented at the AIM Conference. McCulloch, C. (2009, April 19–22). Solvency II Update. Sydney. Paper presented at the Biennial Convention of the Institute of Actuaries of Australia. Müller, H. (1997). Solvency of Insurance Undertakings. Brussels: Report by the Conference of Insurance Supervisory Services of the Member States of the EU. Pool, B. (1990). The Creation of the Internal Market in Insurance. Brussels: European Communities. Quaglia, L. (2011). The Politics of Insurance Regulation and Supervision Reform in the European Union. Comparative European Politics, 9(1), 100–122. Rae, R., Barett, A., Brooks, M., et al. (2017). A Review of Solvency II: Has it Met Its Objectives? British Actuarial Journal, 23(4), 1–72. Ralph, O. (2016, January 3). How Solvency II Works. Financial Times. Sandström, A. (2006). Solvency: Models, Assessments and Regulation. Boca Raton: Taylor & Francis Group. Schmidt, V., & Radaelli, C. (2004). Policy Change and Discourse in Europe: Conceptual and Methodological Issues. West European Politics, 27(2), 183–210. Schneider, J., Oechslin, J., & von Tautphoeous, M. (2009). Solvency II: The Next Revolution. Munich RE, Presentation at the Analysts’ Dinner. Scott, J. (1998). Seeing Like a State: How Certain Schemes to Improve Human Condition Have Failed. New Heaver: Yale University Press. Sharma, P. (2002). Prudential Supervision of Insurance Undertakings. Final Report, Conference of Insurance Supervisory Services of the Member States of the EU. Smith, A. (2016). The Politics of Economic Activity. Oxford: Oxford University Press.

82 

C. BENOÎT

Steffen, T. (2008). Solvency II and the Work of CEIOPS. The Geneva Papers, 33, 60–65. Stoyanova, R., & Gründl, H. (2014). Solvency II: A Drivers for Mergers and Acquisitions? The Geneva Papers, 39, 417–439. Swarup, A. (2012). A Well-Intentioned Folly: The Macroeconomic Implications of Solvency II. Economic Affairs, 32(3), 17–23. Thomson, S., & Mossialos, E. (2002). Voluntary Health Insurance in the European Union: A Critical Assessment. International Journal of Health Services, 32(1), 19–88. Thorburn, C. (2004, February). On the Measurement of Solvency of Insurance Companies. World Bank Policy Research Working Paper, no 3199. Treib, O., Bähr, H., & Falkner, G. (2007). Modes of Governance: Towards a Conceptual Clarification. Journal of European Public Policy, 14(1), 1–20. Tsingou, E. (2008). Transnational Private Governance and the Basel Process: Banking Regulation and Supervision, Private Interests and Basel II.  In J.-C.  Graz & A.  Noelke (Eds.), Transnational Private Governance and its Limits (pp. 58–68). Routledge: London. Van Hulle, K. (2005). Solvabilité II: Une approche ‘risquée’? Revue d’économie financière, 80, 243–252. Vandenabeele, T. (2014, September–October). Solvency II in a Nutshell. Revue bancaire et financière/Banken Financiewezen, n°5. Zhang, V. (2014). Uncalculated Risks: The Transformation of Insurance, the Erosion of Regulation and the Economic and Social Consequences. Toronto: Canadian Scholars’ Press.

PART II

The Impact of the European Union Private Health Insurance

CHAPTER 4

The Uncertain and Differentiated Impact of EU Law on National (Private) Health Insurance Regulations Philippe Martin and Marion Del Sol

1   Introduction Private health insurance (PHI) is fairly well developed in Europe (Lewalle 2006), though in a very heterogeneous way. National backgrounds and environments play a prominent role (Sagan and Thomson 2016). Comparative studies show that the scope of private health coverage in each country—assessed as a share of the population covered by a private/ voluntary plan—varies widely; the function or role of PHI, which can be substitutive, duplicative, complementary (user charges or services) or supplementary, depends on the very structure and organization of the statutory health insurance scheme or national health system in each country (see Fig.  4.1). Health insurers also operate under various legal forms:

P. Martin (*) Centre for Comparative Labour Law and Social Security, Université de Bordeaux, CNRS, Pessac, France e-mail: [email protected] M. Del Sol Intitut de l’Ouest : Droit et Europe, Université de Rennes 1, Rennes, France e-mail: [email protected] © The Author(s) 2021 C. Benoît et al. (eds.), Private Health Insurance and the European Union, https://doi.org/10.1007/978-3-030-54355-6_4

85

86 

P. MARTIN AND M. DEL SOL

Duplicative PHI

Substitutive PHI

Complementary PHI

Supplementary PHI

Private health Provides access to Covers specific categories insurance that services that are already of the population who can complements coverage Generally offers available within the “opt out” of the statutory of government/social services that are publicly financed health scheme (employees with a insured services by not covered under insurance scheme salary exceeding the covering all or part of the statutory (presumably affording contribution assessment the residual costs not scheme faster access, greater ceiling; self-employed otherwise reimbursed choice, or other persons; civil servants) (e.g., cost sharing, amenities) co-payments)

Fig. 4.1  Overview of the different functions of private health insurance (PHI). (Source: authors)

for-profit companies and non-profit organizations (mutual benefit societies, provident institutions, public institutions, etc.), and this has often led European countries to adopt and apply specific regulations, depending on the legal status and purpose of these different sorts of insurers. The generic term “private health insurance” or “voluntary health insurance” includes a wide variety of actors and techniques. In fact, PHI is a galaxy. In the light of this complex and heterogeneous reality, European Union law appears to be a simplified or simplistic framework. PHI is regarded as non-life insurance under the EU legislation—the non-life Insurance Directives—which aims to harmonize national legislations in order to establish an internal market for insurance, governed by the principles of freedom of establishment and freedom to provide services (this includes solvency rules). Health insurance is considered as an economic activity and health insurers as undertakings, which are the criteria for the applicability of EU competition law (Thomson and Mossialos 2007). So there is a kind of European “legal matrix” which shapes voluntary health insurance and imposes its own rules over national regulations. At first glance, one could think that when private bodies or actors provide healthcare coverage, this leads automatically to a uniform application of European economic law, that is, the non-life Directives and competition law (art. 101 to art. 109 TFEU). From this point of view, a clearly identifiable impact of EU law is to be expected, according to the model: private health insurances imply a free and competitive market. In other

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

87

words, though the EU law in itself does not command the marketization of healthcare, it tends to impose the application of market and competition rules when health coverage is provided by private insurers. Nevertheless, a closer examination of the situation leads to the observation that there is no uniform application of EU law. European law leaves a margin of discretion to the Member States so that they can arrange the health insurance market and limit competition to a certain extent. Sagan and Thomson (2016) note: “The EU-level regulatory framework created by the non-life Insurance Directive imposes restrictions on the way in which governments can intervene in markets for PHI. There are areas of uncertainty in interpreting the Directive, particularly with regard to when and how governments can intervene to promote public interests. As in most spheres of EU legislation, interpretation largely rests on European Court of Justice (ECJ) case law, so clarity may come at a high cost and after considerable delay”. This chapter seeks to analyse and discuss the impact of EU law on private health insurance at national level. From a legal point of view, the question of the impact or the influence of EU law on national legislation is rather classical. Due to the supranational nature of EU law, an impact can be observed and assessed when national law is forced to transform or adapt itself through a legal process such as the transposition of a directive or the application of an ECJ ruling. This is the “vertical” or “top down” dimension of European integration. Nevertheless, in this particular area, some national arrangements may challenge the EU legal framework and produce a kind of feedback effect. In many cases, the confrontation between national regulation (of the PHI market) and EU law is provoked by a stakeholder that challenges this regulation and aims to impose perfect competition. It happens that the outcome of this process is the evolution of EU law itself (through ECJ case law). Two main ideas will be developed here. First, though EU law appears to be a “matrix” for PHI activity, it is actually rather flexible and open to interpretations so that there is latitude for national regulation of the PHI market (Sect. 2). This, in turn, generates some uncertainty. Second, there are various forms of arrangements at national level as far as health coverage is concerned, and this is why the impact of EU law is differentiated (Sect. 3). In some countries, the frontier between the public healthcare system and the private health insurance market is well defined and quite impervious. This is generally the case when private health insurance plays a duplicative role, according to pure market mechanisms: people purchase private health insurance in order to get better and faster access to healthcare; they

88 

P. MARTIN AND M. DEL SOL

have a choice among a plurality of insurers who offer and supply health coverage and services in a competitive environment. In these countries— Spain, for instance—the impact is rather “soft” insofar as the Member States only have to make sure that private health insurers comply with the provisions of the EU Insurance Directives. In other countries, private health insurance is integrated into the universal statutory scheme (Belgium, Czech Republic, Germany, the Netherlands and Slovakia) or is closely connected to it and plays a complementary role in social protection (France, Ireland and Slovenia). These are the two faces of health coverage marketization: the adoption of “the methods and values of the market” in the non-profit sector to make it “more market-like in their actions, structures, and philosophies” (Eikenberry and Kluver 2004); the idea of using market forces to finance and provide health services. Those kinds of arrangements are potentially more problematic from the EU law perspective.

2   Health Insurance and the Internal Market: The European Matrix Examining the impact of European Union law on national legislation is common practice, irrespective of the field in question. However, the extent of this impact largely depends on the breakdown of competencies between the European Union and Member States in the field under consideration. Health insurance is undeniably part of social policy, which is dealt with in Title X of the Treaty on the Functioning of the European Union (hereinafter, TFEU). In particular, this Title provides that with respect to social security and the social protection of workers, the Union shall support and complement the activities of the Member States (art. 153§1, c). However, there are two constraints on the action of the EU in this domain. Firstly, while European legislation may implement measures designed to encourage cooperation and the exchange of information and best practice, it cannot implement any harmonization measures of the laws and regulations of the Member States (art. 153§2 a); on the other hand, the provisions adopted “shall not affect the right of member States to define the fundamental principles of their social security systems and must not significantly affect the financial equilibrium thereof” (art. 153§4). From this it follows that the configuration of social protection systems falls within the remit of Member States, as stated very clearly by the Court

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

89

of Justice of the European Union (hereinafter the CJEU) in 1987, in its Duphar ruling: “Community law does not detract from the powers of member states to organize their social security systems”.1 The Court has had the opportunity to reiterate the rules of competency in a number of rulings, including the Snares ruling, which states the following: “in the absence of harmonization in social security matters, the Member States remain competent to define the conditions for granting social security benefits”.2 From this it follows that Member States decide on the characteristics of their social security systems in every respect; this leads to a very high degree of diversity in these systems. National choices in this respect are therefore not constrained by the European Union, within which there are not only national health systems, but also social insurance schemes; these may exist side-by-side with various forms of privatization, including private insurance schemes. Simply put, it must be borne in mind that the fact that national options in terms of social security organization and protection, and, more particularly for our interests, health insurance, are not predetermined at EU level (Lear et al. 2010). This means that the degree of vertical integration provided by EU law is relatively minor, although it does have an impact in the event of the existence of private insurance schemes. However, horizontal integration under EU law is worth examining more closely. Indeed, integration may be indirect, potentially brought about by compliance with “transverse” EU rules that, in theory, apply irrespective of the activity in question. This is particularly the case for internal market law, based on the freedom to provide services, and competition law, the provisions of which are enshrined in the TFEU. This law has led to the creation of a kind of matrix, with its own criteriology (Mossialos et al. 2010). 2.1  The Relatively Minor Impact of EU Insurance Law on Vertical Integration Member States retain most responsibility for deciding how to organize and regulate those areas of activity not covered by EU harmonization in

 ECJ case 28/82, Duphar, 7 February 1984, ECR 1984, p. 523.  ECJ case C-20/96, Snares, 4 November 1997, ECR 1997, p. I-6057. See also ECJ case C-158/96, Kohll, 28 April 1998, ECR 1998, p. I-1931; ECJ case C-247/90, Xhymshiti, 18 November 2010, ECR 2010, p. I-11845. 3  This type of Directive relates to certain sectors of business (e.g. transport, banking and financial markets). 1 2

90 

P. MARTIN AND M. DEL SOL

the form of sector-specific directives.3 This applies, for instance, to social protection, in view of how competencies are allocated between the EU and Member States (see above). With respect to health insurance, this means that the EU does not predetermine what should be covered by social insurance (social security)4 or private insurance. Legally speaking, it is up to Member States to establish “on-market” and/or “off-market” health insurance and the scope of each type (Thomson and Mossialos 2006, 2010). In this respect, national arrangements are often hybrid in nature; this raises specific problems. Indeed, while the EU does not have any remit to harmonize domestic social security, the same is not true of the insurance business, which falls within the scope of the internal market. To create the internal market and make the freedom to establish and provide services a reality, the EU has adopted great many directives, in particular sector-­ specific directives. As a result, a body of EU insurance law has gradually come into existence. It would therefore be tempting to conclude that, in countries that allow private health insurance, these “Insurance” Directives shape the legal environment applicable to this business, so much so that they constitute vertical integration.5 However, this sweeping conclusion does not withstand an analysis of the limits established by the directives themselves. 2.1.1

S tatutory Social Security Schemes: Not Covered by Vertical Integration The Solvency II Directive excludes “insurance forming part of a statutory system of social security” (art. 4) from its scope. As a result, the deployment of social insurance to cover sickness falls outside the sphere of EU competition law (see below). 4  Member States also decide on the procedures governing state payment for care. Some countries, such as the UK, have opted for a national health system that provides direct access to care; others, such as France, have opted to set up social insurance, covering part of the cost of care. 5  Several generations of “Insurance” Directive have been adopted since 1973. They were abrogated by Directive 2009/138/EC of the European Parliament and of the Council of 25 November 2009 on the taking-up and pursuit of the business of Insurance and Reinsurance. Known as the Solvency II Directive, this Directive is now a kind of European Insurance Code. Hereafter, reference will be made to this 2009 legislation, of which many provisions are taken from previous Directives (particularly those of 1973 and 1992). See Chap. 3.

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

Insurance Directives

EU competition law

Statutory social security schemes

Out of the scope regardless the organization

Out of the scope regardless the organization

Private health insurance

All organizations (for-profit or non-profit insurance undertakings)

All organizations (for-profit or non-profit undertakings)

91

Fig. 4.2  Organic scope of EU economic law. (Source: authors)

In a case involving France, the issue of this exclusion was brought before the CJEU. In its Garcia6 ruling, the Court stated that one of the law’s provisions clearly and precisely specified that the Insurance Directive excluded from its scope of application “not merely social security organizations (undertakings or institutions) but also the types of insurance and operations which they provide in that capacity”. In its view of national competency to organize social security systems in the light of the scope of application of “Insurance” Directives, the Court ruled that two health insurance schemes could coexist in Member States: a private-sector scheme, subject to “Insurance” Directives, alongside a social security scheme that was not subject to the rules of the internal market (“Insurance” Directives and competition law) (see Fig. 4.2). As a result, domestic arrangements for health insurance play a determining role in the extent to which EU law achieves vertical integration. The more the national system leaves room for private health insurance, the greater the vertical integration; on the other hand, in countries where social security covers most of the cost of healthcare, vertical integration is weak. From this it follows that the definition of a statutory social security scheme is key. No such definition is to be found in the “Insurance” Directives, which are silent on the matter. However, EU jurisprudence in the field of competition law does provide some definitional aspects. Indeed, the Court referred to this jurisprudence in its Garcia ruling. It cites the landmark Poucet and Pistre7 rulings, which define schemes  ECJ case C-238/94, García, 26 March 1999, ECR 1996, p. I-1673.  ECJ joined cases C-159/91 and C-160/91, Poucet and Pistre, 17 February 1993, ECR 1993, p. I-637. 6 7

92 

P. MARTIN AND M. DEL SOL

founded on the principle of solidarity and to which affiliation is made mandatory in order to ensure application of the principle of solidarity as well as that of financial equilibrium, as being a solely social (as opposed to economic) activity. The actual wording of the Poucet and Pistre ruling states that these schemes have a social purpose, and fulfil the solidarity principle, because they are intended “to provide cover for all the persons to whom they apply, against the risks of sickness, old age, death and invalidity, regardless of their financial status and their state of health at the time of affiliation”. In terms of health insurance, solidarity is expressed in terms of identical services for all social security beneficiaries, with contributions proportional to income, ensuring that the less well-off are not deprived of any social cover that their state of health may require.

The off-market nature of “operations” by provident and mutual benefit societies

Right from the first “Insurance” Directive (1973), a distinction is drawn between insurance activities and “operations of provident and mutual benefit institutions whose benefits vary according to the resources available and in which the contributions of the members are determined on a flat-rate basis” (art. 2§2, cited in article 5 of the Solvency II Directive). Indeed, all such “operations” are excluded from the scope of application of provisions covering insurance other than life insurance. There is no explanation of this exclusion in the recitals to the directives, even though it results in part of the activity of a certain category of insurers qualifying as being “off-market”. In view of the definitional elements set out in the legislation and the insurers in question, it would appear that this is a reference to services provided on the grounds of social action (e.g. one-off assistance to a beneficiary having to cope with an exceptionally large excess to be paid for care that they cannot afford due to their low income).8 Unlike insurance cover, the insurers in question make no commitment to their policyholders, because the services counting as “operations” may be granted only if resources are available. Moreover, policyholders’ contributions to the funding of (continued) 8  Traditionally, provident and mutual benefit institutions have social action funds financed by an annual budget that draws on the contributions received. These funds allow aid, assistance and additional services to be provided to policyholders in the circumstances covered by the insurance policy in question.

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

93

(continued)

operations is on a flat-rate basis; this means that they do not depend on actuarial criteria (the degree of risk the policyholder represents). However, the definitional criteria do not restrict “operations” to the field of social action. Indeed, “operations” are not defined in terms of their nature or purpose, but in terms of how they are funded. It is therefore in theory possible for part of the activity of insurance companies to fall outside the scope of regulations applicable to the insurance sector, including some activities that are somewhat complementary in nature to the insurance business itself. A reform undertaken in Belgium in 2010 made express use of this distinction between the business of insurance and “operations” to keep part of the top-up healthcare activity carried out by mutual benefit societies out of the insurance market (see Chap. 6).

2.1.2 A Lack of Harmonization in Insurance Policy Law The subjects of the harmonization brought about by the “Insurance” Directives offer some insight into the scope of expression of vertical integration. In substance, the purpose of the “Insurance” Directives is to harmonize the conditions for competition between companies present on the market, securing the commitments made by these operators with respect to policyholders and allowing the latter to have all the information required to benefit from the existence of competition. EU law governing insurance companies has thus been built up over time and been applied throughout all Member States via directives. The law governing the terms of exercise of the business of insurance is highly integrated, in order to achieve an internal insurance market. However, insurance policy law has for the most part not been included in the harmonization process. Substantive law covering insurance policies derives mainly from domestic law. The lack of vertical integration regarding the contents of insurance policies grants Member States the scope to influence this content (or choose not to do so).9 Each country decides on the regulatory framework governing all or part of the products that may be offered on the market by insurance companies. Applying this to the 9  “Horizontal” EU Regulations do however cover some aspects of the contractual relationship in insurance policies (information, non-discrimination, etc.).

94 

P. MARTIN AND M. DEL SOL

realm of health insurance, state interventionism is a matter of domestic policy; this may depend largely on the extent to which private health insurance is, explicitly or implicitly, assigned a systemic role in paying for care. In France, for instance, private health insurance is mostly involved as a top-up to the statutory health insurance scheme (social security); it is therefore hardly surprising that regulations structure the relationship between these two levels of payment via configuration constraints that impact health insurance policies.10

When private insurance replaces social security—a special case

EU regulations covering insurance do however apply to one type of health insurance policy: policies taken out on an opt-in basis that wholly or partly replace health insurance provided by a social security scheme. This instance of so-called “substitute” insurance generally applies only to a tiny fraction of the population. For instance, in Germany, employees whose income exceeds a given ceiling may approach a private-sector insurer rather than a health insurance fund (Busse and Blümel 2014). According to recital 84 of the Solvency Directive, “the particular nature of such health insurance distinguishes it from other classes of indemnity insurance and life insurance insofar as it is necessary to ensure that policy holders have effective access to private health insurance or health insurance taken out on a voluntary basis regardless of their age or risk profile”.11 This may legitimize the maintenance of legislation that is binding in quantitative terms, designed to ensure that the cover in question is a genuine substitute (e.g. including an obligation of non-discrimination). Such legal measures must be made binding, in order to protect the general interest with respect to this branch of insurance in the Member State in question (Solvency Directive art. 206). The State in question may require the insurance policy and the insurer’s practices to comply with the provisions (continued)  See Chap. 7.  On the difficulties of clearly defining the notion of substitute insurance, see Sagan and Thomson (2016, p. 85) and Mossialos et al. (2010). 10 11

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

95

(continued)

designed specifically to protect the general interest,12 one aspect of which is that nobody should be deprived of health insurance. Again, in Germany, in the case of substitute insurance, this means, for instance, that insurers cannot increase premiums on the grounds of age or terminate the policy (i.e. their commitment is for life).

From this analysis, it may be concluded that in substantive terms (i.e. policy content, rules governing selection and tariffs, etc.), vertical integration in terms of insurance law is, ultimately, relatively slight. With respect to health insurance, however, closer attention needs to be paid to indirect integration mechanisms. Indeed, transverse EU rules may impact national decisions; in some cases, they may even offer Member States opportunities when it comes to choosing how to organize their social protection system. This is the case not only in view of fundamental rights (e.g. the principle of equal treatment13) but also in view of EU competition law. 2.2  Horizontal Integration via EU Competition Law: Uncertain Criteria and Malleable Instruments Competition law makes reference to a body of rules of which the purpose is to prevent competition between companies that intervene or seek to intervene on the same market from being hindered, restricted or distorted (art. 101-113, TFEU), for instance, through the abuse of dominant position or via state aid granted to some operators. Consequently, its principles have a very broad influence on all sectors of activity. They thus constitute horizontal integration processes with an impact in the field of health insurance. However, understanding these processes and their real impact is complicated, for two reasons in particular. Firstly, the precise outline of the 12  On this concept, see the interpretative communication by the Commission, C (1999) 5046 of 2 February 2000 “Freedom to provide services and the general interest in the insurance sector”. 13  For instance, in a landmark ruling, the CJEU ruled against sex-based discrimination in insurance premiums and services. ECJ case C-236/09, Association belge des consommateurs Test-Achats, 1st March 2011, ECR 2011, p. I-773.

96 

P. MARTIN AND M. DEL SOL

criteria governing the legal spaces created by EU competition law (see below) is unclear. Secondly, when Member States configure their health insurance systems, they rarely provide a detailed explanation of how these fit into the spaces in question; the issue of the compatibility of domestic provisions with EU competition law only tends to emerge in the event of litigation. 2.2.1

 egal Spaces Created by Competition Law: A Broad Spectrum L for (Non)-marketization of Health Insurance Introductory identification of the legal spaces created by EU competition law. Irrespective of the sector of activity in question, examination of CJEU jurisprudence reveals three legal spaces within the regulatory environment (Del Sol 2016; Mossialos et al. 2010) (see Fig. 4.3). Positive law has created an “off-market” space for activities that are not provided by “undertakings” within the meaning of EU competition law. Activities that are exclusively social in nature—a category that has emerged in the Court’s jurisprudence—fall within this space. As such, the latter therefore falls outside the scope of application of TFEU articles 101-113, that is, it is not subject to EU competition law; from this it follows that a State may grant a single institution or body a monopoly on managing this space. The legal space in question may be understood as having come into existence by default, by opposition with the “on-market” space.

EU competition law

Free competition

Competition adjustment (monopolistic situation, selective benefits…)

Undertaking (entity engaged in an economic activity)

Undertaking engaged in an economic activity) and entrusted with the operation of services of general economic interest

Health insurance not solidarity-based

Health insurance with social function and impregnated with considerations of solidarity that justifies regulating market forces

Out of the scope of EU competition law

Activity fullfilling an exclusively social function

Statutory social security schemes structurally solidarity- based and under state control

Fig. 4.3  The three legal spaces resulting from the EU competition law. (Source: authors)

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

97

Activities provided by undertakings (i.e. activities that are not exclusively social in nature, as it were) fall within an “on-market” space. This space therefore falls within the scope of application of TFEU articles 101-113 and is thus subject to EU competition law and its principle of free and fair competition. It may be understood as being the legal space within which activities should theoretically take place, since the European Union promotes the market economy and aims to construct a huge internal market. This quite clearly includes the insurance sector, as the CJEU stated in 1987.14 Within this on-market space, there is also a legal enclave: TFEU articles 106 and 107 specify both exceptions and restrictions to competition rules. For instance, subject to certain conditions, TFEU article 106§2 allows restrictions to market rules to be implemented for “undertakings entrusted with the operation of services of general economic interest” (hereinafter, SGEIs). Such mechanisms provide a way to adjust competition rules for social purposes via what may be referred to as social instruments of competition law (Driguez 2006). Defining the scope of the “on-market” space. The “on-market” legal space is the space that applies in principle. This space is defined by the scope of application of EU competition law, the implementation of which applies only to undertakings. However, primary legislation does not precisely define the concept of “undertaking”. As a result, it has been up to the Court of Justice to define what the EU means by “undertaking” during the course of a number of disputes, a significant proportion of which pertain to the field of social protection. According to the foundational Höfner & Elser ruling in 1991,15 “in the context of competition law […], the concept of an undertaking encompasses every entity engaged in an economic activity…” From this it follows that the existence of economic activity can be used to determine whether the activity in question falls within the scope of EU competition law (the “on-­market” space). The European Union definition of “undertaking” supplied by the Court of Justice considers the characteristics of the activity engaged in rather than those of the entity, body or structure engaging in it. From the Höfner ruling onwards, the Court of Justice has remained agnostic as to the legal nature and form of entities engaging in an activity

14  ECJ case 45/85, Verband der Sachversicherer versus Commission, 27 January 1987, ECR 1987, p. 405. 15  ECJ case C-41/90, Höfner and Elser, 23 April 1991, ECR 1991, p. I-1979.

98 

P. MARTIN AND M. DEL SOL

when it comes to determining whether or not an “undertaking” within the meaning of EU law exists. The organic criterion, that is, looking at the nature of the body in question, cannot therefore be used to determine economic activity in this respect. From this it follows that a non-profit body may be deemed to constitute an “undertaking”. The same may also apply to a mutual benefit societies or a statutory body. The Court’s jurisprudence is based on a functional approach, in which economic activity is the kernel of the definition of an “undertaking” (Dunne 2010).16 Typically, the Court rules that the concept of an undertaking covers any entity engaged in an economic activity that consists in offering goods or services on a given market.17 In using the term “market”, the Court is making reference to situations involving competition. However, defining the existence of economic activity does not necessarily mean that an actual situation involving competition has been determined; a potential situation may suffice. An activity will be defined as economic if it is exercised, or is liable to be exercised, by a private actor for profit. In other words, the reasoning is deductive. For instance, in its Fédération Française des Sociétés d’Assurance (FFSA)18 ruling, the Court states that “the mere fact that the CCMSA is a non-profit-making body does not deprive the activity which it carries on of its economic character” to the extent it competes with life assurance companies. This has led the Court of Justice to clarify that “the fact that the offer of goods or services is made without profit motive does not prevent the entity which carries out those operations on the market from being considered an undertaking, since that offer exists in competition with that of other operators which do seek to make a profit”.19 In theory, it should be checked whether the activity in question is or could be comparable to the activity of a private-sector, for-profit undertaking, or any activity liable to be exercised by such an entity. Generally speaking, the Court adopts a comparative approach, in particular by seeking to verify whether the activity in question is actually managed (or may have been managed recently) by private sector, for-profit entities in other Member States. 16  Following this approach, a single structure may be defined as an undertaking for one type of activity in which it engages and not for another. 17  ECJ case C-118/85, Commission versus Italy, 16 June 1987, ECR 1987, p. 2599; ECJ case C-475/99, Ambulance Glöckner 25 October 2001, ECR 2001, p. I-8089. 18  ECJ case C-244/94, FFSA 16 November 1995, ECR 1995, p. I-4863. 19  ECJ case C-49/07, MOTOE, 1st July. 2008, ECR 2008, p. I-4013; ECJ case C-222/04, Cassa di Risparmio di Firenze, 10 January 2006, ECR 2006, p. I-289.

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

99

Criterion for defining the “off-market” space. Activities that are solely social in nature do not fall within the scope of competition law: since they do not entail any economic activity, they cannot be defined as an “undertaking” and therefore find themselves in the “off-market” space. This makes it all the more important to examine what constitutes an “exclusively social activity”: if no such activity can be identified, the default conclusion is that the activity is economic in nature. This question is particularly acute in the field of social protection and indeed this makes up a substantial part of the Court’s jurisprudence. The founding rulings are the Poucet and Pistre rulings (see above) on independent workers who refused to pay contributions to social security funds, taking the view that for their mandatory basic social insurance, they should be entitled to approach a private insurance company, rather than suffer from the dominant position occupied by social security. The Court excludes from competition law “the organizations involved in the management of the public social security system that fulfil an exclusively social function”. While the social aim of the social protection scheme being challenged is a prior condition for this definition, it is not a sufficient condition, because the activity must be “exclusively social”. The fact is that the defining criteria have not always emerged very clearly from the Court’s jurisprudence. However, the Kattner ruling handed down in 200920 now offers a greater degree of certainty.21 Indeed, in this ruling, the CJEU specifies that, when a social insurance scheme has a social aim, “it remains to be examined, in particular, whether that scheme can be regarded as applying the principle of solidarity and to what extent it is subject to supervision by the State, given that these are factors that are likely to preclude a given activity from being regarded as economic”.22  ECJ case C-350/07, Kattner Stahlbau GmbH, 5 March 2009, ECR 2009, p. I-1513.  The European Commission’s interpretative communications are also useful when it comes to identifying exclusively social activities. In this respect, §20 of the 2016 Communication on the concept of State aid (2016/C 262/01) specifies the following: “Solidarity-based social security schemes that do not involve an economic activity typically have the following characteristics: (a) affiliation with the scheme is compulsory; (b) the scheme pursues an exclusively social purpose; (c) the scheme is non-profit; (d) the benefits are independent of the contributions made; (e) the benefits paid are not necessarily proportionate to the earnings of the person insured; and (f) the scheme is supervised by the State”. 22  “The fact that the body provides insurance services directly does not, of itself, affect the purely social nature of that function, in so far as it does not affect either the solidarity inherent in that scheme or State supervision of it”. See also ECJ case C-218/00, Cisal, 22 January 2002 concerning Italian legislation covering “workplace accidents and professional diseases”, ECR 2002, p. I-691. 20 21

100 

P. MARTIN AND M. DEL SOL

Criteria and conditions for regulating the “on-market” space with respect to SGEIs. Within the field of economic activities subject to competition law, competition rules may be adjusted to allow services in the general interest to be deployed. In other words, the TFEU allows the “off-­market” space to be regulated, subject to certain conditions. To do this, the “social instruments” of competition law specified in TFEU articles 106 and 107 may be used (Driguez 2006). With respect to principle and also, more particularly, when it comes to disputes, the legitimacy of such adjustments to the rules of competition (EU law) depend on being able to identify a general-interest mission, entrusted to an undertaking. Doing so is difficult in the absence of certainty as to the precise definition of SGEIs. Indeed, SGEIs are not defined in primary law or derivative law. However, the European Commission has specified that SGEIs are economic activities carrying out general-interest missions that, in the absence of any State intervention, would not be carried out at all, or would be carried out by the market in different conditions in terms of quality, safety, accessibility, equality of treatment and universal access. It is therefore not surprising to discover that despite the existence of extensive jurisprudence, the law in this respect is not fully predictable, since identifying an SGEI is based on examining a body of evidence which may be difficult to weigh. SGEI-related EU litigation in the field of social protection highlights the fact that the argument rests mainly on the solidarity factors imposed on the operators that manage the schemes in place. Indeed, in most cases, it is argued that if the scheme is characterized by a degree of solidarity imposing significant obligations on the managing operator, the service it provides cannot be delivered at the same price as other operators to whom these constraints do not apply. It is therefore necessary to demonstrate that shouldering the general-interest mission may constitute a form of competitive disadvantage, leading to the service provided being less competitive, because its cost is higher in order to take into account solidarity-based considerations. One illustration of this is the AG2R ruling, on a case pertaining to a collective agreement concluded for the artisanal bakery branch that established a mandatory “health costs” scheme.23 To identify the existence of an SGEI, the Court referred to a number of indicators: funding via fixed-rate contributions such that the rate was not proportional to the risk shouldered and thus

 ECJ case C-437/90, AG2R, 3 March 2011, ECR 2011, p. I-973.

23

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

101

that age, state of health, and the particular risks inherent in any specific job were not taken into consideration;24 the nature of the benefits and the extent of the cover not being proportionate to the amount of contributions paid; and the existence of “free” entitlements that did not directly entail contributions (e.g. temporary maintenance of cover after the termination of a contract or the death of the employee). Secondly, it has to be demonstrated that the regulatory measure does not do more than what is strictly necessary to fulfil the mission (control of proportionality); failing this, an operator may be deemed to enjoy an unlawful competitive advantage with respect to other operators present on the market. In other words, any restriction on competition, in particular if it takes the form of financial compensation, must be very carefully calibrated. The idea is that market intervention should satisfy the general-interest mission, in other words “counter” competition with regulations designed to achieve aims deemed to be indispensable, such as equal access to certain specific benefits. Primary law does not specify the nature of such arrangements,25 but does define a framework for anything that takes the form of state aid (TFEU art. 107-109). It allows restrictions to market rules favouring undertakings responsible for managing SGEIs to be implemented. Broadly speaking, these arrangements may come in one of two forms. One option is to regulate the entire market; any such regulation is thus binding on some or all of the undertakings engaged in competition on the market in question. Regulation imposes obligations on all or part of the offerors such that ultimately, general-interest considerations are satisfied (e.g. the obligation to offer certain services to any citizen approaching an operator).26 Regulation may also require undertakings to provide their products in a manner that would be incompatible with purely market requirements, for instance to correct unequal access to certain goods and services. The alternative approach concerns selective advantages favouring

 In other words, tariffs are not determined on the basis of each employee’s loss exposure.  There is also a wide range of possibilities when it comes to management modes, that is how, practically speaking, general-interest missions are taken up. Public authorities have a high degree of freedom in this respect: they may perform general-interest tasks “by their own means, without being required to enlist third-party entities” (ECJ case C-480/06, Commission versus Germany, 9 June 2009, pt 45, ECR 2009, p. I-4747). Alternatively, they may also opt to outsource the management of general-interest economic activities or entrust general-interest missions to private operators. 26  TPI case T-289/03, BUPA, 12 February 2008 (Irish case), ECR 2008, p. II-81. 24 25

102 

P. MARTIN AND M. DEL SOL

undertakings responsible for an SGEI. Here, adjusting competition rules may take the form of special management rights; these may go so far as to grant exclusive management rights, creating a monopoly to the benefit of the undertaking responsible for the SGEI. Adjustment may also take the form of financial compensation or aid, the benefit of which is reserved solely for undertakings responsible for a general-interest mission. 2.2.2

 ssessing Horizontal Integration: The Many Shades of Health A Insurance Marketization The impossibility of using EU competition law to measure horizontal integration. Legally speaking, significant differences arise depending on which of the three identified spaces a health insurance activity is assigned to. Consequently, if a State organizes health insurance in such a way that a management monopoly is established, or entrusts selective benefits to a particular body or imposes specific obligations on certain operators, the issue of compatibility with EU competition law arises—or more accurately, it could be said that the issue of compatibility is likely to arise. In other words, using EU law to gauge horizontal integration is tricky in conceptual terms, for two main reasons. Firstly, States do not necessarily situate their legislation within the field of EU competition law; the recourse to EU criteriology may only be implicit. For instance, when French law establishes a contractual clause recommending that a body providing insurance to companies in a particular professional sector should provide a high degree of solidarity, it uses the criterion of solidarity at the heart of CJEU jurisprudence with regard to SGEIs, but does not expressly acknowledge the paternity of this concept. At the same time, there are other examples in which EU law has manifestly—and indeed sometimes explicitly—been used as a resource, either to provide support for a particular arrangement or to support a reform. In such examples, this “Europeanisation” mingles strictly domestic considerations and EU legislation, the outcome of which is a complex hybridization of levels. For various reasons, to differing degrees and on different timescales, Belgium, the Netherlands and Ireland have all adopted this type of approach. Secondly, there is no systematic control of the compatibility of domestic legislation with EU law. As a result, the issue is not necessarily raised or discussed. The issue of Euro-compatibility often arises as a result of litigation. Very

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

103

generally speaking, this arises when operators ejected from a market, or believing themselves to be the victims of unfair competition, bring litigation in the form of a civil case.27 In other instances, litigation may be brought by Member States following a ruling by the European Commission, for instance during the course of the latter’s control of state aid. The phenomenon of interaction between the EU and national levels. In this field, the term “Europeanisation” is often used: EU legislation is said to produce a top-down vertical effect that “shapes” the national level. However, the field of health insurance may well serve to demonstrate the limits of this traditional assumption, since EU law does not appear to be making public or private-sector health systems more uniform or harmonized. Despite this, the impact of EU law is sometimes perceived in such a way as to give the impression that application of EU economic law is intended to achieve conformity of national systems with a European model destined to grant a large slice of the pie to private insurance and the free market. This assumption is mistaken. Indeed, when it comes to health insurance, EU law is likely to produce an impact only where there is a health insurance market; this cannot happen unless a market has actually been identified. If one works on the assumption that a market is not a “spontaneous” creation but a legal/political construction, the fact is that EU-level interference depends not only on social and historical characteristics, but also on the political decisions governing the construction of national systems in the realm of health insurance. That said, the contingent nature of the concept of economic activity should be borne in mind. “Apparently, its contents can be determined only on a case-by-case basis [by taking into account] the facts of each matter, and then deducing the existence of economic activity, without having previously defined the scope of this concept” (Bernard 2009, p. 357). This contingency is thus a source of uncertainty, as can be seen from the dispute concerning mandatory health insurance in Slovakia (see box below).

27  For an example, see Chap. 6 (Belgium case). For another example, see the challenge to aid granted by the French State to certain mutual benefit institutions for civil servants following a ruling by the European Commission on 27 July 2005.

104 

P. MARTIN AND M. DEL SOL

A topical example of the contingent nature of the concept of economic activity: the dispute over mandatory health insurance in Slovakia28

In the country of Slovakia, all health insurance bodies are limited companies, the sole purpose of which is to provide mandatory public health insurance. Slovakian residents can choose between three bodies: two private insurance companies and one public insurance company. These bodies are legally required to affiliate any resident that so requests; from this it follows that they cannot refuse to insure an individual on the grounds of their age, state of health or risk of illness. Moreover, the scheme is based on a system of mandatory contributions, the amounts of which are established by law, in proportion to beneficiaries’ income, and thus independently of the benefits used or “individual” risk. Moreover, all beneficiaries are entitled to the same minimum level of benefits, and there is a system to balance risks out between the different bodies. Lastly, these bodies are subject to special legislation. In particular, they cannot carry out any activity other than those specified by law (i.e. mandatory public health insurance) and are controlled by a regulatory agency. In the light of all these elements, the European Commission had concluded that the activity of providing mandatory health insurance was non-economic in nature, taking the view that the above characteristics were evidence of the predominance of social, solidarity-­based and regulatory aspects in this scheme. In other words, in the view of the Commission, there was an exclusively social activity due to the existing solidarity and state control being foundational elements. However, the Court did not come to the same conclusion as the Commission, due to a particular characteristic of the system put in place: the legally recognized possibility for health insurance companies to make, use and distribute profits. For the Court, “[this] possibility… is of a nature liable to compromise the non-economic character of their activity” (pt 63) since it “is, in any event, and independently of the performance of their public health insurance ­mission (continued) 28  General Court case T-216/15, Dôvera zdravotná poistʼovňa versus Commission, 5 February 2018. It should be noted that this ruling has been appealed before the CJEU (262/18 P); the appeal is still pending.

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

105

(continued)

and the state control operated, evidence of the fact that they are pursuing a for-profit goal; on this basis, the activities in which they engage on the market qualify as economic” (pt 64). The Court also emphasized the existence of a degree of competition regarding the quality and extent of the offer: with the profits they make, the bodies may freely supplement mandatory, statutory benefits with related, free benefits (e.g. better cover for some types of treatment or assistance services): “these bodies’ freedom of manoeuvre to engage in competition allows insured parties to benefit from better social protection for equivalent levels of contribution, since the supplementary benefits are offered free of charge” (in other words without the payment of any additional contributions) (pt 66). Competition therefore exists regarding the insurance’s value for money; according to the Court, far from being residual, this competition is “intense”, particularly due to the volatility of the market, with beneficiaries being able to change their affiliation every year and incited to do so by their personal assessment of the quality of service.

Even where a health insurance market exists, the impact of EU law is not uniform from one Member State to the next. As the Commission explains, Member States’ public authorities enjoy extensive room for manoeuvre in defining what they deem to constitute an SGEI, provided they observe EU law and that there are no manifest errors of appreciation.29 In view of this, interference at national level is largely dependent on certain systemic configurations. Indeed, EU competition law criteriology is such that it allows different types of appropriation, the effects of which are liable to vary widely in terms of health insurance organization and regulation (Mossialos and Lear 2012). To paraphrase one author (Bernard 2009, p. 374), Member States may set “any degree of market conditions” they like, with a view to situating the market in one of the two legal spaces made available in competition law: that of a market that is not specially regulated, or that of a regulated market, allowing general-interest 29  Guide to the application of the European Union rules on state aid, public procurement and the internal market to services of general economic interest, and in particular to social services of general interest—Commission SWD(2013) 53 final/2.

106 

P. MARTIN AND M. DEL SOL

considerations to come into play.30 So, for instance, in a field in which competition is predominant, existing legislative instruments may be brought to bear such that significant general-interest considerations, such as health protection or the correction of inequalities and the access to certain services, and off-market values such as solidarity, can be deployed on a particular market. This “explains the fluctuating nature of competition and thus of the market, depending on historical periods, geographical spheres, and above all, the goals pursued by the political and legal powers that be” (Bernard 2009, p. 374). In substantive terms, the solidarity components governing the overall workings of the social protection activity must be identified. This identification is very often expressed in terms of degree, with the Court seeking to identify and assess the extent to which the configuration rules for the scheme affect the workings of traditional insurance. In this case, the question becomes: is it possible to identify solidarity components to a degree that would legally and justifiably allow exclusive or special rights on this insurance market? The degree of differentiation from “conventional” private insurance present on the market, in which there is a direct proportionate link between the contribution and the benefit, is what makes it possible to assess implementation of the principle of solidarity and thus provide grounds for adjusting rules derived from EU competition law. Identifying the degree of solidarity in an insurance scheme involves studying the rules used to configure the proportionality that is normally determined in insurance, between the contribution and the risk on the one hand, and between the contribution and the benefit on the other hand. Despite extensive jurisprudence relating to the application of the concept of SGEIs in the field of social protection, the law is not entirely clear, since this identification uses the approach of a body of evidence, which can be complex to weigh.

3   National Arrangements, Health Coverage and Marketization Under the Treaty on the Functioning of the European Union, article 168§7, the definition of health policy and the organization and delivery of health services and medical care are prerogatives of the Member States. The so-called principle of subsidiarity applies in this area. Traditionally, the 30  This binary alternative may apply not only to an entire market, but also to particular market segments.

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

107

organization of health services and medical care at national level has been directly connected with social security schemes which provide people with (more or less) universal access to healthcare. As pointed out above, the area of social security is in principle exempted from the application of European competition law. Thus, there should be a clear and well-marked frontier between healthcare provided through social security on the one hand, and healthcare and services provided by the market on the other hand. When healthcare is provided by the market through voluntary and private health insurance, the European Insurance Directives are supposed to apply. However, as noted by Thomson and Mossialos (2010), “the European directive reflects the health system norms of the late 1980s and early 1990s, a time when boundaries between social security and ‘normal economic activity’ were still relatively well defined in most Member States. Today, these boundaries are increasingly blurred”. Indeed, some countries have chosen to introduce market forces and competition into the statutory health insurance scheme, while others use and, in a certain way, “instrumentalize” the complementary health insurance market in order to make it an actor of the national health policy. These situations may challenge the EU legal framework. 3.1  Marketization within the Statutory Health Insurance Scheme In a number of countries, public health insurance is administrated by a wide variety of operators such as public or quasi-public bodies, non-profit organizations and commercial companies, so that people—that is, “consumers”—have a choice. Theoretically, consumer choice in the health insurance market is assumed to discipline insurers to make them increase their efficiency and be responsive to consumers’ preferences. In Europe, this kind of arrangement can be observed in particular in Belgium, the Czech Republic, Germany, the Netherlands and Switzerland.31 What these national systems have in common is that private insurers are legally entrusted with public interest missions. This means that the scope, the content and the conditions of the health coverage they offer to the population are strictly regulated, so one may wonder whether there is room for market and competition rules. Actually, marketization is more or less obvious, depending on specific national arrangements.  In this chapter, only EU member States will be reviewed.

31

108 

P. MARTIN AND M. DEL SOL

3.1.1 Private Insurers Entrusted with Public Interest Missions Private actors within social security. In countries with a Bismarckian welfare tradition, private actors have been entrusted with administrating statutory health insurance. Traditionally, the social partners—that is, employees’ and employers’ representatives—as well as mutual benefit societies and other non-profit organizations, have been at the centre of the system, which is a “non-State” one (Pennings et al. 2013). According to the Bismarckian model, health insurance funds are ruled by the self-­ governance principle (Busse et  al. 2017). In most cases, though those actors are private entities, the national legislation or their own articles of association do not allow them to make a profit. There are no shareholders to reward. The Belgian, German and Czech systems are of this kind. Nevertheless, two European countries are known to have opened statutory health insurance to for-profit companies: the Netherlands and Slovakia. In Belgium, as early as the nineteenth century sickness funds were held by mutual benefit societies or trade unions. The system became compulsory in 1944 with the creation of social security, but the health insurance funds remained in the hands of the mutual societies and the unions. Currently, seven health insurers are entitled to operate: the five main Belgian mutual benefit societies,32 a national public body called Caisse Auxiliaire d’Assurance Maladie-Invalidité (CAAMI), and the railway mutual fund Société Nationale des Chemins de Fer Belges Holding (SNCBHolding). Belgian mutual benefit societies also offer complementary coverage to their members. In Germany, since the beginning of social insurance (1883), the legislation has deliberately opted for a plurality of health insurance funds and has required people to be affiliated to one of them (Kaufmann 2004). Germany has a wide variety of funds: local funds, mutual benefit societies, company funds, corporative funds and sectoral funds (agriculture, mines, railway and seafarers), but they are all autonomous bodies governed by public law.33 These funds (113 in 2017) are usually held organized within associations. Population coverage by the statutory health insurance scheme is

 Alliance nationale des mutualités chrétiennes, Union nationale des mutualités neutres, Union nationale des mutualités socialistes, Union nationale des mutualités libérales, Union nationale des mutualités libres. 33  This does not preclude the determination of whether these sickness funds are or are not undertakings in the sense of EU competition law (ECJ case C-59/12, BKK versus Zentrale zur Bekämpfung unlauteren Wettbewerbs, 3 October 2013). 32

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

109

almost universal, but some categories of people34 can opt out and purchase private health insurance that plays a substitutive role. Private plans are offered by commercial companies and also by non-profit companies under market conditions, though the German substitutive PHI has to comply with certain legal requirements (see below). The Czech and Slovakian paths have been rather tortuous due to historical circumstances. The two countries were one—Czechoslovakia— from 1918 to 1992. In the early twentieth century, sickness funds covering workers and their families were made mandatory (1919) and were then unified and centralized (1924). Following the Second World War, Czechoslovakia fell under the economic and political influence of the Soviet Union, resulting in health legislation on national insurance in 1948. This system was totally funded by employers’ contributions. Then, in 1951, a “Semashko-type” healthcare system was introduced: centralized, managed by general government and financed through general taxation. A Bismarckian system was nevertheless reintroduced after 1989. Czechoslovakia was dissolved in December 1992. The Czech Republic and Slovakia were formed in 1993. The Czech health insurance system is currently managed by seven health insurance companies. These insurers are quasi-public, self-­governing bodies. The biggest one, the General Health Insurance Company, covers around 60% of the population and its payments are guaranteed by the State. Its activities are governed by a special law called the Act on the General Health Insurance Company. The activities of the other health insurance companies are governed by the Act on Employee Insurance Companies. None of these health insurance companies are allowed to make a profit. In Slovakia, the 1994 Act on Health Insurance introduced multiple health insurance funds. This scattered system was considered as inefficient and an extensive reform took place in 2004. The reform was based on managed competition, with the introduction of services related to healthcare and the possibility of user fees, the liberalization of networks (eligibility for permits and licences), selective contracting, independent oversight by Health Care Surveillance Authority (HCSA) and the transformation of all health insurance funds into joint-­ stock companies (Smatana et  al. 2016). It should be noted that this was not a “straight ahead” reform and that successive legislatures seemed hesitant: in 2007, the possibility of 34  Civil servants, self-employed individuals and high-income employees (the annual earnings threshold was €59,400 in 2019).

110 

P. MARTIN AND M. DEL SOL

profit-making in health insurance was banned, but was reintroduced by parliament in 2011 in order to comply with a ruling of the Constitutional Court of the Slovak Republic, under which the provision of health insurance could take place in the sphere of competition and insurers could make profits. Then, in 2012, the Slovakian government planned to reform the system and create a single public institution. This plan was stopped in 2014. In the Netherlands, the statutory health insurance scheme is rooted in the Bismarckian tradition of social insurance. Early predecessors of the country’s health insurance funds were mutual benefit societies, which appeared in the nineteenth century. These voluntary arrangements were replaced by mandatory state health schemes (Sickness Act 1913). Then, in 1941, the Germans occupying the country put in place a compulsory insurance system with sickness funds for employees earning less than a certain income level. Under this system, social health insurance covered only two-thirds of the population, those with lower incomes. For the other one-third a private health insurance scheme applied. Thus, until 2006, it was a hybrid system based on social insurance with a dominant role for not-for-profit sickness funds, combined with a long-standing role for private insurance covering the better off. The major healthcare reform of 2006 not only brought with it the long-desired unified insurance scheme, it also drastically changed the role of actors in the healthcare system. Multiple private health insurers are now supposed to compete in a regulated environment (Kroneman et al. 2016). The 2006 reform is now emblematic of the so-called Dutch model. Currently, there are 25 health insurers (in nine groups) and even though the legislation considers them as commercial companies, mutual benefit societies and non-profit organizations remain present and active on the market. Health insurance under public interest regulations. In the above-­ mentioned countries, the statutory health insurance scheme is plural, with the participation of multiple private actors administrating health insurance funds, some of them commercial companies acting in a quasi-market. This theoretically includes user choice and provider competition to a certain extent (Legrand 2011). Nevertheless, because health insurance is mandatory, as a legal consequence of solidarity,35 health insurance business has to 35  The principle of solidarity is at the core of social security programmes and systems. Technically, it means that contributions or premiums paid to social insurance schemes are not related to individuals’ risk profile. Healthy persons subsidize people in poor health, young people subsidize older ones, and so on.

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

111

comply with public interest regulations. In other words, within statutory health insurance schemes, private actors are entrusted with public service missions: they must guarantee a universal coverage governed by the principle of solidarity. Universality means that everybody has access to healthcare coverage, either for free or under financially affordable conditions. Access for all is the flipside of the mandatory nature of statutory health insurance. We should keep in mind that in Bismarckian “corporatist” systems, healthcare coverage is traditionally achieved through sickness fund membership based on occupational situation or position. In this respect, insurance enrolment is not fully open. Nevertheless, these systems have evolved so that everybody, regardless his/her position, is currently entitled to register and have access to healthcare coverage. Universality also implies that health insurances cover a uniform and standardized basic benefits package—the “health benefit basket”—which is generally determined by legislation or regulation,36 through procedures that differ from one country to another. So there can be no competition between private insurers as regards the range or level of benefits included in the mandatory scheme (though private health insurers try to attract customers with “extras” and supplementary benefits proposed in their voluntary health benefit package, which supposes extra premiums). Another typical aspect of health insurance in Bismarckian systems is copayment (or user charges). This means that public universal health insurance does not necessarily cover the whole cost of healthcare and services. This is generally well accepted—and leaves room for complementary private health insurance—but can pose problems in countries in which the population was used to cost-free healthcare. This was the case of Slovakia. The 2004 reform introduced user charges into statutory health insurance and the Constitutional Court had to decide if this was in line with the Slovakian constitution. The Court stated that user fees for health services were in conformity with the constitutional guarantee of cost-free healthcare (Constitutional Court of the Slovak Republic 2005). Solidarity requires pooling of heterogeneous risks. This implies that private insurers providing statutory health coverage cannot use risk selection: they are not allowed to risk-rate their premiums depending on health, age, sex, and so on. This kind of requirement generally fits with the ethos 36  The content of the basic benefit package is not uniform in Europe, though it is fairly comparable. In many countries, optical care is totally excluded and dental care is partially covered.

112 

P. MARTIN AND M. DEL SOL

of mutual benefit societies and non-profit organizations. Moreover, national legislations usually oblige sickness funds to accept all eligible applicants irrespective of their health status. In countries such as the Netherlands or Slovakia, although health insurance is now governed by commercial law (under the EU Insurance Directives), private insurers providing statutory health coverage have a duty to accept each person’s application and cannot impose risk selection. In countries in which the legislation strives to combine competition with solidarity, a risk equalization system (RES) is usually in place, which compensates insurers for predictable variations in medical spending. Indeed, the riskier portion of the population—older people—is covered by a specific sickness fund, for historical and sociological reasons, so that this fund’s “competitors” have an advantage as they cover those who make fewer medical claims (the healthy). Risk equalization or compensation is achieved through a national subsidy fund that collects contributions and subsidies each health insurer according to the risk profile of the population it covers. So, in theory, risk adjustment is supposed to make both good and bad risks equally attractive to insurers through adjusted financial compensation. Risk equalization schemes are to be found in the Czech Republic, Germany, The Netherlands and Slovakia, among others.37 In addition, it should be noted that in countries in which certain categories of the population can opt for voluntary health insurance that substitutes the public scheme, some legal provisions may be applicable, under article 54(2) and recitals of the 1992/49 Directive: open enrolment, community rating, lifetime cover, policies standardized in line with the cover provided by the statutory health insurance scheme at a premium rate at or below a prescribed maximum, participation in risk equalization schemes (referred to as “loss compensation schemes”) and the operation of private health insurance on a technical basis similar to life insurance. In Germany, voluntary health insurance has to comply with certain requirements: lifelong insurance guarantee, similarity to statutory insurance, level premium (premium rates must not increase because of increasing age, though health status is taken into account at the time of application).

37  These are the EU countries in which managed competition has been introduced into the statutory health insurance scheme. There are also RES in voluntary health insurance markets such as in Ireland.

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

113

3.1.2 Room for Both Market and Competition? Some EU Member States have decided to introduce managed competition into their statutory health insurance scheme through market-based reforms. In this model, health insurance agencies compete with each other and selectively contract providers on behalf of their insured, under narrowly defined rules set by government. This is particularly manifest in the Netherlands and Slovakia, while to a certain extent, managed competition has also been introduced in the Czech Republic and Germany. As noted regarding the 2006 Dutch reform, health insurers are expected to negotiate with providers and purchase efficient, good-quality care; patients are expected to critically assess and select the health insurer and provider of their choice (Van Ginneken et al. 2011). A comparative overview shows some legal characteristics which reflect this trend, though it is ultimately unclear whether it leads to real competition from an economic perspective. Two main characteristics or criteria will be examined here: user choice, that is, the possibility to choose and the freedom to switch health insurance; and insurers’ ability to differentiate themselves in the market and to offer better coverage. User choice among a plurality of health insurers. Freedom of choice of the health insurer is not truly a principle of the so-called Bismarckian model. Some countries such as France opted (in 1945) in favour of a single health insurance fund, so people have no choice: they must register with the local health insurance fund that provides them with basic health coverage. Note that before 1945 and under the French 1930 Social Insurances Act, people had a choice among a variety of organizations and bodies entrusted with statutory health coverage. At that time, mutual benefit societies played a prominent role. The immediate postwar political and social context explains why the French opted for a single fund mainly managed by workers’ trade unions representatives,38 while mutual benefit societies and provident funds were given the possibility to offer complementary (voluntary) health coverage (Martin 2013). In Belgium, unlike in France, an important place was given to mutual benefit societies within the statutory health insurance scheme set up in 1944-1945. Everyone is legally bound to be affiliated to a health insurance fund for basic coverage and has a choice in this respect. User choice might be guided by the health insurer’s offer in the sense that mutual 38  Employers’ representation was progressively introduced into the health insurance boards during the 1960s.

114 

P. MARTIN AND M. DEL SOL

societies traditionally propose complementary coverage (user charges, hospital fees, etc.) to their members. Nevertheless, since 2012 complementary coverage has become compulsory, which means that a person affiliated to a certain mutual company for basic health coverage must pay an extra premium for the complementary plan “offered” by this company. This creates a very strong bond between the user and the insurer. From a pure customer and market point of view, this might seem a little unorthodox, but it fits with the ethos of mutual societies based on membership (members sit on management boards). In Germany, though insured persons used to be automatically affiliated on the basis of their profession, they currently can choose their health insurance fund (since the 1993 Healthcare Structure Act).39 This implies that they can switch from one fund to another. It seems that despite this legal possibility, most people with poor health or chronic diseases do not change, or only do so rarely. This has consequences on competition. Conversely, some categories of people—the healthiest—appeared to be very mobile, so in 2002 the lawmaker restricted the freedom to switch from one insurer to another, imposing an 18-month probation period before a switch is allowed (Kaufmann 2004). Recent national legislations have tended to regulate the freedom to switch health insurer. In the Czech Republic, each insured person can change his or her health insurance company twice a year. In Slovakia, the 2004 health reform gave health insurance companies tools to compete for clients. The insured may change their health insurance company once a year. The deadline to switch health insurance companies is 30 September, becoming effective from 1 January the following year. The switch can only be refused if the insured person has applied to more than one health insurance company at the same time. Other administrative barriers that could have hindered this switching of health insurance companies (e.g. a written notice) were removed. The mobility of the insured has varied significantly over time and is influenced by marketing activities and by the extent to which the insured exercise their freedom of choice. In the Netherlands, the insured can change health insurer at the end of each year (before 1st January). In fact, mobility is determined by the structure of the health system: in countries in which health insurers cover the public scheme and also offer supplementary plans, people tend to be more “captive” because of the risk selection applied to the supplementary health insurance they have purchased.  The 1993 Act has been in force since 1997.

39

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

115

Insurers’ ability to differentiate themselves. Key elements in differentiation are: price freedom; selective contracting; “special offers”; and extra benefits. Price freedom, that is, the possibility for each health insurer to set the price of contributions or premiums, is not fully recognized in the countries under review here. In some countries such as Belgium, the Czech Republic or Slovakia, all health insurers in the statutory health insurance scheme apply a uniform contribution rate determined by the State. In Germany, before 2007, each sickness fund had the freedom to set the contribution rate, so local funds were able to differentiate themselves from the others. Generally, the funds that cover “good risks” (healthy persons) are able to propose low contribution rates and thus attract customers. In 2007, Germany reformed its system and since then, a single and unique contribution rate is established by governmental decision (sickness funds are nevertheless allowed to demand a flat-rate complement from their members). In the Netherlands, private health insurers in the statutory scheme are free to set the contribution rate, but as mentioned above, because they are not allowed to engage in risk selection, they must charge all their policyholders the same premium (community rating system).40 Residents have market choice, though very few variations are to be observed in prices between the different health insurers, because the legislation requires them to offer a standardized basic benefits package and thus prevents risk selection. Another factor or element of differentiation in the market is health insurers’ capability to control the costs of healthcare and services. This can be achieved by selective contracting, that is, contractual arrangements between health insurers (sickness funds) and healthcare providers. Under certain legal conditions, health insurers can negotiate the prices, volume and quality of medical and pharmaceutical care provided. This is “strategic purchasing”. These arrangements often take the form of integrated care, which means that a group of providers is contracted by an insurer to provide services across sectoral boundaries. They are “selective”, as they usually limit beneficiary freedom of choice to those providers within the selected network. Among EU Member States, selective contracting has been implemented in Germany, the Netherlands and Slovakia, though as pointed out by Thomson et  al. (2013), “Dutch insurers have access to 40  “Community rating” means that the insurance company must charge the same rate for a given level of service, regardless of age, sex or health status.

116 

P. MARTIN AND M. DEL SOL

many more efficiency-enhancing tools than do their counterparts in other countries”. In Germany, before 2000, there were restrictions to selective contracting and, like in Belgium, sickness funds were only allowed to have collective agreements with healthcare providers (Busse et al. 2017). This restriction was removed in 2000 and in the mid-2000s, sickness funds were permitted to negotiate discounts with pharmaceutical drug manufacturers. In 2007, they also began to offer a choice of tariffs, which was a feature previously reserved for private health insurance companies. The funds can adapt to the individual needs of their members by offering different benefits packages and prices (Busse et al. 2017). Though it remains rather low, the share of statutory health insurance expenditures in selective contracts increased significantly between 2002 and 2015.41 In the Netherlands, health insurances propose basically two kinds of plans to enrollees: in-cash benefits plans (the insurance reimburses healthcare expenses and the enrollee is totally free to choose the physician, the hospital and all kinds of health professionals); and in-kind benefits plans with selectively contracted providers (the providers are directly paid by the health insurance). It should be noted that premiums are often lower if enrollees choose in-kind benefits plans. Theoretically, Dutch health insurers can negotiate the prices and quality of health services. Actually, the negotiation margins are quite narrow, as for a wide range of healthcare services, tariffs are established by the Dutch Healthcare Authority (DHA). As far as hospital care is concerned, tariffs are partially regulated by the DHA, and about 70% of hospital care is provided through selected networks. However, Dutch health insurers seem to be rather reluctant to implement selective contracting as they believe their enrollees will not accept it (Bes et  al. 2013). In Slovakia, purchaser-provider relations are based on selective contracting regulated by the Ministry of Health. Since 2012, this regulation has been strengthened as the Ministry of Health has set a list of selected state providers, which is in fact a compulsory network that has to be contracted by all health insurance companies. All in all, selective contracting does not appear to be a strong element of differentiation in European countries. A common way to differentiate and attract health insurance customers is to offer extra benefits. As already mentioned, private health insurers in the statutory health insurance scheme often propose supplementary or complementary plans. This market is well developed in countries such as  From 0.02% in 2002 to 1.5% in 2015 (Busse et al. 2017).

41

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

117

Belgium and the Netherlands. Economically speaking, it can be analysed as a driver of competition. In legal terms, however, a distinction is made between a health insurer’s activity as a provider of statutory (public) coverage, which in principle is beyond the scope of EU competition law, and voluntary health plans which, in principle, are an economic activity under EU competition law.42 Nevertheless, it can happen that health insurances are allowed by national legislation to complement the statutory basic benefits package for free, which means that enrollees do not purchase a complementary plan: extra benefits are part of the statutory coverage offered by certain private health insurers. This obviously provides an opportunity for differentiation and competition as regards quality and service offerings. The Slovakian legislation exemplifies this situation. Indeed, despite the strong regulations governing the scope of covered services within the statutory scheme, health insurance companies can attract new customers by offering additional services such as medicine discounts; reimbursing co-payments for some medicines, vitamins or non-healthcare services; shorter surgery waiting times; broader preventive examinations; or a variety of electronic support services. These circumstances, plus the fact that Slovakian health insurance companies are for-profit organizations, led the Court of 1st instance of the European Union to consider that there was competition between the various Slovakian health insurance companies as noted above. The Court notably states that “even if there is no competition within the Slovak compulsory health insurance system in respect of either the compulsory statutory benefits or formally on the amount of contributions, there is nevertheless intense and complex competition due to the market volatility resulting from insured persons’ power freely to choose their health insurance provider and to switch insurance company once a year, and the fact that health insurance bodies are competing in terms of the quality of service, which is assessed individually by the insured persons”. In this regard, according to the Court, health insurance in Slovakia is economic in nature and competition law has to apply.43 From this comparative overview, a first conclusion can be drawn: the different features analysed—the presence of private insurers, user choice, market-oriented reforms and policies—do not lead to a systematic

42  With a possible exemption based on the exception for services of general economic interest (BUPA case). 43  General Court case T-216/15, Dôvera zdravotná poist’ovňa, a.s. versus Commission, 5 February 2018.

118 

P. MARTIN AND M. DEL SOL

application of EU competition law. Players actually implement social security schemes widely controlled by the State. EU competition law appears to apply exceptionally and due to peculiar circumstances (cf. the Slovakian case). 3.2  Complementary Health Insurances and Mixed Market-­social Security Patterns As pointed out by Thomson and Mossialos (2010), “Governments in several member States recognize that complementary cover of statutory user charges can contribute significantly to social protection”. So they tend to regulate and manage this market—the complementary cover of user charges—as a tool for social protection or welfare policy. This is particularly manifest in Belgium, Ireland, France and Slovenia.44 Note that this approach is quite ambiguous given that on the one hand, national regulations aim at correcting market effects which tend to create or increase inequalities (by risk selection), and on the other hand those regulations serve public policies that seek to deal with the flaws of statutory public health insurance. In a context of government budget constraints, cost sharing between public and private health insurance might be the way to go for some countries.45 Moreover, this creates a blurry situation as regards the EU legal framework. This section aims to outline some features of national regulations that possibly pose or have posed problems from the EU law perspective. 3.2.1

 he Presence of Non-profit Insurers and Mutual Benefit T Societies (and the Recognition of a Special Status) In the field of health insurance, non-profit organizations and mutual benefit societies play an important role in the EU countries under review here, as statutory health coverage providers and/or as complementary health insurers. Mutual societies are prominent in the Belgian and French 44  We shall focus here on countries in which voluntary health insurance plays an explicitly complementary role, covering user charges for publicly financed health benefits. 45  Overall, private health insurance does not make a significant contribution to total health spending in the European Union. It accounts for less than 5% of total health expenditure in two-thirds of the Member States, although it plays a considerably larger role in Slovenia (13.1%), France (12.8%) and Germany (9.3%). Source: European Commission, Private health insurance in the European Union: Final report prepared for the European Commission, Directorate General for Employment, Social Affairs and Equal Opportunities, 24.06.2009, p. 6.

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

119

complementary health insurance markets, for historical reasons. In Slovenia, when private health insurance was introduced (in 1993), two entities offered complementary plans covering user charges: Vzajemna, a mutual society specializing in health, and Adriatic Slovenica, a commercial joint-stock company offering a range of insurance products. The mutual society—Vzajemna—was linked with the statutory health coverage. In Ireland, the market for private health insurance was instituted by the Voluntary Health Insurance Act 1957 which created a semi-public (non-­ profit) body, the Voluntary Insurance Board, now known as VHI Healthcare, which remained monopolistic for a long time. Legally speaking, mutual and non-profit organizations that sell voluntary/complementary health plans fall under the scope of the EU Insurance Directives and competition law. Nevertheless, national legislations have tended to grant them a special legal status which may be incompatible with EU law requirements. This has led to some adjustments. A comparative view reveals different national approaches to the EU Insurance Directives and to their transposition. In France, for a long time, mutual benefit societies, their federations and unions as well as the French government resisted the applicability of the EU Insurance Directives. In the 1970s and 1980s, mutual benefit federations did not consider themselves as “undertakings” or insurance businesses as defined by first- and second-generation Insurance Directives. Nevertheless, they finally had to comply with the new legal framework designed by the French legislation through the 1985 Act which reformed the Mutual Societies Code. This reform opened the market and competition to for-profit insurance companies which were allowed to offer complementary health coverage. The reform of the Mutual Societies Code was followed by an important Act, in 1989, regulating the market for complementary health coverage46 but also confirming the principle of a competitive market open to commercial companies. However, at that time, the federation of mutual societies still believed that they could maintain a certain legal specificity, such as the fact that mutual societies could have several activities combining health coverage and the management of welfare services and institutions. Just before the adoption of the 1992 Non-life Insurance Directive, a discordance appeared between the different French federations of mutual benefit societies about whether or not they entered into the EU legal framework. The biggest federation opted for an 46  The French legislation limits risk selection within collective complementary health plans set by employers or sectoral collective bargaining.

120 

P. MARTIN AND M. DEL SOL

adaptation of the legal status of mutual benefit societies in order to comply with the EU Insurance Directives. But as soon as the 1992 Directive was adopted, the federations mobilized in order to delay the transposition procedures and obtain recognition of their specificities from the French government and the EU. This situation lasted a few years, but in 1998, the European Commission appealed to the EU Court of Justice (ECJ). According to the Commission, by not transposing the Insurance Directives in respect of mutual societies governed by the Mutual Societies Code, the French Republic had failed to fulfil its obligations under the EC Treaty and under the directives. Two main points were underlined by the European Commission. First, French mutual societies were not subject to the prudential and financial requirements laid down in Directives 92/49 and 92/96 (adequate technical provisions and solvency margin). Second, their insurance activities as such were not legally separate from their “philanthropic” activities, particularly in relation to pharmaceuticals, optical centres, holiday centres and the renting of meeting rooms, in breach of the principle of specialization of insurance companies laid down in Directives 92/49 and 92/96, which requires that the commercial and philanthropic activities pursued by mutual societies should not be managed by the same legal entity; a distinction would be drawn in national law between mutual insurance activities carried out in the form of services provided for consideration in cash or in kind and falling within the scope of the prudential rules in Directives 92/49 and 92/96 on the one hand, and activities carried out by mutual societies which are unconnected with insurance and which ought to be managed by subsidiaries, on the other. The ECJ agreed with the European Commission and ruled that the French Republic had failed to fulfil its obligations.47 In Belgium as in France, mutual societies’ activities had remained out of the scope of the EU Insurance Directives for years. Oddly enough, this situation was not challenged until 2008 when the European Commission initiated an infringement procedure against the Kingdom of Belgium for incorrect transposition of the EU Insurance Directives. The reason for this late reaction from the EU Commission might be that in Belgium, unlike in France, mutual societies administrated the statutory health insurance scheme, so that even if they usually offered complementary plans to their members, this activity tended to be considered as provident services and welfare fund management. In other words, under the Belgian legislation, 47  ECJ case C-239/98, Commission versus France, 16 December 1999, ECR 1999, p. I-8935.

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

121

those “extra benefits” were a kind of extension of the statutory scheme and this justified the exemption from EU internal market law. The problem was that in the “complementary package” offered by the mutual societies to their members, coverage of user charges and hospitalization fees were generally included. According to the European Commission, alerted by ASSURALIA—a Belgian union of insurance companies—these activities were economic in nature and must fall within the scope of the Insurance Directives. The European Commission observed that the package of complementary benefits as well as the related premiums were freely determined by each mutual society, that solidarity only applied within the scope of the mutual society (not at national level) and that insurance companies offered the same kind of coverage, so there was competition. The European Court of Justice agreed with the European Commission and stated that the Kingdom of Belgium had failed to fulfil its obligations under the EU Insurance Directives.48 As a consequence, the Belgian government and parliament had to modify the legislation on complementary health coverage. What is remarkable when comparing the French and the Belgian pathways is that the Belgian reform passed in 2010 managed to keep certain complementary plans out of the scope of insurance law. Under certain legal conditions, complementary coverage of hospital fees by mutual societies could be considered as “operations” in the sense of art. 2.2. b. of the First Non-life Directive (73/239).49 Thus, in Belgium, complementary health coverage was not totally “marketized” as insurance products and activities offered in a competitive market. Ireland provides another example of resistance by non-profit organizations towards alignment with the EU legal framework. In Ireland, between 1957 and 1997, the semi-public body VHI Healthcare was the only private health insurer in the Irish market.50 Following the introduction of the EU’s Third Non-life Insurance Directive in 1992, the VHI market was opened up to competition in 1994. Private and for-profit health insurers then entered the market but VHI Healthcare remained exempted from the regulations of the Insurance Act. Actually, VHI Healthcare had a high profile in the media and political debate and substantial influence in the

 ECJ case C-41/10, Commission versus Kingdom of Belgium, 28 October 2010, ECR 2010, p. I-138. 49  “Operations of provident and mutual benefit institutions whose benefits vary according to the resources available and in which the contributions of the members are determined on a flat-rate basis”. 50  Alongside a number of smaller occupational schemes with restricted membership. 48

122 

P. MARTIN AND M. DEL SOL

health system. A key issue was the discussion about the status of VHI Healthcare in relation to its competitors (Turner 2016). VHI Healthcare’s competitors claimed that its exemption from the regulations of the Insurance Acts gave it unfair advantages because it was not required to hold a minimum level of reserves to guarantee solvency and did not need to establish subsidiaries to engage in other business activities. Changing the regulatory status of VHI Healthcare to bring it in line with its competitors was initially proposed in 1999 and again in 2008, but has been fully implemented only recently,51 after the ECJ found the exemption to be unlawful.52 The Slovenian case shows a much quicker and less conflictual alignment with the EU legal framework. As mentioned, private health insurance was introduced in 1993 and Vzajemna, a mutual society specializing in health, started selling voluntary health insurance as an integral part of the Health Insurance Institute of Slovenia (HIIS), the statutory body responsible for purchasing publicly financed health coverage. To a certain extent, the situation could be compared to the Belgian one.53 But in 1999, the Slovenian legislation was modified in order to comply with EU law: following amendments to the Slovenian Health Care and Health Insurance Act, the HIIS separated its mandatory health insurance from its VHI business. Vzajemna’s activity is currently exclusively dedicated to complementary/ voluntary health insurance and though it remains a non-profit organization, its legal status is in line with the EU Insurance Directives. 3.2.2 PHI Market under State Control Some countries regard complementary health insurance as a strategic element—a sort of “pillar”—of social protection and public health policies. Thus, national legislations can be prone to imposing rules that tend to put this activity under control. There are various ways to do so: using legal and fiscal incentives; imposing tight regulation of insurers’ activity; regulating the proper market and neutralizing selective effects. Some of these regulations have been found to breach EU rules 51  VHI had to submit an authorization application to the Central Bank, like all health insurers, with a deadline at May 2014. VHI finally complied with the solvency requirements in January 2016. 52  ECJ case C-82/10, Commission versus Ireland, 29 September 2011, ECR 2011, p. I-140. 53  In the sense that there was a blurry zone between mandatory and voluntary health coverage.

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

123

Legal incentives to develop PHI as a complement of social security. In France, public policies use a full range of incentives in order to make complementary health coverage universal and congruent with national objectives, especially the containment of health costs: Universal complementary health coverage (a health benefit package for free and with no user charges for low income people);54 “responsible and solidary” complementary health plans supported by fiscal incentives;55 and employers’ legal duty to implement complementary health coverage plans.56 Nevertheless, those measures do not appear to challenge EU law. In Slovenia, the legislation is quite restrictive. Insurers are required to cover user charges for all publicly financed health services. However, children are exempt from user charges and so do not require complementary PHI. In a way, complementary health coverage appears to be compulsory for adults, as they may pay penalties if they do not purchase complementary PHI as soon as they become liable for user charges. For each complete year (12 months) they do not have PHI, the penalty is equal to 3% of the premium. The maximum penalty is 80% of the premium. Tight regulation of complementary health insurance activity. Slovenia is the most illustrative example of State control over voluntary health insurance activity. Some authors consider Slovenian complementary health insurance as a SGEI (Nikolic 2015). Indeed, the Slovenian market for 54  This coverage is offered to all residents whose earnings are under a certain threshold. It is financed through a national fund that collects special taxes from the complementary health insurers (profit and non-profit organizations) which sell voluntary health insurance in France. 55  Private health insurers are induced to offer complementary packages whose content is determined by law, the so-called “responsible and solidary contracts”: these plans must cover all user charges for healthcare provided in the scope of the mandatory scheme (with some restrictions for a certain range of drugs) as well as hospital fees; they must also include minimum services and benefits for optical care. The insurer cannot engage in risk selection. Conversely, these plans must not cover user charges when the patient does not respect the “healthcare pathway” (obligation to consult one’s own general practitioner (GP); limitations in the possibility to consult physicians who charge extra fees not covered by the mandatory health insurance, etc.). 56  The Employment Act 2013, in accordance with the national collective agreement signed on January 2013 by the French national employers’ associations and workers’ unions, aims to extend complementary health coverage. It states that the trades unions and employers’ associations bound by a sectoral collective agreement have a duty to bargain about the implementation of collective complementary health insurance, if such a plan has not yet been implemented in their professional sector. Then, employers at the company level have a duty to propose to their employees a complementary health coverage which complies with the collective agreement requirements in this respect (deadline on January 2016). Under the 2013 Act, employers have to financially cover 50% of the premium charges.

124 

P. MARTIN AND M. DEL SOL

complementary PHI covering user charges is subject to relatively tight regulation. Some of these regulations have been found to breach EU rules. The Health Care and Health Insurance Act (2006) introduced a number of controls on PHI business. Under this Act, health insurers from other countries had an obligation to establish a branch office in Slovenia. The legislation also stipulated controls over the use of profits, systematic notification of product changes and prior approval of premium increases. In 2011, the European Commission found these rules to be in breach of EU competition and free movement rules (specifically, the First and Third Non-life Insurance Directives). In response, the Ministry of Health planned to bring Slovenian regulation in line with EU rules, but reform efforts were halted due to early elections held in December 2011 and the Commission subsequently referred Slovenia to the European Court of Justice. In 2012, the ECJ found that Slovenia had failed to transpose the Non-life Insurance Directives correctly57 Regulations correcting market effects: open enrolment, community rating, and so on. In countries in which complementary health insurance is regarded as a driver for better access among the population to healthcare and a key element of the right to healthcare, the legislation tends to impose rules that correct or neutralize possible market effects (the exclusion of certain categories). Various techniques are used, such as open enrolment, community rating and risk equalization schemes. Open enrolment means that health insurance companies must accept anyone who wishes to join, regardless of age, sex or health status. Irish and Slovenian legislations stipulate open enrolment. The Irish legislation also stipulates “lifetime cover”, which means that once a person joins a private health insurance and continues to pay his/her premiums, the insurance company cannot refuse to provide him/her with cover. In France, the 1989 Act on complementary health plans for employees limits the possibility for insurers to engage in risk selection (there can be no risk selection in the collective plans) and provides workers with some guarantees as regards the continuity of their complementary health insurance coverage when they quit their job. Community rating means that the insurance company must charge the same rate for a given level of service, regardless of age, sex or health status. So, all enrollees pay the same amount for the same benefits. This system

 ECJ case C-185/11, Commission versus Slovenia, 26 January 2012.

57

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

125

applies in Ireland and in Slovenia. In France, there is no legal prohibition from differentiating premiums rates, so private insurers can modulate rates according to age. Nevertheless, as mentioned, French legislation—since 2002—has encouraged health insurers to offer “responsible and solidary” contracts or plans, with fiscal incentives. In this kind of plan, which is currently the most widespread in the market, there can be no risk selection (no medical questionnaire) and premiums cannot be related to health status. Risk equalization or compensation schemes for the complementary health insurance market have been set up in Ireland and Slovenia. Risk equalization is needed to support community-rated health insurance. Basically, health insurers receive credits or subsidies from a national fund or authority to compensate for the additional cost of insuring older and less healthy members. As explained and developed in Chap. 8 of this book, the Irish risk equalization scheme was challenged by a private for-profit insurer, British United Provident Association (BUPA), which strongly contested this system: BUPA claimed that the scheme was a State-enforced subsidy to VHI and was anti-competitive. In 2003, the European Commission said the scheme did not constitute a State aid or subsidy to VHI58 and in 2008, the Court of First Instance upheld the Commission’s decision.59 In Slovenia, Adriatic Slovenica and Vzajemna challenged the risk equalization scheme in the Constitutional Court and eventually at the EU level, claiming that it distorted competition in the VHI market. All disputes were rejected by the Constitutional Court in 2006 and the risk equalization scheme remained in place after the ECJ ruling in 2012 which did not address this issue. One can speculate about the uncertainty of ECJ case law as regards the concept of SGEI applied to national risk equalization schemes. Would the ECJ have made the same ruling on the Slovenian scheme as on the Irish one, according to BUPA case law? One may think that in the Slovenian legal environment, a SGEI appears to be even clearer as regards complementary health insurance than in the Irish case, so the Slovenian risk equalization scheme should not be challenged by EU Law (Thomson and Mossialos 2010).

58  More precisely, the Commission considered that even if the risk equalization scheme could be seen as a State aid, it was nonetheless compatible with the internal market. 59  Court of First Instance of the European Communities, case T-289/03, BUPA e.a. versus Commission, 12 February 2008, ECR 2008, p. II-81.

126 

P. MARTIN AND M. DEL SOL

References Bernard, E. (2009). L’activité économique, un critère d’applicabilité du droit de la concurrence rebelle à la conceptualisation. Revue Internationale de Droit Economique, 353–386. Bes, R., Wendel, S., Curfs, E., Groenewegen, P., & de Jong, J. (2013). Acceptance of Selective Contracting: The Role of Trust in the Health Insurer. BMC Health Services Research, 13(375). Busse, R., & Blümel, M. (2014). Health Systems Review—Germany. Health Systems in Transition, 16, 1–296. Busse, R., Blümel, M., Knieps, F., & Bärnighausen, T. (2017). Statutory Health Insurance in Germany: A Health System Shaped by 135 Years of Solidarity, Self-governance, and Competition. Lancet, 390, 882–897. https://doi. org/10.1016/S0140-6736(17)31280-1. Del Sol, M. (2016). Protection sociale: le droit de l’Union européenne entre monopole et concurrence. In M.  Borgetto, A.-S.  Ginon, & F.  Guiomard (Eds.), Quelles(s) protection(s) sociale(s) demain ? (Thèmes et commentaires) (pp. 107–125). Paris: Dalloz. Driguez, L. (2006). Droit social et droit de la concurrence (868 p). Bruxelles: Bruylant. Dunne, N. (2010). Knowing When to See It: State Activities, Economic Activities, and the Concept of Undertaking. Columbia Journal of European Law, 16(3), 427–463. Eikenberry, A. M., & Kluver, J. D. (2004). The Marketization of the Nonprofit Sector: Civil Society at Risk? Public Administration Review, 64, 132–140. Kaufmann, O. (2004). La concurrence dans l’assurance maladie légale allemande. In P.  Hassenteufel & S.  Hennion-Moreau (Eds.), Concurrence et protection sociale en Europe (pp. 245–264). Rennes: Presses Universitaires de Rennes. Kroneman, M., Boerma, W., Van den Berg, M., Groenewegen, P., de Jong, J., & van Ginneken, E. (2016). The Netherlands: Health System Review. Health Systems in Transition, 18(2), 1–240. Lear, J., Mossialos, E., & Karl, B. (2010). EU Competition Law and Health Policy. In E. Mossialos, G. Permanand, R. Baeten, & T. Hervey (Eds.), Health Systems Governance in Europe: The Role of European Union Law and Policy (pp. 337–378). Cambridge: Cambridge University Press. Legrand, J. (2011). Quasi-Market versus State Provision of Public Services: Some Ethical Considerations. Public Reason, 3, 80–89. Lewalle, H. (2006). A Look at Private Health Care Insurance in the European Union. Revue française des affaires sociales, 2006(6–7), 33–157. Martin, P. (2013). The Role of Non-public Actors in French Social Security: The New Features of Solidarity. In F.  Pennings, T.  Erhag, & S.  Stendahl (Eds.),

4  THE UNCERTAIN AND DIFFERENTIATED IMPACT OF EU LAW… 

127

Non-Public Actors in Social Security Administration. A Comparative Study (pp. 37–61). Alphen aan den Rijn: Wolters Kluwer. Mossialos, E., & Lear, J. (2012). Balancing Economic Freedom Against Social Policy Principles: EC Competition Law and National Health Systems. Health Policy, 106(2), 127–137. Mossialos, E., Permanand, G., Baeten, R., & Hervey, T. (2010), Health Systems Governance in Europe: The Role of European Union Law and Policy. Cambridge University Press. Nikolic, B. (2015). Slovenian Complementary Health Insurance as a Service of General Economic Interest. International Public Administration Review, 13(1), 49–67. Pennings, F., Erhag, T., & Stendhal, S. (Eds.), (2013). Non-public Actors in Social Security Administration. A Comparative Study. Wolters Kluwer. Sagan, A., & Thomson, S. (2016). Voluntary Health Insurance in Europe. Role and Regulation (no. 43, 50). World Health Organization, European Observatory on Health Systems and Policies. Smatana, M., Pažitný, P., Kandilaki, D., Laktišová, M., Sedláková, D., Palušková, M., van Ginneken, E., & Spranger, A. (2016). Slovakia: Health System Review. Health Systems in Transition, 18(6), 1–210. Thomson, S., Busse, R., Crivelli, L., Van de Ven, W., & Van de Voorde, C. (2013). Statutory Health Insurance Competition in Europe: A Four-country Comparison. Health Policy, 109, 209–225. Thomson, S., & Mossialos, E. (2006). Choice of Public or Private Health Insurance: Learning from the Experience of Germany and the Netherlands. Journal of European Social Policy, 16(4), 315–327. Thomson, S., & Mossialos, E. (2007). Regulating Private Health Insurance in the European Union: The Implications of Single Market Legislation and Competition Policy. Journal of European Integration, 29(1), 89–107. Thomson, S., & Mossialos, E. (2010). Private Health Insurance and the Internal Market. In E. Mossialos (Ed.), Health Systems Governance in Europe. The Role of EU Law and Policy (pp. 419–460). Cambridge: Cambridge University Press. Turner, B. (2016). Ireland. In A. Sagan & S. Thomson (Eds.), Voluntary Health Insurance in Europe. Country Experience (pp. 77–82). Geneva: WHO. Van Ginneken, E., Schäfer, W., & Kroneman, M. (2011). Managed Competition in the Netherlands: An Example for Others. Eurohealth, 16(4), 23–26.

CHAPTER 5

An Increasing Homogenisation of Private Health Insurers Under Solvency II? Philippe Abecassis and Nathalie Coutinet

1   Introduction The voluntary private insurance market in Europe has been strongly impacted by European regulatory changes as well as by changes in national health policies in almost all European countries. Various European directives have abolished the insurance monopoly and the specificities of mutual benefit societies that assimilate their status to that of the insurance companies. These changes have broadened the market for the various providers of health insurance contracts and also introduced competition between players whose markets had hitherto been national. These first directives on the freedom of establishment and the freedom to provide services were accompanied by legislation to ensure the solvency of insurers. Solvency I (73/239/EEC) and Solvency II (2009/138/EC, L35A) introduced enhanced solvency rules (see Chaps. 2 and 3). In addition to these new European regulations, there have been changes in health policies

P. Abecassis (*) • N. Coutinet CEPN (UMR CNRS 7234), Université Sorbonne Paris Nord, Villetaneuse, France e-mail: [email protected] © The Author(s) 2021 C. Benoît et al. (eds.), Private Health Insurance and the European Union, https://doi.org/10.1007/978-3-030-54355-6_5

129

130 

P. ABECASSIS AND N. COUTINET

implemented in each European country. These health policies are intended to reduce the public and/or social costs of health, while their weight in the gross domestic product (GDP) has risen steadily until the end of the 2000s. The common feature of these policies has been a reduction in public and social health expenditures in all European countries. This was achieved either by reducing public and social health expenditures directly or by disengaging the State from social protection. This disengagement with multiple shapes has sometimes left a vacant space for private insurers and households, and sometimes organised its withdrawal by contracting or regulating with these private insurers. In any case, private health insurance operators have been involved—on their own initiative or not—in new health insurance markets based on common European rules. In preliminary analysis, it is often accepted that the combination of the new rules and the change in the size of the private insurance markets has constituted a new situation to which insurers would naturally have responded by reorganising themselves. This reorganisation would have been carried out in a canonical manner, that is, through the normal functioning of markets and competition: external and/or financial growth by insurers aimed, for each insurer, to take possession of new markets in its own country as in other European countries, to consolidate or increase its own market share, to expand and diversify its offer in order to maintain or even strengthen its solvency, and so on. This involves a wave of mergers and acquisitions (M&A) and strategic alliances between players. However, this first analysis is hampered by the particular characteristics of the private health insurance market. This depends directly on the organisation of the compulsory health care system; the number and status—for-profit or non-­ profit—of insurers; local rules, specific to each country, which regulate and organise the market and assign a well-defined role to private insurers; the individual or collective nature of health insurance contracts, and so on. As a result, the impact of European regulations on private health insurance markets is uncertain. In particular, how should the many M&As conducted over the past 20 years be interpreted? Are there responses undertaken to adapt to new local and European conditions? If that is so, then what is the weight of European regulations, what is the weight of local regulations? And if not, what are the determinants that may explain waves of concentrations? The aim of this chapter is to shed light on this question based on an analysis of M&A carried out in private health insurance during the 2000s and 2010s. In Sect. 2, we focus on showing how changes in the insurers’

5  AN INCREASING HOMOGENISATION OF PRIVATE HEALTH INSURERS… 

131

environment, in particular the public disengagement from health care coverage, are creating a favourable environment for the emergence or development of more or less profitable national markets for private health insurance. Section 3 focuses on the nature and determinants of M&A in this sector over the past 20 years.

2   The Transformations of the Environment of Private Insurance Providers in Europe Reforms carried out since the 1980s and more particularly in the 1990s in the context of mainly liberal policies have led to attacks on public spending and on social protection, said to be too high. Health care systems are considered too generous. On the one hand, they are accused of leading patients to overconsuming care and, on the other hand, of hindering the competitiveness of the companies that finance it. Thus, on the basis of these arguments, many European countries have taken measures to contain public health care expenditures and have gradually withdrawn from health care coverage. The many measures to reduce or control compulsory health care insurance budgets have often been accompanied by other measures, generally fiscal, to encourage the development of a voluntary private insurance offer that promotes the growth of the insurance market. However, the diversity of compulsory health insurance systems and forms of disengagement (2.1) have modulated the effects of these measures on the development of private insurance provision. The nature and development of markets is thus highly dependent on history and national specificities (2.2). 2.1  The Disengagement of Public/Mandatory Systems Frees Up Market Space In the 1990s, many governments in Europe initiated policies to reduce public funding of health expenditures (André et al. 2016). This disengagement, which has been particularly visible in European countries since the early 2000s, takes three main forms: reduction of public spending, reduction of the share of public funding in health expenditures and under-­ financing of public health care provision. The first form consists of a set of measures justified by the reduction in public expenditures deemed necessary. These include reducing the

132 

P. ABECASSIS AND N. COUTINET

packages of health care covered by the public system, developing co-payment schemes, reducing the proportion of the population covered or providing incentives to contract voluntary health insurance. Most European countries have implemented this type of measure (Paris 2014). Thus, copayment has become widespread in all countries. During the 2000s, countries such as France, Bulgaria and the Czech Republic redefined the basket of care provided, excluding certain medical care and goods, particularly dental care. Other countries such as the Czech Republic and Malta have excluded certain populations from public health coverage, such as nonEuropean migrants. Finally, many European countries have implemented tax measures to promote the development of voluntary private insurance. This is the case in Hungary, Denmark or Italy. The main consequence of these measures is an increase in the Out of pocket (OOP) payments.1 For the European Union (EU) as a whole, these remaining costs represent one fifth of health expenditure. The OECD estimates that in 2019 they will be around 10% in France, Luxembourg or the Netherlands, and almost 40% in Bulgaria, Latvia and Cyprus. The evolution of the OOP payments is also quite disparate according to the countries (Fig. 5.1). Over the period 1991–2018, its average annual growth rate is positive for the majority of the countries. Particularly high in the Czech Republic (6.18%) and the Slovak Republic (4.14%), it is close to 3% in Hungary and Poland and slightly above 2% in Luxembourg and Estonia. However, some countries are succeeding in reducing the OOP payments over the same period. This is particularly the case in France, Denmark, Belgium and Ireland. These changes in the OOP payments are independent of the role of the private health insurance sector. Thus, in France and Denmark, where private health insurance plays a mainly complementary role, the decline in the OOP payments is mainly due to incentives to promote the private health insurance market. These incentives are the generalisation of compulsory company complementary insurance contract (2016) in France or the introduction, in Denmark, of a tax exemption for the purchase of company supplementary insurance contract (2002). Conversely, neither Greece nor Ireland, where health insurance is mainly supplementary, has developed policies targeted at the development of the private health insurance market. In these two 1  This corresponds to the expenses that patients have to pay when no public or private health insurance system covers the total cost. It includes the share of expenditures paid by households to caregivers, including informal payments.

5  AN INCREASING HOMOGENISATION OF PRIVATE HEALTH INSURERS… 

133

Fig. 5.1  OOP payments evolution, 1991–2018. Austria, Lithuania and Sweden: 2001–2018; Belgium: 2004–2017; Denmark, Czech Republic and the United Kingdom: 1991–2017; Estonia: 2000–2018; Finland, Luxembourg, Ireland and France: 1991–2017; Germany: 1993–2018; Greece: 2009–2017; Hungary: 1992–2018; Italy and Poland: 1991–2018; Latvia: 2001–2017; the Netherland: 1999–2018; Portugal: 2001–2018; Slovak Republic: 1998–2017; Slovenia: 2004–2018; and Spain: 1992–2017. (Sources: OECD data, 2019.)

countries, only the self-development of the local insurance market explains the decrease in the OOP payments. Overall, the measures associated with the first form of government disengagement lead to a significant increase in household voluntary private insurance spending. Between 2000 and 2014, these expenditures increased particularly in Croatia (+24.2%), Ireland (+15.3%) and Poland (+11.4%). The increase is even more striking when measured in per capita expenditure: 2000% in Hungary, 1600% in Lithuania, 1040% in Poland and 771% in Bulgaria2 (Sagan and Thomson 2016a). The second form of disengagement corresponds to a decrease in the share of public funding in health expenditure. This form of disengagement has been favoured in some European countries such as the Netherlands or, to a lesser extent, Germany. Thus, the share of public financing of health expenditures in the Netherlands has fallen from nearly 70% before the 2006 reform to less than 30% until 2015 and reached 2  However, these expenditures fell slightly in some countries, particularly in the Czech Republic (−1.3%) and Slovenia (−1.1%).

134 

P. ABECASSIS AND N. COUTINET

26.4% in 2018. German reforms have been less radical than in the Netherlands, and Germany illustrates the evolution observed in many European countries in recent years (Turquet 2012). The competition between insurers contributed to the significant decrease in the share of public expenditures in health expenditures in the early 2000s. This share then increased slightly from 2009 onwards. It rose from 80.3% in 1995 to 75.1% in 2006 and 2012 and 77.2% in 2017. The third form of disengagement, the underfunding of public health care provision, is more specific to the Beveridgian model. In this type of model, the underfunding of public health care provision encourages patients to seek private providers and to subscribe to substitutive or supplementary health coverage. The United Kingdom is an archetype of this form of disengagement. Outsourcing care to the private sector was seen in the early 2000s as a pragmatic response to the chronic underfunding of the health care system (Kober-Smith 2010). The growth in private health care provision has been accompanied by an increasing competition in the private health insurance market, leading to a reduction in insurance premiums. Thus, nearly 12% of patients who were dissatisfied with the quality and timeliness of the mandatory system were able to subscribe to an alternative private individual or collective insurance at a rate deemed acceptable (Foubister et al. 2006). Disengagement from compulsory health insurance, taking one or more of the three forms of disengagement identified, is present in all European systems. This phenomenon argues in favour of the hypothesis that private insurance would fill the space thus abandoned by public systems. However, as the evolution of OOP payments shows, the trajectories of each country may diverge according to local policies or measures. 2.2  From Disengagement to the formation of a Private Health Insurance Market An analysis of the trends in public and private financing of health expenditures in each country shows that public disengagement has taken a different scale in each European country. These developments provide information on the scale, scope, nature and evolution of private financing of health care and help to characterise the private insurance market in each country. Broadly speaking, changes in public health care financing allow a variable share depending on the country, with different distributions between households and private insurers. The question is then to determine the criteria on which this distribution is based.

5  AN INCREASING HOMOGENISATION OF PRIVATE HEALTH INSURERS… 

135

Figure 5.2 shows the first set of answers by cross-referencing the coverage rate of public and social health expenditures by country in 2017 with the evolution of the share of public financing of health expenditures in the EU countries over the period 1991–2017.3 This figure confirms, first, the relatively high level of the share of public and social financing in health (It is less than 65% for only five countries and only the Netherlands has a rate less than 50%). It also illustrates the general trend towards a disengagement from public systems (only five countries increased their share of public and social health expenditures over the period). However, neither the level of this share of public and social health expenditure, nor its evolution over time, nor the nature—insurance or not—of the health system is sufficient to explain the development of the private health insurance market. For example, in Fig. 5.2, there is relatively little proximity between France, Ireland and Slovenia, while these three countries have a commonality in their large and rapidly growing private insurance markets. Indeed, the share of public health expenditures in these three countries, although

Fig. 5.2  Level and evolution of share of public spending on health (1991–2017). (Sources: OECD & World Health Organisation (WHO) data, 2019) 3

 Except Luxembourg, for which the data cannot be used.

136 

P. ABECASSIS AND N. COUTINET

high, differs significantly (79% in Slovenia, 76% in France and 72% in Ireland) and the trajectory of expenditures diverges completely over the period (significant decrease in expenditures in Slovenia, stability in France and growth in Ireland). Finally, the French system is more insurance-based than the Slovenian or Irish systems. In his 1982 article, Smith proposes three ingredients to characterise a market: the environment, institutions and behaviours. The environment describes the market structure. These are the nature of competition (monopoly, oligopoly, etc.), the resource and information allocations of agents and their preferences. Institutions are the set of rules that organise the market as well as the institutions and mechanisms—including legal ones—responsible for their implementation. Behaviours refer to the expectations, anticipations and actions of market participants. In the context of private health insurance, the principles proposed by Smith (1982) lead to an interest in the initial structure of the market, its size and effectiveness in characterising its environment, as well as the role of the health insurance system and the nature of contracts (individual or collective) in characterising institutions4. The initial market structures vary widely from country to country (Table 5.3). The private health insurance market ranges from oligopolistic situations such as in the Baltic countries (two to five insurers in 2011) and Ireland (four insurers) to open and highly competitive markets, such as in France (713 insurers) or Poland (220). However, the coexistence of forprofit and non-profit insurers makes characterisation quite complex. Thus, only one of the two Estonian insurers is for-profit. On their side, while France has a larger number of insurers than Poland, for-profit insurers are more than twice as numerous in Poland (219) as in France (92). To characterise the environment, we prefer to use the size and efficiency of the market (Fig. 5.3). The market size is measured by the product of the share of the population covered by voluntary private health insurance (vertical axis) by the share of voluntary health expenditures (horizontal axis).5 Market efficiency is approximated by the ratio of these two indicators. In  Behavioural analysis is not studied in this work. It is assumed that the European countries selected are economically close enough to each other not to consider significant differences in the behaviour of the insurers on the health insurance markets, ceteris paribus. 5  The diversity of health systems does not allow this indicator to be constructed on the basis of a distinction between private and public insurance. This makes no sense, for example, when private insurers participate in compulsory health coverage (the Netherlands and Germany). To overcome this difficulty, only voluntary private insurance is taken into account here. 4

5  AN INCREASING HOMOGENISATION OF PRIVATE HEALTH INSURERS… 

137

Fig. 5.3  Relationship between private health insurance coverage and voluntary health insurance (VHI) spending. Data 2011—(e) estimate. (Sources: OECD data, 2019)

this figure, countries above the linear regression line have a less efficient market than the average, and those below this line have a more efficient market than the average. For example, the Maltese market appears to be more efficient than the Cypriot and Finnish markets because it covers the same population as the other two countries (about 21.5%) for a lower share of voluntary health expenditures in total health expenditures (1.7% against 4.1% and 5% respectively in Cyprus and Finland). The juxtaposition of the three criteria used to characterise the environment leads to five groups of countries (Table 5.1). The first group includes countries where all indicators suggest that the health insurance market has significant growth potential: the share of public expenditures in total health expenditures is relatively lower than that in other European countries, the OOP payments is very high and the share of the population covered by private insurance is particularly low. In addition, the private insurance market seems to be more efficient than the average for European countries. The next three groups include countries in which the private health insurance market has strengths but suffers from one or two disabilities. For example, the share of mandatory public health expenditures is

a

The underlined characteristics correspond to advantages for the class health insurance market

Source: authors, based on OECD and Sagan and Thomson data, 2016

– Mostly supplementary –  Complementary (Denmark and Croatia) –  Substitutive (Czech Republic and Germany)

Less than 15%

More than 75%

High

Varies – Mostly Complementary according –  Supplementary to country increasing

More than 80%

Below 18%

More than 75%

More than 30%

Average to – Mostly supplementary high –  Complementary increasing

Less than 25%

– Mostly supplementary (Bulgaria, Latvia) –  Substitutive (Cyprus) – Mostly supplementary

Between 18% and 20%

Average

High

Role of private insurance

More than 70%

More than 45% Less than 25%

Proportion of the Market population covered efficiency by voluntary insurance

Less than 25%

Less than 70% Less than 70%a

1. Narrow market with high potential Bulgaria, Cyprus and Latvia 2. Narrow market with medium potential Malta, Lithuania, Hungary, Greece and Portugal 3. Narrow market with medium potential in a complementary role Spain, Italy, Poland, Estonia, Romania, Finland, Slovak Republic and Austria 4. Large market with average potential in a supplementary role Slovenia, Netherlands and France 5.  Large market with low potential Sweden, Belgium, United Kingdom, Czech Republic, Croatia, Germany, and Denmark

OOP payments, as a % of total health expenditure

Between 70% and Between 30% 75% and 40%

Share of the compulsory system in total health expenditure

Class, type of market and countries involved

Table 5.1  The five types of private health insurance markets in Europe

138  P. ABECASSIS AND N. COUTINET

5  AN INCREASING HOMOGENISATION OF PRIVATE HEALTH INSURERS… 

139

higher than in group 1 and the OPP payments are lower. The combination of these two elements reduces the growth potential of a possible private health market. However, this growth seems to be concentrated on the complementary role of private insurance. The last group is diametrically opposed to group 1. It corresponds to an almost saturated market. The share of public health expenditures is very high; a large part of the population is covered with relatively low OOP payments. However, the maturity of this market is reflected in a high degree of efficiency that can create ‘niches’ depending on the roles played by the health insurance system in each country. The role of voluntary private insurance and the type of contracts, either collective or individual, differ from one country to another. In France and Slovenia, the countries with the highest coverage rates, it mainly plays a complementary role, highly regulated by the public authorities and accompanied by tax incentives. In France in particular, by making private insurance compulsory for employees of private companies, the public authorities have institutionalised it next to public health insurance. At the same time, supplementary insurance is developing, covering goods and medical care that are not covered by regulated complementary contracts. In Croatia, health insurance, authorised in 1993, also plays a mainly complementary role. Until 2001, it could only be offered by the public system. Only substitutive insurance was the responsibility of private actors. Since 2004, supplementary insurance has emerged but has only grown slowly due to the weight of the public system. Since then, private insurance has played a complementary role and is offered by the public system as well as by private actors, increasing by 5% between 2000 and 2014. In the Czech Republic, Cyprus and Estonia, private insurance has a substitutive role for certain categories of the population, mainly non-­ European migrants, excluded from the public system. It also plays a supplementary role by providing faster access to certain practitioners. The place of private insurance is marginal in the Czech Republic and Estonia, more important in Cyprus. In these countries, private insurance market developments are highly dependent on those of health policies and in particular on the size of future OOP payments. In other European countries, voluntary private insurance mainly plays a supplementary role. Overall, it provides faster access to better quality and coverage for care not covered by public systems, particularly dental care. This supplementary coverage is dominated by collective contracts. In some countries, these can represent more than 90% of contracts, as in Denmark, Bulgaria and Poland (Table 5.2).

140 

P. ABECASSIS AND N. COUTINET

Table 5.2  Nature of voluntary health insurance contracts in countries where supplementary voluntary insurance prevails Contracts mainly collective

Malta (70%); Portugal (64%); Bulgaria (98%); Denmark (90%); Finland, Spain, Greece, Lithuania and Poland (90%); Slovak Republic, Sweden and United Kingdom (75 %)

Contracts mainly individual

Italy, Germany, Cyprus (60%); Austria

Source: Sagan and Thomson (2016a, b) Data are not available for Belgium, Croatia, Estonia, Hungary, Latvia and Romania

Supplementary private insurance is the predominant form of insurance in Europe, but it is developed in various ways. It is almost non-existent in Lithuania or Sweden and in Bulgaria where the people covered are a few employees of multinational companies. In these countries, the growth potential of the market differs. It has been high in Bulgaria since the 2005 law which modified the basket of care offered by the public system by distinguishing between basic and complementary care. In the long term, additional care can be covered by voluntary insurance. In this country, where the public system covers less than 60% of health expenditure, the number of actors increased between 2001 and 2011, from 2 to 11 actors. Despite these elements witch would promote market growth, the market is currently struggling to develop. In Lithuania, since 1996 almost all governments have included the development of voluntary insurance in their programmes. Its development has been encouraged since 2007 by tax subsidies; however, it is hampered by the population’s low purchasing power, the importance of public health services and the relatively low OOP payments. In both countries, given the low purchasing power of households, the development of voluntary insurance depends on the development of collective contracts in companies. Finally, the growth potential of voluntary insurance in Sweden is very low given the quality of care provided by the public system. In a number of countries, the coverage rate of private voluntary insurance increased significantly in the years 2000 and 2010. Thus, in Greece, since 1998, various measures have been taken to encourage the development of private insurance. A first round, abolished in 2007, consisted of tax incentives. Since 2011, a second set of measures has allowed public hospitals to contract with private insurers. As a result, the coverage rate increased from 2% of the population in 1980 to 12% in 2016. In Portugal,

5  AN INCREASING HOMOGENISATION OF PRIVATE HEALTH INSURERS… 

141

the private insurance market was opened in 1980 and, since 1998, tax incentives have allowed households to deduct their health spendings and premiums paid to insurers from their taxable income. Although these tax benefits were reduced in 2012, they have allowed the market to grow. The proportion of the population covered has increased from 17% in 2011 to more than 26% in 2016. In these two countries, and without possible correlation, the market has been developed mainly as a result of tax incentives. Hungary is also a country in which the share of private financing of health expenditures has increased significantly. In Hungary, since 2012, the government has encouraged the development of voluntary insurance through significant tax incentives, particularly for companies that offer contracts to their employees. In other countries, however, particularly the most Beveridgian countries such as Spain, Italy or the United Kingdom, the private insurance market is underdeveloped. In Italy, the private insurance market was opened in 1992. While the objective was to develop a substitutive insurance, it actually plays a supplementary and complementary role, encouraged by tax incentives for households and firms. Despite these measures, the share of private voluntary insurance financing increased by only 0.05% between 2000 and 2014. In the United Kingdom, tax relief for people over 60 years old was adopted by the Conservatives in the 1990s and abolished in 1997 by the Labour Party. Since then, private voluntary insurance has not benefited from any special measures. This is also the case in Spain, where the coverage rate varies by region but remains relatively low. These examples illustrate the situation in countries where private insurance is developing only very slowly despite the growing OOP payments. Romania, Estonia and Latvia, where the compulsory health system is largely insurance-based, are also characterised by low rates of private insurance coverage. In these countries, no dedicated legislative measures are implemented. Health insurance is therefore regulated in the same way as all insurers. In Latvia and Romania, the low level of health coverage can be explained by the low purchasing power of households and the particularly high cost of individual health contracts. The existence of informal care payment mechanisms also seems to explain the persistent weakness of private insurance. The development of private insurance in Europe since the early 1990s coincides, in many countries, with the disengagement of public or social health systems. This disengagement leads everywhere, whatever its scope, to an increase in private financing of health expenditure. In a number of

142 

P. ABECASSIS AND N. COUTINET

European countries, this private financing almost entirely corresponds to the OOP payments. In others countries, however, it is composed, in various proportions, of private insurers and households spendings. The increase in the share of voluntary private health insurance is often encouraged by governments through various tax incentives for households and firms. However, these incentives are only part of the explanation of the dynamics of private health insurance markets. The initial state of the health insurance market, the role assigned to it, the local rules to which it must comply and the behaviour of the actors also play a major role. An analysis of this environment shows that the European directives and solvency rules determine a common European legal framework that has had little impact on these markets. Indeed, only the few countries in which private health insurance was developed—the mature markets—had to adapt. For most countries, particularly in Central and Eastern European countries, these regulations are constraints or directions to give to future markets.

3   The Concentration of Private Insurers in Europe In most mature health insurance markets, environmental changes, regulatory changes and government disengagement have led private health insurers to a significant movement of concentration since the early 2000s. This movement was marked by M&A and alliance transactions. Most of them do not go beyond the borders of each country. Despite its magnitude, this phenomenon of concentration is not or only partially studied. The most extensive work concerns mainly France because of the particularly high number of private health insurers and the scale of the concentration movement (Abecassis et  al. 2014). In other countries, when mentioned, the concentration of health insurance is not distinct from the concentration of the insurance sector as a whole. Finally, some European studies focus on the concentration of private health insurance in quantitative terms (Sagan and Thomson 2016a). This analysis of concentration in Europe is intended to be more qualitative. It is based on the systematic statistical exploitation of two financial databases, SDC platinum (Thomson Reuter/Refinitiv) and Zephyr (Van Dijk Office). These databases inventory all types of operations like

5  AN INCREASING HOMOGENISATION OF PRIVATE HEALTH INSURERS… 

143

mergers, acquisitions, joint ventures, strategic alliances, and so on.6 The quality of these databases is not uniform over time or space. Barnes et al. (2014) have thus shown that the SDC Platinum base is incomplete before 1984. After this date, the database is complete on transactions in which American firms are involved. The Zephyr database is more complete and accurate on small European transactions, but does not cover transactions prior to 1997 (Huyghebaert and Luypaert 2010). Finally, according to Bollaert and Delanghe (2015), the quality of information in the two databases is comparable, although the Zéphyr database is more precise on transactions involving several buyers or several targets. In view of these constraints, this study has selected the actual M&A transactions carried out during the period 1997–2018 in which a European health insurer is involved. A firm was considered as a health insurer when it indicated the activity ‘Accident and Health Insurance’ (industrial economics usually use USSIC code 6321) in the list of its main activities. After verification and cleaning of the database, 506 M&A transactions were retained over the period. These operations, which are relatively numerous, contribute to reduce the number of private health insurers, particularly in countries where it already played an important role, without contributing massively to a concentration of the sector (3.1), with the exception of certain countries. M&A have more generally reconfigured the markets of each country, at rates that are not very coordinated and not very correlated with the great wave of M&A observed in other sectors of activity (3.2) but explained by the micro-economic behaviour of actors that are more specific to the environment and legislation of each country (3.3). 3.1  A Reduction in the Number of Insurers that Does Not Significatively Increase the Concentration The concentration movement of private health insurers in Europe, which has been underway since the beginning of the twentieth century, experienced strong growth at the beginning of the twenty-first century (Fig. 5.4). After a peak in 2011, the main driver of the concentration, the number of M&A seems to have slowed down since then. At the beginning of this period, the European private health insurance market was already relatively concentrated. In 2006, in many Member States, the three largest 6  The Mergr database (Mergr.com) has been partially used as a complement or confirmation of some M&A transactions.

144 

P. ABECASSIS AND N. COUTINET

Fig. 5.4  Change in the number of M&A transactions inside and outside the EU (1997–2018). (Source: authors (Zephyr data))

suppliers held more than half of the market share. But this high average level of concentration hides wide disparities between countries and local regulations. Figure 5.5 illustrates these disparities: with a CR3 concentration ratio of 9%,7 France is particularly low concentrated while Lithuania, Denmark and Slovenia have very high concentration ratios (98%, 99% and 100%, respectively). The level of concentration of providers is not correlated with the weight of private voluntary insurance but rather results from national specificities. This can be illustrated by France and Slovenia. In both countries, the proportion of the population covered by private insurance is among the highest in Europe. However, in France the concentration ratio is low, or one of the lowest, while it is very high in Slovenia. The evolution of concentration over the past 20 years has been driven more by business partnership and M&A than by the internal growth of some insurers. These operations have naturally significantly reduced the number of health insurance operators. They were particularly numerous in 7  The CR3 is a concentration ratio representing the combined market share of the three largest firms in a sector.

5  AN INCREASING HOMOGENISATION OF PRIVATE HEALTH INSURERS… 

145

Fig. 5.5  The top three insurers’ market shares by country (CR3) in 2006 and 2011. (Source: authors, based on Thomson and Mossialos 2009; Sagan and Thomson 2016a, b.)

France,8 Finland, Greece, Italy, Spain and Portugal. In some of these countries, M&A is part of a more general trend towards concentration in the insurance sector as a whole. This is the case, for example, in the United Kingdom (Pearson 2013), Spain (Pons Pons 2013) or Germany (Borscheid 2013). For these countries, despite some local specificities, the concentration of insurance in the 2000s is mainly linked to the formation of global insurance companies. These large groups, mainly German (Allianz), British (Aviva, Prudential), French (Axa) or Italian (Generali), were created by acquiring subsidiaries abroad, particularly in ‘young’ EU countries such as Spain, Portugal or Greece (Harford 2005, Desombre et al. 2017). In other countries such as France, they can partly (but not exclusively) be analysed as a response to changes in legislation, in particular the application of European solvency standards (Defalvard and Laguérodie 2014). While the trend is towards a decrease in the number of actors, the opposite movement has been observed in some countries, in particular in 8  In France, where non-profit organisations largely dominate the complementary insurance market, the number of M&As recorded in databases largely underestimates the phenomenon of concentration. Indeed, the status of non-profit actors prevents them from accessing traditional forms of M&A. The actors then opted for original forms of rapprochement outside the markets, creating more or less strong interrelationships between them (Abecassis et  al. 2014). Unfortunately, the databases used do not identify these types of rapprochement.

146 

P. ABECASSIS AND N. COUTINET

the Central and Eastern European countries such as Romania, Lithuania, Bulgaria and Malta. In these countries, the degree of concentration has decreased due to their opening to the market economy and the arrival of new suppliers in an initially highly concentrated market. Overall, the number of private health insurers has actually declined in Europe. For the 19 countries studied by Sagan and Thomson (2016a, b) and Thomson and Mossialos (2009), it rose from around 1650 insurers in 2006 to 1216 in 2011 (Fig. 5.5). Paradoxically, while this decrease in the number of players has led to a better distribution of market shares, the average CR3 concentration rate has not increased. On the contrary, over the period, it went from 69.3% (2006) to 64.1% (2011). Nevertheless, the most active countries in terms of M&A (the United Kingdom, Germany, France and the Netherlands) have an increasing concentration between 2006 and 2011 (CR3 increases by 10.6%) associated with a significant reduction (−26.23%) in the number of insurers over the period. This development is linked to the coexistence of multiple statutes of private health insurers (for-profit insurers; non-profit insurers—mutual benefit societies, cooperatives, provident institutions—public insurers). In general, non-profit actors, especially mutual benefit societies, have a high degree of specialisation, while insurance companies are diversified over several risks—life and non-life—in order to use health insurance contracts as premium products for other types of contracts (Table 5.3). This distinction affects the way insurers respond to solvency legislation. In general, more specialised non-profit players are confronted with less complex technical risks than diversified insurers. On the contrary, for lucrative insurers, the complexity of operations constitutes a significant barrier to market entry and technical profitability is more difficult to preserve. For-profit or non-­specialised insurers must therefore consolidate their equity capital, in particular through insurer rapprochements enabling them to reach a sufficient critical size (ACPR 2016). Historically, non-profit insurers have dominated the market in many European countries. This is still the case in Belgium, Denmark, France, Ireland, Italy, Malta and Slovenia. However, their market share has declined in several countries since the 1990s due to competition from commercial companies and the disappearance of a large number of mutual benefit societies. This is the case in Finland, Denmark, Ireland, Malta, the Netherlands, Slovenia, the United Kingdom and France. In many European countries, commercial insurance companies are the only providers of private health insurance (Bulgaria, Greece, Hungary, Iceland, Latvia,

5  AN INCREASING HOMOGENISATION OF PRIVATE HEALTH INSURERS… 

147

Table 5.3  Number of private health insurers by status (2011) Country

Commercial insurers

Non-profit-making Total (year) Insurers specialising insurers in health

Austria Belgium

8 26

Mutuals: 1 Mutuals: 13

Bulgaria Croatia Cyprus Czech Republic Denmark

20 6 17 52

0 Statutory fund: 1 0 Statutory funds: 8

10

Mutuals: 1

Estonia Finland France

1 10 92

Georgia Germany Greece Hungary Ireland

14 24 ≈24 5 3

Italy

65

Statutory fund: 1 Mutuals: 140 Mutuals: 587 Provident funds: 34 0 Mutuals: 19 0 0 Quasi-public entity: 1 Mutuals: 3 Cooperatives: 1

Latvia Lithuania Malta

8 7 7

Netherlands Poland

9 (2011) 39 (2010)

No. Mutuals and very few commercial 19 (2012) No (2013) 21 (2010) Statutory fund 17 (2010) No. 60 (n/a) Statutory funds and very few commercial 11 (2011) Mutual and some commercial 2 (2013) Statutory fund 150 (2010) Mutuals 713 (2011) Most mutuals, half provident funds, some commercial 14 (2012) No. 43 (2012) n/a 24 (2011) Very few 5 (2012) n/a 4 (2015) Quasi-public entity 69 (2010)

0 0 Provident funds: 1 0 Statutory fund: 1

8 (2012) 7 (2011) 8 (2011)

Portugal Romania

33 Subscription: ≈200 Commercial: 15–20 19 0 Commercial: 12 Statutory fund: 1

33 (n/a) ≈220 (2012) 19 (2011) 13 (2012)

Slovakia Slovenia

3 3

0 Mutuals: 1

3 (2012) 4 (2010)

Spain

22

Mutuals: a few

≈30 (n/a)

All non-profit-making; very few commercial No. Very few Provident funds Few Subscription and statutory fund No. Subscriptions and statutory fund One Mutuals and one commercial Mutuals (continued)

148 

P. ABECASSIS AND N. COUTINET

Table 5.3 (continued) Country

Commercial insurers

Non-profit-making Total (year) Insurers specialising insurers in health

Sweden United Kingdom

17 11

n/a 7

17 (2013) 18 (n/a)

Very few Some of them

Source: Sagan and Thomson (2016a, b) Notes: Not all Czech commercial insurers offer VHI. Mutual funds in Hungary provide voluntary medical savings accounts and are not included For Ireland, we have excluded restricted membership undertakings, which limit membership to occupational groups and accounted for about 2% of those covered by VHI in 2010, and HSF Health Care (trading as Hospital Saturday Fund), which only sells cash plans In Italy, commercial insurers include collective private insurance funds (Fondi Integrativi Sanitari del Servizio Sanitario Nazionale) which may be for-profit or non-profit-making. The total number of entities selling VHI in Romania excludes subscriptions Some UK firms underwrite policies and do not sell VHI policies directly n/a = information not available

Lithuania, Netherlands, Norway, Portugal, Sweden and Switzerland). In other countries, these companies dominate the supply in number (Austria, Czech Republic, Finland, Spain and the United Kingdom). As a consequence, in many countries, the private health insurance market was concentrated on non-profit actors. However, their number has decreased much faster than that of commercial companies. The number of mutual benefit societies in the 19 countries included in Fig. 5.4 fell by 30% between 2006 and 2011, that of other non-profit insurers fell by 64%, while the number of for-profit insurers decreased slowly over the same period (−8.3%). These broad features of the concentration of private health insurance in Europe, based on global data, require a more detailed analysis. The objective is to better characterise the contours of the M&A wave observed in private health insurance and to measure the extent to which M&A operations are linked to the waves of concentration in the insurance sector in general. Finally, it is a matter of specifying the determinants and motivations that drive these operations. Most neo-classical economic research explains M&A waves by major technological or regulatory changes (Coase 1937). However, it is only when the liquidity of capital is sufficient to allow the reallocation of assets that a sectoral shock generates a wave of M&A (Harford 2005). Institutionalist work emphasises the major role of financial markets in M&A waves. This work assumes that bull markets

5  AN INCREASING HOMOGENISATION OF PRIVATE HEALTH INSURERS… 

149

encourage firms whose shares are overvalued to use these shares to buy real assets from undervalued targets through mergers (Shleifer and Vishny 2003). 3.2  A Mainly National M&A Rhythm M&A occurs most often in waves. These are generally observed in a context of strong technological and financial innovations, changes in market size, and economic and stock market growth. Europe in the early 2000s experienced a large wave of this type, driven by global competition, technological change and deregulation (Weston and Jawien 1999). Thus, the number of M&A transactions grew at an average annual rate of 13.42% between 1997 and 2011, from around 1200 transactions in 1997 to 7000 in 2011 (Holm 2014).9 Like other sectors, the insurance sector has benefited from this wave. According to Harford (2005, p. 539), it began in November 1998 and is mainly justified on the grounds of ‘Bigger is safer, leading to consolidation, especially in reinsurers’. Although no study distinguishes transactions by the type of risk insured, this study shows that the private health insurance sector has not escaped this trend. The average annual growth in the number of M&As was comparable to that of other sectors, at 14.91% over the period 1997–2011. The number of operations has thus increased from 7 in 1997 to 49 in 2011. However, this wave of M&A in private health insurance is not uniform across Europe (Table  5.4). Seven European countries account for three quarters of all transactions (France, Germany, the United Kingdom, the Netherlands, Spain, Italy and Denmark), whether the firms are acquirors or targets. Each of the seven countries driving the concentration process has a different profile. Although Fig.  5.6 should be analysed with caution, two indications can be drawn from it. On the one hand, in France, Germany, Spain and the United Kingdom, the Solvency I directives seem to have initiated a process of rapprochement. However, the Solvency II directives only gave rise to one in France. On the other hand, the data do not confirm a homogeneous reaction at the European level to the shocks constituted by the European directives insofar as there is no wave at this level.

9  The author points out the inaccuracy of these data. This is due to the discrepancies in the information provided by the source databases used.

Table 5.4  M&A deals by country (number and percentage: 1997–2018) Country

Number of Total Cumulated Number Total Cumulated acquirors acquirors (%) (%) of targets targets (%) (%)

United Kingdom Germany Spain France Netherlands Italy Denmark United States of America Austria Bulgaria Finland Czech Republic Ireland Luxembourg Poland Cyprus Norway Sweden Belgium Malta Switzerland Lithuania Romania Slovakia Australia Brazil Latvia Republic of Moldova Slovenia South Africa Argentina Portugal India Turkey International consortia (acquirors)a Other countries (targets)b Total

112 95 60 52 41 15 12 11

22.18 18.81 11.88 10.30 8.12 2.97 2.38 2.18

22.18 40.99 52.87 63.17 71.29 74.26 76.63 78.81

109 94 58 43 39 16 12 10

21.58 18.61 11.49 8.51 7.72 3.17 2.38 1.98

21.58 40.20 51.68 60.20 67.92 71.09 73.47 75.45

10 10 9 6 6 6 5 4 4 4 3 3 3 2 2 2 1 1 1 1

1.98 1.98 1.78 1.19 1.19 1.19 0.99 0.79 0.79 0.79 0.59 0.59 0.59 0.40 0.40 0.40 0.20 0.20 0.20 0.20

80.79 82.77 84.55 85.74 86.93 88.12 89.11 89.90 90.69 91.49 92.08 92.67 93.27 93.66 94.06 94.46 94.65 94.85 95.05 95.25

3 9 7 6 8 2 9 1 1 5 3 3 4 4 3 6 4 3 6 –

0.59 1.78 1.39 1.19 1.58 0.40 1.78 0.20 0.20 0.99 0.59 0.59 0.79 0.79 0.59 1.19 0.79 0.59 1.19 0.00

76.04 77.82 79.21 80.40 81.98 82.38 84.16 84.36 84.55 85.54 86.14 86.73 87.52 88.32 88.91 90.10 90.89 91.49 92.67 92.67

1 1 – – – – 22

0.20 0.20 0.00 0.00 0.00 0.00 4.36

95.45 95.64 95.64 95.64 95.64 95.64 100.00

1 – 3 3 3 3 –

0.20 0.00 0.59 0.59 0.59 0.59 –

92.87 92.87 93.47 94.06 94.65 95.25 –

–

–

–

24

4.75

100.00

505

100

100

505

100

100%

Sources: authors (Zephyr & SDC Platinum Data) Only transactions in which the acquiror or the target is in an EU country are included a At least one firm of the consortium is in an EU country b Albania (1), Bahrain (1), Botswana (2), Cambodia (1), Canada (1), Chile (1), China (2), Estonia (2), Gibraltar (1), Hong Kong (1), Liechtenstein (1), Montenegro (1), Republic of Korea (1), Russian Federation (1), Serbia (2), Singapore (2), Sri Lanka (1), Thailand (1), Ukraine (1) and Virgin Islands (1)

5  AN INCREASING HOMOGENISATION OF PRIVATE HEALTH INSURERS… 

151

Fig. 5.6  Pattern of evolution of the M&A count in the five most active countries (1997–2018). Quadratic regression. (Sources: authors (Zephyr & SDC Platinum Data))

Only national M&A movements are emerging, explained by the specificities of each health system. In France, the first wave is explained by the opening of the market following the new mutuality code of 2001, itself imposed by the 1992 EU insurance directives. Before 2013, the second wave is to be linked to the predominantly non-profit status of the actors, which forces them to reconcile in order to meet solvency requirements. In Germany, following the Seehofer reform of 1994, which aimed to better control and reduce public health expenditure, a series of measures were introduced between 1997 and 2002 to increase competition between health insurers and to introduce a co-payment that can be paid by insurers. It is also one of the countries driving the M&A waves in the insurance sector both as an acquiror and a target. These two elements may explain the strong movement of operations in the early 2000s, involving rather large players. In the United Kingdom, a country in which there are very few private health insurers, transactions are strongly correlated with financial market developments. It is in this country that the percentage of transnational transactions, often involving large insurers, is highest (43% of the total). Spain experienced a high concentration of the insurance sector (all risks combined) after its entry into the EU in 1986. This trend continued overall until 2010. The number of insurers has thus increased from 554 in 1991 to 287 in 2010. This is mainly due to the interest of European insurers in Spanish insurers,

152 

P. ABECASSIS AND N. COUTINET

which are the subject of numerous acquisitions and whose effect is to demutualise the sector. It is also the result of Spain’s adaptation to European prudential standards (Pons Pons 2013). In addition, the decentralisation of the Spanish health system, the first phase of which was completed in 2002, probably played a role in this wave. In the Netherlands, the European directives do not seem to have had an impact despite a large and relatively stable number of M&A transactions. The existence of an old competitive regulation of the health insurance system, dating from the end of the Second World War, may be one explanation. Private health insurers, which are exclusively for-profit, have a long-standing history of strong competition on premium levels, leading to early concentration of the sector. However, this increased after the 2006 reform. Thus, in 2005, there were 57 social and private insurances, for 33 at the beginning of 2006 and four in 2011 (Bernstein 2011). The analysis of M&A movements in these five countries shows that, while European regulatory policy does have an impact, the national specificities, their evolution and the conditions for exercising health insurance are superimposed with greater intensity. They are locks and/or accelerators of European regulation. In other words, in the health insurance market, players seem to react more to micro-economic determinants than to European regulations or, more generally, to macroeconomic variables usually used to explain M&A waves. 3.3   The Preeminence of the Micro-economic Determinants of M&A The economic literature has shown that M&A have micro-economic determinants related to firms’ external and financial growth strategies. The relevance of each one varies from one sector or company to another. Among the most frequent determinants are the seek for economies of scale and scope, the looking for synergies, the reduction of competition, the diversification, the penetration of new markets, the acquisition of knowledge and the behaviour of managers (Coutinet and SagotDuvauroux 2003). More specifically, acquisitions allow for rapid growth, if the target is active in the market and has its production capacity, its distribution network and its customers. In addition, acquisitions may be less costly than organic growth, particularly when the replacement cost of targeted assets exceeds their market value. Finally, and unlike internal growth, M&A can be partially or entirely financed by shares. They are then

5  AN INCREASING HOMOGENISATION OF PRIVATE HEALTH INSURERS… 

153

preferred by companies with low liquidity reserves and/or limited debt capacity (Huyghebaert and Luypaert 2010). Some of these reasons also determine the type of transactions. For example, the penetration of new markets or the search for synergies gives rise to horizontal transactions while the diversification of activities will result in vertical transactions. An analysis of M&A transactions carried out over the period 1997–2018 in the private health insurance sector shows that some of these determinants are particularly relevant in this sector: the search for economies of scale and scope; the diversification of activities and risks as well as the penetration of new markets. Thus, in France, Germany and to a lesser extent in Italy and the United Kingdom, the share of horizontal operations is significant and indicates that the operations are motivated by the search for economies of scale and scope (Fig. 5.7). French insurers have sought to reduce their management costs (claims management, customer acquisition, ‘tiers-payant’ management, etc.). In addition, thanks to common platforms, for-profit private insurers were able, for example, to quickly penetrate the health insurance market and non-­profit insurers improved their positions on this market. The merger

Fig. 5.7  Deals by type of concentration, in %. Number of transactions in which the country is the acquiror. (Sources: authors (Zephyr & SDC Platinum Data))

154 

P. ABECASSIS AND N. COUTINET

has also provided companies with new opportunities to reshape the market by developing new strategies. As a result, more-concentrated health insurers have developed patient selection processes to strengthen their solvency and have partnered with service providers to increase their profitability (Abecassis et al. 2014). The diversification of activities and risks through vertical operations dominates in Belgium and Denmark. In this country, M&A with nonhealth insurers accounts for half of vertical operations. This may result from solvency constraints that lead insurers to merge with health insurers because the associated risk is short-lived (such as motor vehicle risk, property damage, natural disasters and transport) and requires relatively small provisions to be made, given the speed at which these provisions are liquidated. In addition, this type of risk allows for rapid premium adjustments. To refine the analysis, the reason for penetration of new markets through external growth can be analysed on the basis of the share of transnational operations in the total (Fig.  5.8). On average, a third of the

Fig. 5.8  Opening and vulnerability of health insurance markets in Europe. (Sources: authors (Zephyr & SDC Platinum Data))

5  AN INCREASING HOMOGENISATION OF PRIVATE HEALTH INSURERS… 

155

operations are transnational, mainly between geographically neighbouring countries (France, Germany and the United Kingdom), despite differences in their national health systems. Denmark stands out with an external operations rate of almost 74%. In this country, while complementary insurance has existed for many years, supplementary insurance has grown strongly since 2002, following measures to encourage the acquisition of group contracts. Private domestic and foreign insurers have positioned themselves in this new market through external growth operations (Sagan and Thomson 2016b). Although French transnational operations are slightly below average, French insurers clearly pursue a strategy of market conquest: while transnational M&A represents 24% of total operations, the situations in which a French insurer is an acquiror are 4.5 times more important than those in which it is a target (20.37% as an acquiror and 3.7% as a target). To a lesser extent, the Netherlands, Spain and the United Kingdom are also countries in which insurers use M&A to penetrate new markets. On the contrary, Italian insurers are mainly targets: Italy is the only country with a negative difference between the number of transactions in which a health insurer is an acquiror and the number in which it is a target (−5.66%). The role of private health insurance affects the organisation of the market. Countries in which private insurance plays a supplementary role have significantly more operations outside the EU than others. Thus, while the number of operations is 9.8% in countries where health insurance plays a supplementary role, it is only 3.7% for countries with duplicative health insurance and 0.8% for those in which insurance is complementary. This is likely related to the level of dependence on the mandatory system and the regulations associated with each role. Supplementary insurance, which is largely independent of the compulsory system or even of the other roles assigned to health insurance, is thus the least regulated and most standardised. Substitutive insurance, by offering alternative insurance in comparison with the compulsory system, is, if not legally at least economically dependent on the reference compulsory system. Finally, complementary insurance depends largely on the functioning of the compulsory system, whose reimbursements it complements. Under these conditions, it is not surprising to observe that supplementary insurance is the most active in terms of M&A. The supplementary insurance is also, in countries where several roles overlap, the segment of the private health insurance that is growing more rapidly.

156 

P. ABECASSIS AND N. COUTINET

Table 5.5  Type of concentration according to the role of private health insurance Sector and concentration type

Complementary role (%)

Substitutive role (%)

Supplementary role (%)

All roles (%)

Health insurance  Vertical  Horizontal Other sectors  Vertical  Horizontal Provision of care  Vertical  Horizontal Total

40.68 11.02 29.66 50.00 25.42 24.58 9.32 9.32 0.00 100.00

83.65 11.54 72.12 13.46 7.69 5.77 2.88 2.88 0.00 100.00

54.62 22.27 32.35 27.73 13.87 13.87 17.65 13.45 4.20 100.00

57.61 16.96 40.65 30.22 15.43 14.78 12.17 10.00 2.17 100.00

Sources: authors (Zephyr & SDC Platinum Data)

The role of health insurance does not affect the form of rapprochement (Table 5.5). The majority of M&A transactions are horizontal (40.65%) and are carried out between specialised health insurers (57.51%), regardless of the role assigned to private insurance. However, this result is more pronounced when the insurance is substitutive (72%). On the contrary, while private insurance plays a complementary role, operations are carried out with other insurers whose main activity is not focused on health insurance. Overall, relative to other health insurance roles, complementary health insurance markets are not very active in M&A. However, they mainly promote national and transnational diversification. Supplementary health insurance markets strongly encourage mergers to penetrate new markets or benefit from synergies between insurers. Finally, substitutive insurance markets do not particularly favour rapprochement operations. When they are carried out, the vast majority of M&A are aimed at seeking economies of scale and synergies in their main health insurance activity.

4   Conclusion The European directives applying to insurers were intended to organise a large insurance market guaranteed by strict common solvency rules. The various European countries have, for their part, delegated part of their prerogatives in terms of health coverage to private insurers. Private health insurance markets have emerged or are in the process of emerging from

5  AN INCREASING HOMOGENISATION OF PRIVATE HEALTH INSURERS… 

157

this evolution. These elements should have constituted a regulatory ‘shock’ generating a wave of European concentration in health insurance. The analysis of the sector’s M&A does not confirm this effect. While waves of M&A are clearly visible, they are not coordinated with each other and do not coincide with the financial M&A wave observed in the insurance sector in general. On the contrary, it appears that M&A operations respond to local and micro-economic determinants. This contribution has shown the importance of the initial environment and local institutions regulating markets. In particular, the role assigned to private health insurance determines an important part of M&A behavior. This result does not, however, totally exclude the possibility of an influence of European regulatory policy. It can indeed be inferred from this work that three levels of determinants influence concentration operations. At the first level, the least important in terms of impact, M&A transactions are part of a wave of financial M&A affecting all sectors from which they can only benefit. This wave, linked to the growth of the financial market, has also been beneficial to the insurance sector as a whole in penetrating new markets and reducing solvency risks. At a second level, European policy influences M&A operations, not in the form of a regulatory shock but rather as an accelerator (Insurance Directives) or as a brake (solvency constraints) on opportunities. These led insurers as a whole to invest in foreign markets. Constraints led them to adopt diversification strategies, including approaching other health insurers because of the short-term risk associated. Finally, at a third and most significant level, national reforms and regulations provide M&A opportunities. This is the case, for example, for French (generalisation of complementary health insurance) or Danish (tax measures favouring supplementary company insurance) regulations.

References Abecassis, P., Coutinet, N., & Domin, J.-P. (2014). Les principes mutualistes confrontés aux modalités de regroupement des organismes complémentaires d’assurance maladie. Recma/Revue Internationale de l’Économie Sociale, 93(331), 60–75. ACPR. (2016). Éléments d’analyse assurance non-vie des cycles en assurance non-vie. Paris: ACPR-Banque de France. André, C., Batifoulier, P., & Jansen-Ferreira, M. (2016). Health Care Privatization Processes in Europe: Theoretical Justifications and Empirical Classification. International Social Security Review, 69(1), 3–23.

158 

P. ABECASSIS AND N. COUTINET

Barnes, B. G., Harp, N. L., & Oler, D. (2014). Evaluating the SDC Mergers and Acquisitions Database. Financial Review, 49(4), 793–822. Bernstein, D. (2011). Pays-Bas : la théorie de la concurrence régulée à l’essai. Les Tribunes de la santé, 30(1), 127–148. Bollaert, H., & Delanghe, M. (2015). Securities Data Company and Zephyr, Data Sources for M&A Research. Journal of Corporate Finance, 33, 85–100. Borscheid, P. (2013). La concentration de l’industrie de l’assurance en Allemagne. Entreprises et histoire, 72(3), 21–35. Coase, R. H. (1937). The Nature of the Firm. Economica, 4(16), 386–405. Coutinet, N., & Sagot-Duvauroux, D. (2003). Économie des fusions et acquisitions. Paris: La Découverte, coll. Repères. Defalvard, H., & Laguérodie, S. (2014). Les effets sur l’emploi de la concentration des Mutuelles de santé. Etude de cinq cas de fusions. Marne la Vallée: Chaire d’Économie Sociale et Solidaire. Desombre, N., Duverne, D., & de Montchalin, A. (2017). Les stratégies d’internationalisation en assurances. Revue d’économie financiere, 126(2), 51–64. Foubister, T., Thomson, S., Mossialos, E., & McGuire, A. (2006). Private Medical Insurance in the United Kingdom. Copenhagen: European Observatory on Health Systems and Policies. Harford, J. (2005). What Drives Merger Waves? Journal of Financial Economics, 77(3), 529–560. Holm, C.T. (2014). Anatomy of Recent European Industry—Level Merger Wave. Aarhus University: Business and Social Sciences. Huyghebaert, N., & Luypaert, M. (2010). Antecedents of Growth Through Mergers and Acquisitions: Empirical Results from Belgium. Journal of Business Research, 63(4), 392–403. Kober-Smith, A. (2010). “We Will Save the NHS”: la réforme du système de santé anglais sous le New Labour. Observatoire de la société britannique, 8, 69–86. Paris, V. (2014). L’impact de la crise économique sur les systèmes de santé des pays de l’OCDE. Médecine/sciences, 30(10), 910–915. Pearson, R. (2013). Mergers and Concentration in the UK Insurance Industry. Entreprises et histoire, 72(3), 7–20. Pons Pons, J. (2013). The Insurance Market in Spain: From Fragmentation to Concentration (1880–2010). Entreprises et histoire, 72(3), 36–54. Sagan, A., & Thomson, S. (2016a). Voluntary Health in Insurance in Europe: Role and Regulation. Copenhagen: World Health Organization, Regional Office for Europe. Sagan, A., & Thomson, S. (Eds.). (2016b). Voluntary Health in Insurance in Europe: Country Experience. Copenhagen: World Health Organization Regional Office for Europe. Shleifer, A., & Vishny, R. W. (2003). Stock Market Driven Acquisitions. Journal of Financial Economics, 70(3), 295–311.

5  AN INCREASING HOMOGENISATION OF PRIVATE HEALTH INSURERS… 

159

Smith, V.  L. (1982). Markets as Economizers of Information: Experimental Examination of the “Hayek Hypothesis”. Economic Inquiry, 20(2), 165–179. Thomson, S., & Mossialos, E. (2009). Private Health Insurance in the European Union. London: London School of Economics and Political Science. Turquet, P. (2012). Réformes du financement des systèmes d’assurance maladie aux Pays-Bas, en Allemagne et en France: quelles conséquences en matière de couverture sociale et de redistribution ? Revue internationale de sécurité sociale, 65(1), 33–56. Weston, F. J., & Jawien, P. S. (1999). Perspectives on Mergers and Restructuring. Business Economics, 34(1), 29–33.

PART III

Shifting the Public-Private Mix in Healthcare? Multifaceted Paths Towards Europeanization

CHAPTER 6

Private Health Insurance in Belgium: Marketization Crowded Out? Cyril Benoît and Marion Del Sol

1   Introduction Health coverage in Belgium is provided through a system of compulsory health insurance schemes involving, with very few exceptions, the entire population. Its organization and management are delegated to private mutual benefit societies and (to a much lesser extent) a public sickness fund (see below; Schokkaert et al. 2010). In practice, health coverage by these non-profit entities proceeds through the reimbursement of a wide array of health benefits and services. In this context, affiliation with a mutual benefit society or public sickness fund is compulsory, social contributions (consisting of a mix of employer and employee contributions) to levied rates are identical for all funds, and while there is freedom of choice with regard to the funds, the sector is closed to new entrants by law

C. Benoît (*) Centre for European Studies and Comparative Politics, Sciences Po (Paris), CNRS, Paris, France e-mail: [email protected] M. Del Sol Intitut de l’Ouest : Droit et Europe, Université de Rennes 1, Rennes, France e-mail: [email protected] © The Author(s) 2021 C. Benoît et al. (eds.), Private Health Insurance and the European Union, https://doi.org/10.1007/978-3-030-54355-6_6

163

164 

C. BENOÎT AND M. DEL SOL

(Schokkaert and Van de Voorde 2005). Contrary to what is observed elsewhere—like France, as Coron, Benoît and Houssoy explore in this volume—mutual benefit societies (organized similarly to their counterparts in other European countries) thus play a pivotal role in the Belgian system. As discussed at greater length below, this overall structure reflects long-­ term institutional dependencies and the persistent effect of initial choices that have proven to be enduring. Grouped into five associations at the federal level, mutual benefit societies themselves developed in line with political and religious boundaries. For example, “Christian” and “Socialist” Mutualities cover around 75% of the Belgian population (Schokkaert and Van de Voorde 2005). Like other continental European countries, Belgium has also seen the development of complementary health insurance over the last 40 years. In this respect, the Belgian case has followed the same general storyline as other Bismarckian welfare systems, whereby complementary schemes increased as statutory coverage receded or stagnated (Mossialos and Thomson 2004). This situation essentially originated amid growing public concerns about rising healthcare costs that resulted in several attempts since the 1980s to impose co-payments and tariff and supply restrictions. A distinctive feature of the Belgian system, however, is that most of the complementary and (initially) voluntary coverage was offered and organized by mutual benefit societies themselves. Some of these entities started developing complementary insurance at the beginning of the 1980s; by the end of the 1990s, all mutual benefit societies were providing their own complementary coverage, with notable variations in fees and covered benefits and services. The coverage was actually highly differentiated from one provider to the next, ranging from co-payment reimbursement to home and geriatric assistance, and even financial support for a sports club membership. Comparative research on health policy has consistently emphasized this dual role of mutual benefit societies in Belgium. This same statement applies to the overall legal framework in which they operate, since policymakers have clearly defined their role for statutory social security (i.e., public health insurance), but not for their complementary coverage, the existence and scope of which they were, until recently, free to determine (Stevens et al. 1998). The two integrated driving forces that the introduction of this book identified as a major source of change for the public-private mix in healthcare—namely, Europeanization and marketization—thus appear to play a

6  PRIVATE HEALTH INSURANCE IN BELGIUM: MARKETIZATION CROWDED… 

165

limited role in Belgium, due to the particular features of its healthcare system. Although they are private entities, mutual benefit societies were not initially considered trustees of the statutory social security scheme in the 1992 Insurance directives. The development of complementary health insurance did not significantly increase competition between mutual benefit societies, since it mostly developed as a quasi-extension of statutory coverage, even though its origins lie in a gradual retrenchment and drive to reduce healthcare expenditures covered by social security contributions. Considering the development of some managed care and risk-­management techniques over this time period, it appears that these initiatives were of limited scope and effect (Stevens et al. 1998). However, Europeanization, and to a lesser extent marketization, unexpectedly disturbed the public-private mix in Belgium in the 2000s. As co-payments increased, particularly for hospital care, where the most significant cost-containment efforts were made (see below; Schokkaert and Van de Voorde 2005; Perelman and Closon 2004), several mutual benefit societies considerably expanded their complementary health coverage, notably in this particular area. This trend mirrored similar developments initiated by for-profit insurance companies, which had started offering broad complementary coverage for hospital care. However, these companies were not competing with mutual benefit societies on a level playing field, since they were required to comply with the provisions of the 1973 and 1992 (non-life) Insurance directives. To facilitate their entry into this highly protected market, in 2005, insurance company representatives called on the European Commission to restore conditions for free and fair competition and requested that the complementary activities of Belgian mutual benefit societies be included in the scope of the Insurance directives. At both the national and European levels, this marked the beginning of a four-year period of intense political battles that ultimately yielded a healthcare reform to (re-)define the institutional boundaries of health insurance in Belgium. While several options were explored at the beginning of the process, mutual benefit societies succeeded in securing their overall position in the system. Insurance companies’ political efforts thus resulted in limited institutional change that translated into a 2010 reform. A range of complementary services offered by mutual benefit societies were considered to be “operations” that fell outside of the European directives’ remit. In effect, mutual benefit societies secured a monopoly position for a wide range of benefits. Yet the societies also defined a small area (mainly complementary coverage in hospital care) where mutual

166 

C. BENOÎT AND M. DEL SOL

benefit societies and insurance companies competed on an equal footing. As a result, the former were required to create specific societies in compliance with European law in order to continue offering complementary coverage in hospital care, as most of them would do in subsequent years. European influence and marketization ultimately grew and helped change the public-private mix in the Belgian health insurance system. But this influence was explicitly limited to a specific part of the system. The 2010 reform can thus best be described as a process of “path departure” in the public-private mix in Belgian healthcare (Ebbinghaus and Gronwald 2011). Indeed, the 2010 reform largely preserved the most salient features of the system’s institutional architecture. This assertion particularly holds in relation to mutual benefit societies, as most of their complementary activities (i.e., non-hospital) are now fully recognized and protected from competition by law. However, it is also true that Europeanization simultaneously resulted in a growing reliance on a newly institutionalized market including insurance companies for complementary hospital care. New institutional arrangements with divergent orientations were consequently added to the system in a typical case of institutional “layering” (Thelen 2003). The system’s core principles did not dramatically change, but some adjustments are apparent, since a private and marketized segment was explicitly established. As a by-product of the gradual retrenchment in statutory coverage, this trend might have greater effects than initially expected. Indeed, recent figures suggest an increase in complementary coverage by for-profit insurance companies. Given the general context, this chapter primarily focuses on two interrelated questions. First, it will provide an overview of long-term evolutions in the Belgian healthcare system, to better account for the development of complementary and private health insurance. The objective here is to emphasize the timing of the different historical sequences as well as the mutually reinforcing paths that led the insurance companies to file a complaint with the European Commission in 2005. A second objective is to explain why the 2010 reform adopted as a response to this complaint ultimately resulted in a path departure that reinforced several features of the Belgian system, while introducing new principles and segments through institutional layering. In line with Woll and Jacquot (2010), we assume that long-term institutional dynamics and the asymmetrical distribution of resources between the different players involved fall short of explaining this outcome (see also Itçaina et al. 2013). Rather, the strategic, cognitive and legitimating political work that

6  PRIVATE HEALTH INSURANCE IN BELGIUM: MARKETIZATION CROWDED… 

167

insurance companies and mutual benefit societies undertook early in the process played a critical role in shaping the 2010 reform. More specifically, these actors, using a variety of legal and political instruments, struggled to “use” the legal provisions in European legislation to advance their respective causes. While insurance companies initially imported some of these provisions at the domestic level to challenge rules governing the system, mutual benefit societies eventually reframed them in such a way that the bulk of their activities would qualify as compulsory insurance (and therefore be excluded from the directive’s scope). Still, concessions were made to European authorities and insurance companies, resulting in limited marketization. This reveals the high contingency of EU influence and marketization on the public-private health insurance mix, since it is (at least partly) conditional on the political work of a range of different actors, especially at the domestic level (see also Graziano et al. 2011). As such, our finding also provides some insights into the burgeoning literature on the role of private insurers in the structuration of social policy (see Naczyk 2013). The rest of this chapter is organized as follows: On the basis of prior historical research as well as legal and official documents, it outlines evolutions in the Belgian public-private mix in health insurance until the mid-­2000s (Sect. 2). Section 3 draws on a range of interview data and reports produced by governmental entities, mutual benefit societies, private insurers and European institutions to describe the process that resulted in the adoption of the 2010 reform, as well as its aftermaths.

2   Privatization Without Marketization 2.1  Healthcare Reform in Belgium: An Overview In close relation with the European legal framework, the Belgian public-­ private mix in healthcare underwent an important reform in 2010, when debates related to the marketization of mutual benefit societies topped the political agenda. The context of this reform, covered in the next section, created a risk of “path departure” (see Ebbinghaus and Gronwald 2011) from the big, slow-moving institutional arrangements (Pierson 2003) that had governed the Belgian public-private mix in health insurance since the post-war period. To a large extent, most issues related to competition between sickness funds and managed care, among others, have scarcely

168 

C. BENOÎT AND M. DEL SOL

been raised, let alone problematized, over the past 50 years—that is, since the foundation of the modern Belgian healthcare system. As comparative historical research has repeatedly shown, early legacy and path-dependent post-war dynamics in the formation of European Welfare states resulted in a mix of public and (more or less explicitly) private provisions in several areas of social protection ranging from pensions to healthcare. Belgium was no exception to the rule. The emergence of a new institutional framework in the form of a health insurance system in 1944–1945 generated long-term, path-dependent and self-reinforcing processes—and might be considered a critical juncture in the history of Belgian healthcare (see Ebbinghaus 2005). Indeed, a couple of decisions made during this period had both immediate and long-term implications. They entrenched frozen institutional landscapes that proved difficult to escape (Esping-Andersen 1996). The organization and the management of social security were delegated to mutual benefit societies essentially formed before the war (see Engels 1970; Geerkens 2002; Rezsohazy 1957). This initial choice contrasts the Belgian case with the French one— another country with a strong mutualist movement—where the governance of social security was delegated to labour unions at that time, relegating mutual benefit societies and provident institutions to a smaller role in the area of complementary health insurance (see Benoît and Coron 2019a). Seventy years later, this general structure has barely changed. Organized on a professional basis (employees, self-employed and civil servants), statutory (i.e., mostly through mutual benefit societies) health insurance in Belgium provides a high level of coverage. The unemployed, students and retirees also benefit from statutory system coverage. Membership in a mutual benefit society or public sickness fund (which assumes the same functions as mutual benefit societies) is compulsory, but the choice of an insurer is free (Schokkaert and Van de Voorde 2005). Seven entities are responsible for the organization and management of the statutory regime. Five are private mutual benefit societies: Alliance nationale des mutualités chrétiennes (ANMC, Christian), Union nationale des mutualités neutres (UNMN, Neutral), Union nationale des mutualités socialistes (UNMS, Socialist), Union nationale des mutualités libérales (UNML, Liberals) and Union nationale des mutualités libres (UML, Free). A sixth entity is the Service régional de la Caisse Auxiliaire d’Assurance Maladie-Invalidité (CAAMI), that is, the public sickness fund, and the last one is the Caisse

6  PRIVATE HEALTH INSURANCE IN BELGIUM: MARKETIZATION CROWDED… 

169

Table 6.1  Number of affiliates per Sickness Fund in Belgium (2013) Sickness Fund Mutual benefit societies ANMC

UNMN UNMS

Public UNML

UML

SNCB

CAAMI SNCB Holding 4.538.095 491.763 3.102.255 575.842 2.067.497 87.275 108.980

Total 10.969.707

Source: INAMI, Service du contrôle administratif

des soins de santé-SNCB Holding, which only covers national railroad company employees (Table 6.1). In many respects, the Belgian healthcare system’s path is similar to that of other Bismarckian Welfare states, such as France, Germany and the Netherlands. The stages are almost identical, although they were reached at different points in time. These steps relate to three broad dimensions (Hassenteufel and Palier 2007). The first is universalization, that is, the gradual inclusion in the social insurance system of categories initially excluded from statutory coverage. The second one relates to the growing coordination of healthcare system governance at the national (here federal) level. In Belgium, these two trends essentially started after the so-­ called Leburton Law passed in 1963. This law established the Rijksinstituut voor Ziekte en Invaliditeitsverzekering-Institut national d’assurance maladie et invalidité (“National Institute for Sickness and Disability Insurance”, NISDI) “a government agency responsible for health and disability insurance”, in charge of “supervising the sickness funds, negotiating fee schedules with practitioners and coordinating health policy in general” (Schokkaert and Van de Voorde 2000). In addition, NISDI determines the range of benefits covered and the level of spending in statutory insurance following consultation with practitioners and sickness funds (ibid). Meanwhile, statutory coverage of inpatient care was extended to the self-­ employed and was then further expanded over the next decades (Thomson et al. 2013). This long-term development was both accompanied, and to some extent reinforced, by a gradual strengthening in federal institutions’ coordinating capacity. Here, and just as in other Bismarckian Welfare systems, universalisation and federalization proved to be interdependent dynamics. This was the case in 1990, when a new law expanded the regulation of mutual benefit societies and increased their participation in the NISDI. The societies were also required to provide at least one additional

170 

C. BENOÎT AND M. DEL SOL

service in the field of social protection, assistance to ill and disabled people and healthcare. The third dimension of this evolution relates to the introduction of cost-containment policies and the development of regulated competition in healthcare—a trend that essentially started in the mid-1980s in Bismarckian welfare systems. In Belgium, it was essentially embodied by two laws adopted in 1993 and 1995 under the coalition government led by Prime Minister Jean-Luc Dehaene (Flemish Christian-Democrat Party, centre-right). These two policies were based on the same broad principles as the 1963 Health Insurance Law, but sought to organize the system in a more “rational and transparent way”, notably by enhancing risk equalization to make sickness funds shoulder some financial risk (Schokkaert et al. 2018). The prospect of introducing financial responsibility to sickness funds was hotly debated. Ultimately, partial financing was implemented. However, and contrary to what occurred in other countries, the goal of this reform was certainly not to “foster competition among sickness funds”; rather, the objective was simply “to place them on an equal footing and encourage them to control health care expenditure” (Thomson et al. 2013; Schokkaert and Van de Voorde 2000). The reform actually resulted from a compromise at the federal level between the political sphere and the entities governing the healthcare system, mostly mutual benefit societies. Christian mutual benefit societies in particular considered that the containment of rising healthcare costs necessitated the introduction of individual financial responsibility to increase efficiency. Socialist mutual benefit societies did not completely reject this principle. However, the idea was only acceptable to them on the condition of clearing healthcare deficits that have accumulated since the previous major reform (1963). Despite the introduction of prospective funding of sickness fund, the financial responsibility for budgetary non-compliance was “still taken collectively by all insurance funds” after this reform, as a result of the compromise between mutual benefit societies and the government at the time (Schokkaert and Van de Voorde 2005). Most healthcare reforms require several compromises to pass—between the government and mutual benefit societies and between mutual benefit societies as well. Most observers of the Belgian healthcare systems later described the conflictual and slow adoption of these changes as reflecting an overall preference over a system of cost-containment policies in bid to prevent the introduction of real microeconomic incentives and quality controls and, more generally, ambitious systemic reforms. The rationale behind most

6  PRIVATE HEALTH INSURANCE IN BELGIUM: MARKETIZATION CROWDED… 

171

policies adopted during the 1990s has even been depicted as driven by a “fear of market”, in a system where values of equity, solidarity and universal access to care are largely shared in the general population (Schokkaert and Van de Voorde 2005). Without contesting this broad-brush picture, it may be more accurate to posit that these sets of values endured primarily because of their mobilization in the policy process (Smith 2016) and their staunch defenders amid a range of key participants and veto players in Belgian healthcare— notably mutual benefit societies and their representatives. Interestingly though, the mobilization did not prevent the introduction of cost-­ containment measures during this period, especially in the hospital sector. After several tariff reductions (mostly between the end of the 1970s and the beginning of the 1980s), prospective financing schemes and supply restrictions were introduced, stemming the sharp increase in hospital expenditures observable through 1986 (Schokkaert and Van de Voorde 2005). These latter facts suggest that, despite an illusion of stability, the Belgian healthcare system experienced several processes of de-­ institutionalization and re-institutionalization that did not leave post-war compromises unchanged. 2.2  The Recent Surge in Complementary Health Coverage and the Growing Presence of Insurance Companies The difficulty of imposing systemic healthcare reform and the relative ease of introducing cost-containment measures are not necessarily contradictory. The combination is a recurrent feature of Bismarckian systems, as compared to Beveridgean systems. It is indeed well established that in Bismarckian systems, reforms tend to combine retrenchment measures with the continuation, expansion and improvement of existing programmes (Bonoli 2001). In Belgium, this broad development was amplified by some of the structural features of the country’s heavily fragmented constitutional system, characterized by numerous veto points—a fragmentation reflected in the overall organization of the healthcare system as well. The persistent influence of mutual benefit societies combined with cost-containment policies adopted between the 1980s and 1990s resulted in singular developments. From the 1980s onwards, several mutual benefit societies started offering complementary health coverage—a situation that generalized during the 1990s. Due to their position in the provision of health insurance under the statutory regime, these complementary

172 

C. BENOÎT AND M. DEL SOL

plans, though legally voluntary, became de facto compulsory for individuals affiliated with a mutual benefit society. Complementary health insurance emerged in Belgium without any formalized legal boundaries or defined coverage, which significantly varies from one provider to the next. Another peculiarity of complementary health insurance in Belgium is that it does not solely reimburse the share of health expenditures that is not covered under the statutory regime, as is most often the case with complementary health insurance (Mossialos and Thomson 2004). In line with the values of “social inclusion” and “solidarity” claimed by mutual benefit societies, a large array of benefits and services are covered in addition to the partial or full reimbursement of co-payments. These include, but are not limited to, medical transportation, home care, equipment leases in case of disability, reimbursement of homeopathic medicines, childbirth benefits and financial support to join a sports club. The fees and benefits covered significantly vary from one mutual benefit society to the next. More importantly, mutual benefit societies offer “small” risk coverage for the self-employed (such as visits to a general practitioner), who were not included in statutory insurance until recently, as well as hospital care— essentially offering superior accommodation rather than increasing choice or providing faster or better access (Thomson et  al. 2013). Until the beginning of the 2000s, larger mutual benefit societies thus developed increasingly large and complex complementary health coverage. Their offers, also increasingly integrated, embrace lower statutory coverage in some areas, especially hospital care. A prominent example of this trend is the case of Christian mutual benefit societies, which covered around 45% of the Belgian population (Lewalle 2006) by the early 2000s. As a growing share of their affiliates faced difficulties coping with out-of-pocket expenditures in hospital care, these societies created a whole complementary service called Solimut. Through this entity, they created a system of deductibles that members would need to pay for hospitalizations, based on a deductible per hospitalization as well as an annual deductible per family, independent of the number of hospitalizations. This complementary scheme covers all co-payments, supplements, drugs and medical devices (e.g. prosthetics). This service was offered on the basis of modest, family-based monthly contributions, to all members affiliated with Christian mutual benefit societies without age or medical restrictions, in line with mutualist principles. Between 1987 and 1995 alone, the total amount of premiums paid for complementary health insurance rose from five billion Belgian Francs to 12 billion

6  PRIVATE HEALTH INSURANCE IN BELGIUM: MARKETIZATION CROWDED… 

173

Belgian Francs—a figure that can be directly linked to cutbacks in statutory coverage (Stevens et al. 1998). This figure has continuing to grow ever since. Mutual benefit societies, however, are not the only players offering complementary coverage for hospital care. Gradual restrictions to statutory contributions have also created fertile ground for the increased presence in healthcare of for-profit insurance companies, which also developed offerings in this area. Several companies, such as DKV, have entered the complementary health insurance market since the beginning of the 1990s. They differ from mutual benefit societies in a number of ways, since they apply risk-selection techniques to set coverage and premiums, within the framework of individualized (instead of essentially family-based) contracts. However, their expansion in this particular segment of social protection was initially limited due to the dual position of mutual benefit societies and the growing coverage offered by their own schemes. In addition, and more fundamentally, insurance companies and mutual benefit societies were not competing on an equal footing. Indeed, the players needed to operate under the framework established by the EU’s 1973 and 1992 (non-life) “Insurance” directives, which imposed a number of requirements on them, especially in terms of solvency rules. The fact that the complementary health insurance coverage offered by mutual benefit societies has become de facto compulsory is an additional source of difficulties for the growth of insurance companies in the healthcare system. As a result, the stabilized institutional features and more recent developments in the Belgian healthcare system provide both a number of opportunities and barriers to insurance companies’ expansion in the area. In the 2000s, this issue was raised by several firms that mobilized within their representative body—Assuralia—to change the rules (or, more exactly, to institutionalize some rules) in the Belgian complementary health insurance sector. While these large companies have the strategic resources to promote their claims, they lack the legitimacy of mutual benefit societies to shape or change the principles governing the healthcare system as a whole. As aforementioned, the latter also have a range of prerogatives that give them solid veto powers and capacities to guide healthcare reform. To get around these institutional hurdles, insurance companies represented by Assuralia focused their political work on the legal provisions of the Insurance directive, in order to challenge the dominant position of mutual benefit societies in complementary health insurance. In this sense, they tried to make strategic “usage of Europe” (Woll and Jacquot

174 

C. BENOÎT AND M. DEL SOL

2010) in a mobilization that would ultimately constitute the basis of Europeanization of the public-private mix in Belgian healthcare. In 2006, insurance companies filed a complaint with the European Commission through Assuralia. It explicitly denounced the position of mutual benefit societies in complementary health insurance, especially their hospital care offerings. According to the insurance companies, the activities of mutual benefit societies did not comply with European law, as established in the Insurance directives. The situation was therefore one of unfair competition. In its response, the Commission sent a formal request to Belgium asking for additional information to assess the compatibility of the Belgian legal framework (especially the 1990 law on mutual benefit societies in the healthcare system) with EU law (especially the 1973 and 1992 Insurance directives). Without contesting “the structure of the Belgian social security system” or the right of sickness funds to provide complementary health insurance, the Commission expressed concern “that this could result in differing levels of policyholder protection and market distortions” (EC 2006). Accordingly, in 2008 the Commission asked Belgium “to amend its national rules so that the private sickness funds comply with the EU insurance directives when they offer complementary health insurance cover outside the scope of obligatory social security” (EC 2008). The European Court of Justice endorsed this opinion in 2010, in an order condemning Belgium for incompletely transposing the Insurance directives (CURIA 2010). Without frontally challenging the institutional background of the healthcare system at the domestic level, the insurance companies succeeded. Indeed, the “detour through Europe” (Woll and Jacquot 2010) forced Belgian authorities to introduce a number of changes in the rules governing the public-private mix in healthcare that initially appeared to favour insurance companies. European authorities called for an alignment of complementary health insurance activities with the provisions stipulated in non-life insurance legislation— that is, of mutual benefit societies on the norms and principles already governing insurance companies’ activities. The European Commission and the Court of Justice developed a threefold formal argument. It first argued that complementary health insurance could not be considered as part of a so-called “statutory social security regime” (i.e., public or quasi-public)—a qualification that would have legitimately excluded it from European legislation. To reach this first conclusion, the Commission argued that a statutory social security regime requires a genuine, solidarity-based organization at the national

6  PRIVATE HEALTH INSURANCE IN BELGIUM: MARKETIZATION CROWDED… 

175

level—rather than at the level of private mutual benefit societies, as is the case for complementary health insurance arrangements. Second, it argued that the complementary insurance could be considered as economic activities in the European sense. To demonstrate this, the European Commission established a causal relationship of sorts between different statements, mostly on the basis of negative evidence. It assumed a de facto market of complementary health insurance in Belgium without providing details about its features; then, it deduced the qualification of complementary health insurance as an economic activity from the inapplicability of the principle of (nationally based and defined) solidarity in this market; and from the existence of actual competition between mutual benefit societies and insurance companies. A third point is the absence of justification for the exclusion of mutual benefit societies’ complementary health coverage from the Insurance directives, especially since these texts provided superior protection to policyholders (notably in terms of solvency requirements) than the existing Belgian law. On this basis, Belgium was asked to adapt its legislation. As a product of the progressive and contingent alignment of several dynamics, the Europeanization of the public-private mix in Belgian healthcare thus took an unexpected path. Over time, some initial institutional choices granted a central position to mutual benefit societies in the governance and the organization of the healthcare system. This position solidified partly in spite and partly as a result of a growing integration of the system at the national level. In the medium term, mutual benefit societies, as pivotal actors, blocked or reshaped several reforms attempting to restructure overall funding of the healthcare system. This channelled most cost-containment and budgetary efforts towards areas where governmental leverage was greater, including (though not limited to) hospital care. As in other Bismarckian welfare systems, mutual benefit societies managed this gradual retrenchment by developing their own complementary health coverage. However, this coverage was based on the exact same principles as the statutory coverage they were already offering. It mostly consisted of contracts based on solidarity principles; coverage was provided at the family level and managed according to values claimed by the mutualist movement. As a result, a multi-pillar system per se did not form. This was also the case in terms of competition between health insurers and the introduction of risk-management techniques, which remained quite limited in Belgium (Stevens et  al. 1998). But the transformation also paved the way for a growing presence of for-profit insurance companies in

176 

C. BENOÎT AND M. DEL SOL

this sector—a development that was largely unanticipated. In response to their difficulties penetrating a market largely dominated by mutual benefit societies due to the institutional features of the Belgian system, insurance companies made strategic use of European legal provisions in 2005 to change existing rules in alignment with their operations. As a result, Europeanization in this case was not a vertical process guided by a conception of how the health insurance market should be governed. It was essentially used as a vehicle by challengers at the national level to contest disadvantageous hierarchies. At first, and according to the Commission and CURIA orders, the idea was to implement rules and regulations likely to promote competition between mutual benefit societies and insurance companies, as well as a range of transformations in the former’s governance and operating structures. But mutual benefit societies also then sought to reshape European constraints using resources they wielded within the Belgian healthcare system.

3   Path Stability or Path Departure? The Internal Process of Marketization and Its Aftermaths 3.1  Negotiating Europeanization Quite suddenly, European Commission rulings threatened the institutional arrangements that had governed the public-private mix in Belgium since the 1980s at risk. While the complaint raised by insurance companies explicitly focused on complementary health insurance for hospital care, the European Commission considered that the Insurance directives should apply to all complementary health insurance activities. The Commission does not call into question the role of the private sickness funds in the statutory health insurance scheme […] This is a matter which falls outside the scope of harmonised EU insurance legislation. However, the Commission believes that [complementary health insurance activities] must be conducted in accordance with the pertinent provisions of [Insurance] Directives because they offer better guarantees for policyholders and their insured relatives, in particular in the form or more stringent solvency requirements. European Commission (2006)

All non-statutory health activities developed by mutual benefit societies could thus be exposed to competition with insurance companies. Mutual

6  PRIVATE HEALTH INSURANCE IN BELGIUM: MARKETIZATION CROWDED… 

177

benefit societies clearly saw the writing on the wall and mobilized to shape the implementation process. When the European Commission attacked us, it did not distinguish between complementary health insurance and hospital insurance. So we had to clarify what should be considered as health insurance and what should not be removed from our field of competencies, as mutual benefit societies. The whole fight was about this. Interview, Head of European and International Affairs, Socialist Mutual Benefit Society Author’s translation from French

In response to the European Commission’s May 2008 letter, the Belgian government initiated a negotiation phase involving the representative bodies of insurance companies (Assuralia) and mutual benefit societies (CIN, for Collège inter-mutualiste national/ Nationaal Intermutualistisch College—“National Inter-mutualist College”). In this context, the political work of insurance companies proceeded alongside similar efforts by mutual benefit societies. While the former defined existing arrangements as problematic in light of European rules and regulations, mutual benefit societies tried to problematize the debate based on a polar opposite approach. They argued that the European legislative framework itself potentially threatened the Belgian healthcare system. Thus, the distinctive routes to implementing the Insurance directives were legitimized using two separate lines of argument. Insurance companies defended fair and equal competition and the protection of policyholders. On the opposite side, mutual benefit societies emphasized the need to protect (if not expand) the (mutualist) principles governing the healthcare system. Both discourses were firmly tied to the positions of their respective promoters, with insurance companies unambiguously challenging the incumbents, namely mutual benefit societies. Negotiations started on the basis of these two general conceptions, which barely budged during the process. Insurance company representatives used European Commission backing to legitimize their views. Mutual benefit societies did the same, drawing on their historical position within the Belgian healthcare system. Both groups of actors struggled to define the boundaries of the complementary coverage that should be subject to European law and as such, to competition. Four months later, a compromise was finally reached. Mutual benefit society representatives secured

178 

C. BENOÎT AND M. DEL SOL

the exclusion of most covered benefits and services from the scope of European legislation. To do so, they proposed that a share of complementary health coverage become mandatory but not integrated within the scope of the statutory regime per se. Meanwhile, insurance companies secured the application of the principle of “same level playing field” to particular segments of complementary health coverage where they were already developing activity (notably for hospital care, but also for other benefits such as dental care). At the end of this process, mutual benefit societies thus managed to protect a sizeable share of their complementary offerings from competition. However, from a legal perspective, this political achievement fell short of meeting the European authorities’ demands. A new chapter in the process thus opened in September 2008. The translation of the agreement into law mostly involved mutual benefit society representatives, the coalition government led by Prime Minister Yves Leterme (Christian-­ Democratic Party), and officials from the Belgian Ministry of Public Health and Social Affairs led by Laurette Onkelinx (Socialist Party). The Ministry’s strategic group unit in charge of European Affairs was especially involved in this process (MASSP 2009). This second phase can also be considered as a domestic usage of Europe, since mutual benefit societies and the Belgian government engaged in political work to align the boundaries of the healthcare system with European legislation, in accordance with the agreement negotiated with insurance companies. This process was quite different from the negotiation phase between the relevant parties, since the different actors (notably the CIN) tried to mobilize technical (mostly legal) arguments to meet the demands made by European authorities. To this end, they tried to invoke Article 2, § 2, b of the 1973 Insurance directives to exclude most of the complementary health coverage activities of mutual benefit societies from EU law. Art. 2—This Directive does not apply to […] 2) the following operations […] b) Operations of provident and mutual benefit institutions whose benefits vary according to the resources available and in which the contributions of the members are determined on a flat-rate basis 73/239/CEE We tried to use a subtlety of the Insurance directives to ensure that the complementary benefits and services offered by mutual benefit societies did not fall within the scope of European legislation. [We stated] that they

6  PRIVATE HEALTH INSURANCE IN BELGIUM: MARKETIZATION CROWDED… 

179

should comply with a range of conditions related to the very operations and nature of mutualist activity. Interview, Head of European and International Affairs, Socialist Mutual Benefit Society Author’s translation from French

This article grants member States great leeway in defining what should qualify as operations of this kind. Accordingly, a large share of mutual benefit societies’ complementary activities were defined as such in Belgium, and were thus protected from competition, while a smaller share (mainly related to hospital care) was integrated into the European legal framework. In this respect, the Belgian case illustrates States’ relatively significant capacity to shape the implementation of European law in this area. More fundamentally, it also reveals the strong influence of mutual benefit societies and their representatives in this overall process. Indeed, in other countries (typically in France), non-profit entities failed to use and obtain the activation of this article during the Insurance directives’ implementation process (see Benoît and Coron 2019b). The resulting law of 26th April 2010 had three main purposes. First, it defined the entities allowed to offer operations of this kind, namely mutual benefit societies and mutualist societies that do not offer any insurance products (see below). The law thus defined operations as all complementary health coverage activities already offered by mutual benefit societies, and thereby granted these entities a monopoly in the area. Second, it delineated these operations, defined as services providing their members and dependants with financial support in preventing or treating illness, or in providing compensation in the event of work disability or in situations where physical, social and psychological well-being could be nurtured. These operations also covered the provision of assistance, information and guidance to promote physical, social and psychological well-being. Thus, the scope covered most of the complementary coverage already offered by mutual benefit societies; in addition, the law did not define the form that these schemes should take. The mutual benefit societies offering the services were therefore afforded great flexibility, on condition that the services be related to health lato sensu (prevention or treatment of illness, but also the promotion of well-being) and offered and implemented in accordance with a set of precisely defined requirements. Accordingly, a third dimension of the 2010 law related to requirements on qualifying the services offered by mutual benefit societies as operations.

180 

C. BENOÎT AND M. DEL SOL

The first was mandatory affiliation. All members affiliated with a mutual benefit society under the statutory regime became legally obliged to pay a fee for these complementary services (this requirement was previously determined by the status of each mutual benefit society). In line with mutualist principles, the affiliation could not be limited due to age or health conditions. By the same token, fees were set on flat-rate basis and consequently not based on age or risk profiles, as they are for insurance contracts. Derogation from this principle was allowed in cases of high co-­ payment, for example, for low-income and disabled affiliates—that is, to lower and not to increase fees for particular segments or categories covered. The benefit provided had to be the same for all affiliates. More importantly, the law explicitly stated that the general assembly of a mutual benefit society had to approve possible adaptations of these operations’ provisions. So the legislation “obligated” mutual benefit societies to offer a range of operations to their affiliates, while granting them a high level of discretion in determining their own qualitative and quantitative investment in this area. Financial management did not change, since it was based on pay-as-you-go (see also Ebbinghaus 2015 for a discussion on pensions). This principle precluded the creation of risk-based technical provisions, even if financial reserves accumulated. Moreover, there was a disconnect between contributions and benefits, since the latter were theoretically provided from available resources. No insurance-like relationship existed since benefits were based on solidarity principles both in terms of their funding and delivery. Just as insurance companies used the European legal framework to contest the rules governing the health insurance market in Belgium, mutual benefit societies eventually succeeded in preserving their dominant position using the same strategy. While the introduction of market mechanisms (namely, competition with insurance companies on the basis of the rules governing these firms) was a clear possibility after Assuralia’s complaint filing, this alternative scenario was progressively quashed by mutual benefit societies’ political work. This was first achieved through a negotiation process with insurance companies wherein concessions were made in the area where for-profit firms were already players. Then, and together with the Ministry of Public Health and Social Affairs, technical and legal efforts were made to frame the 2010 Law using the leeway offered by European directives. Ultimately, the Belgian government’s response to the European Commission reflected this overall process, and most mutual benefit society demands were met (MASSP 2010).

6  PRIVATE HEALTH INSURANCE IN BELGIUM: MARKETIZATION CROWDED… 

181

During this process, mutual benefit societies also successfully managed to further secure their position in complementary health coverage. A major issue related to the framing of the 2010 Law was indeed the limit to place on the contributions that mutual benefit societies could collect to fund their operations. A relatively high limit would give mutual benefit societies the option of offering large benefits to their members; on the flipside, a relatively low limit would give insurance companies an opportunity to provide superior coverage. Mutual benefit societies obtained a limit set at 250€ per “mutualist household” in a 2011 Royal Decree. This amount was much higher than the basis negotiated with insurance companies in 2008. Mutualist representatives justified the figure as a means of securing most of the benefits they were already covering. From a governmental point of view, strengthening mutual benefit societies’ monopoly also paved the way for the introduction of cost-­ containment policies and cuts in health spending without disturbing the market—since it could be done through an increase in contributions collected by mutual benefit societies without involving provider changes. Thus, it institutionalized a public-private mix without changing the fundamental nature of the system. Insurance companies strongly opposed this choice. Despite their protests, insurance companies’ representatives were unable to change this figure after it was implemented. It is in the area of operations that the insurance industry accepted the most important concession, and it is precisely on this point that the government deviated the most from the agreement […]. The legislator created a potential monopoly of 1,75 billion euros for mutual benefit societies. This amount is four times higher than the revenues from insurers’ individual contracts in hospital care! Philippe Colle, Assuralia CEO (2014)

3.2  The Path to Departure? The limited Europeanization of the Belgian healthcare system is hardly surprising. As actors dominating both statutory and complementary health coverage, mutual benefit societies have enjoyed a wide array of legitimating and strategic resources to make a more effective use of the European framework than the insurance industry—an outsider in this area. At the end of this process, the role of mutual benefit societies in complementary health insurance was explicitly recognized and guaranteed by

182 

C. BENOÎT AND M. DEL SOL

law. Governmental support also helped them secure a monopoly over a significant number of benefits. To a large extent, their position was strengthened by the 2010 reform, contrary to expectations that the sector would open to more competition. However, this general process was not just an adaptation to environmental changes that unfolded without affecting the system’s core principles. By introducing marketization to a very narrow segment of private health insurance, the 2010 reform resulted in a limited redirection of core principles and a partial renewal of existing institutional arrangements (Ebbinghaus 2005)—that is, it induced a path departure instead of path stability. Earlier decisions effectively narrowed the choice made (strengthening the position of mutual benefit societies and marketizing a small segment of complementary health insurance). But given the provisions that were eventually established, they did not necessarily determine the next adaptive step (Ebbinghaus 2006). Indeed, the 2010 reform also created a space where European law and regulations applied, and where mutual benefit societies competed on a strictly equal basis with insurance companies. As such, it developed a new “pillar” in the Belgian public-private mix in healthcare that can be defined as a form of voluntary and supplementary health insurance. All activities beyond the scope of mutualist operations and their other activities that were not operations per se fell into that category, which mostly encompassed health insurance in dental and hospital care. Simultaneously, the reform defined a new framework of operation for mutual benefit societies in this area, in accordance with the Insurance directives and in anticipation of the subsequent Solvency II directive. The most prominent change was the application of the specialty principle, meaning that mutual benefit society activities considered as insurance (i.e., non-operations) now had to operate through a dedicated and separate entity—a (non-profit) mutualist insurance society (SMA, Société mutualiste d’assurance/ Maatschappijen van onderlinge bijstand). All (11) mutual benefit societies have created a SMA since the reform came into force. To apply the same level playing field principle and guarantee that SMAs and insurance companies compete on an equal footing, several laws were changed. SMAs notably needed to secure an agreement to develop their activities, just like insurance companies, by complying with the solvency requirements, guarantee funds and technical provisions stipulated in the non-life Insurance directives, and subsequently, Solvency II. Still, some specificities remain. SMAs and insurance companies do not have the same prudential supervisory authority. While the Bank of Belgium

6  PRIVATE HEALTH INSURANCE IN BELGIUM: MARKETIZATION CROWDED… 

183

oversees the latter, the Control Office of Mutualities (Office de contrôle des mutualités/Controledienst Voor de Ziekenfonden) oversees the former. In addition, mutual benefit societies can distribute the products offered by their SMAs through their own networks, even if they are now formally disjoined entities. Thus, the 2010 reform did not crowd out market logic and market dependency but rather restrained it. Regulated competition now governs an entire segment of Belgian healthcare in a marketization that resulted from long-running retrenchment in several areas and, in the short run, the 2010 reform. The rise of supplementary health insurance in dental and hospital care attests to this, as it created a new pillar whose emergence has mostly benefited for-profit insurance companies. Indeed, 9.5 million people subscribed to a supplementary health insurance contract in 2014—4 million with an SMA, and 5.5 million with an insurance company (Assuralia 2018). This situation is the product of long-term gradual changes (Pierson 2000), notably a gradual retrenchment in public coverage as well as the steadily strengthening position of mutual benefit societies within the system. As Europeanization collided and ultimately merged with these simultaneous change processes (Streeck and Thelen 2005), institutional layering eventually occurred, through the addition of new institutional arrangements with divergent orientations (Thelen 2003; Ebbinghaus 2005)— namely, a third pillar in health insurance based on competition between for-profit and non-profit providers. These transformations have several implications for the institutional equilibriums that have governed the Belgian healthcare system since its foundation. Indeed, only mutual benefit societies are legally obliged to offer complementary health insurance coverage. This is not the case for the public sickness fund (CAAMI), which performs the same functions as mutual benefit societies under the statutory regime but cannot offer complementary health coverage. Moreover, Belgian citizens still have the option of leaving their mutual benefit society and joining the public sickness fund. In this context, people who do not need complementary coverage might migrate from mutual benefit societies to the CAAMI.  Slightly but noticeably perceptible (Table 6.2), this move could, depending on its magnitude, affect the overall funding structure of the complementary health services provided by mutual benefit societies (their operations being funded “according to available means”).

4.538.095 4.597.650 +1%

491.763 536.652 +9%

3.102.255 3.187.006 +3%

575.842 549.664 −5%

Source: INAMI, Service du contrôle administratif, Authors’ calculations

2013 2017 Change

UNML

UML 2.067.497 2.149.961 +4%

87.275 101.511 +16%

CAAMI

UNMS

ANMC

UNMN

Public

Mutual benefit societies

Sickness Fund

Table 6.2  Number of affiliates per Sickness Funds in Belgium (2013–2017)

108.980 104.852 −4%

SNCB Holding

SNCB

10.969.707 11.227.026 +2%

Total

184  C. BENOÎT AND M. DEL SOL

6  PRIVATE HEALTH INSURANCE IN BELGIUM: MARKETIZATION CROWDED… 

185

Another dimension relates to the general balance between statutory and complementary health benefits and services. The 2010 reform did not define a basket of care that should be covered under the statutory scheme and that would automatically define the range of products offered under complementary coverage. Depending on the respective social contributions and fees, this could affect the overall structure of the statutory-­ complementary balance (and, accordingly, of the public-private mix in Belgium). The same applies to the evolution of competition between mutual benefit societies, which was quite limited until recently. By giving them broad discretion to organize the benefits and services provided under the complementary coverage, the 2010 reform also granted mutual benefit societies greater leeway to differentiate in order to attract new members. This did not result from the reform per se, since this development grew with the reduction in statutory coverage, mostly since the 1990s. However, and because it legally recognized mutual benefit societies’ discretion in this area, the 2010 reform staked an explicit position on this development by legitimating it.

4   Conclusion The evolution of the Belgian healthcare system from the post-war period to the early 1990s is marked by several junctures where mutual benefit societies appear as the pivotal actors in the public-private healthcare mix. Despite (but also as a result of) the system’s growing integration at the federal level since the 1963 reform, these entities gradually imposed themselves as both the main trustees of the system and the major players in healthcare reform, as exemplified by the failure of several attempts in the early 1990s to introduce more systemic changes. On several occasions, mutual benefit societies proved to be key actors in the policy process by shaping policies’ overall orientation, especially since the 1980s, when budgetary issues became more salient and an important governmental priority. As in other Bismarckian healthcare systems, cost-containment policies and (though relatively moderate) retrenchment measures were adopted by circumventing these dominant players and veto points. This partly explains why most health spending cuts in Belgium were not obtained through the development of truly regulated competition, but rather focused on more direct healthcare cost reductions in specific areas, notably the hospital sector. These two dynamics—the strengthening of mutual benefit societies in the long run and the more recent yet accelerating trend towards

186 

C. BENOÎT AND M. DEL SOL

cost-­containment policies—largely account for the development of mutualist complementary health coverage from the 1980s onwards. The public-­ private mix in Belgian healthcare gradually developed on the basis of statutory and complementary schemes organized by identical actors (mutual benefit societies), which were more regulated and legally bound in the former schemes than in the latter ones. Unsurprisingly, the processes that led to the formation of large complementary health schemes strengthened mutual benefit societies’ position in the general system. They also helped expand the principles claimed by these entities under the statutory regime to a wider range of (complementary) benefits and services. Mutual benefit societies’ successful, gradual adaptations to changing environmental conditions essentially resulted in path stability (Ebbinghaus 2005). Shaped by European law and regulations, the 2010 reform, by contrast, indicates a path departure. At first glance, it did not spur a radical transformation. It even increased mutual benefit societies’ leeway to organize complementary health coverage. However, and in accordance with the 1973 and 1992 Insurance directives, it also created new institutional arrangements with divergent orientations for supplementary health insurance in dental and hospital care. Marketization was thus explicitly introduced in this area. The discretion afforded to mutual benefit societies to organize complementary health coverage also increased competition, since they were incentivized to differentiate and more intensively compete in this segment to attract new members. Prior policy choices thus narrowed the 2010 reform, but it is unclear which trend will determine the next adaptive step. For this book’s purposes, the overall process reveals that while adaptation to European law and regulations was a major factor of change in Belgian healthcare, its influence was all but automatic. Institutional change to path departure from path stability would have been very unlikely absent the political efforts of insurance companies to challenge the position of their mutualist competitors. This process thus reflects usage of Europe in its stricter sense. The Insurance directives, from which mutual benefit societies were explicitly excluded, were brought back to the national level after Assuralia filed a complaint, and following rulings from the European Commission and European Union Court of Justice. Mutual benefit societies also initially tried to use Europe to limit the impact of the Insurance directives and then, throughout the policy process, to increase their potential market share. This illustrates the leeway that countries possess when it

6  PRIVATE HEALTH INSURANCE IN BELGIUM: MARKETIZATION CROWDED… 

187

comes to implementing EU legislation, particularly when the policies in question are likely to affect the overall organization of the Welfare state. This also reveals the highly contingent influence of the European level on the public-private mix in healthcare. While the transformations described in this chapter ultimately resulted in path departure rather than stability in the Belgian case, they must be understood as the combined product of various simultaneous processes, each with its own timing and unexpected effects.

References ASSURALIA. (2018). Les dépenses nationales en soins de santé (13th ed). Bulletin Hebdomadaire Assurinfo, no. 12. Benoît, C., & Coron, G. (2019a). Expliquer l’échec de la mobilisation d’intérêts économiques à l’échelon européen: l’identité mutualiste aux prises avec la régulation du secteur de l’assurance. Politique européenne, 63(1), 8–35. Benoît, C., & Coron, G. (2019b). Private Health Insurance in France: Between Europeanization and Collectivization. The Milbank Quarterly, 97(4), 1108–1150. Bonoli, G. (2001). Political Institutions, Veto Points, and the Process of Welfare State Adaption. In P.  Pierson (Ed.), The New Politics of the Welfare State (pp. 238–264). New York: Oxford University Press. Colle, P. (2014). Quelques réflexions sur la loi du 26 avril 2010. Revue Belge de Sécurité Sociale, 3th trimestre, 321–327. CURIA. (2010). Judgment of the Court (8th Chamber) of 28 October 2010— European Commission vs. Kingdom of Belgium. Luxembourg, C-41/10. Ebbinghaus, B. (2005). Can Path Dependence Explain Institutional Change? Two Approaches Applied to Welfare State Reform. MPIfG Discussion Paper, 05/02. Ebbinghaus, B. (2006). From Path Dependence to Path Departure in Welfare Reform Analysis. Newsletter of the European Politics & Society Section of the APSA, 5(2), 1–4. Ebbinghaus, B. (2015). The Privatization and Marketization of Pensions in Europe: A Double Transformation Facing the Crisis. European Policy Analysis, 1(1), 56–73. Ebbinghaus, B., & Gronwald, M. (2011). The Changing Public-Private Pension Mix in Europe: From Path Dependence to Path Departure. In B. Ebbinghaus (Ed.), The Varieties of Pension Governance: Pension Privatization in Europe. Oxford: Oxford University Press. Engels, J. (1970). Evolution of the sickness and disability insurance 1945–1970. National Institute for Sickness and Disability Insurance: Brussels.

188 

C. BENOÎT AND M. DEL SOL

Esping-Andersen, G. (1996). Welfare States Without Work: The Impasse of Labour Shedding and Familialism in Continental European Social Policy. In G.  Esping-Andersen (Ed.), Welfare States in Transition (pp.  66–87). London: Sage. European Commission. (2006). Commission Scrutinises Belgian Law on Supplementary Health Insurance Provided by Private Sickness Funds. Brussels, IP/06/1781. European Commission (2008). Reasoned Opinion C2008/1458. Geerkens, E. (2002). Les mutualités professionnelles, un axe majeur de la politique sociale patronale pendant l’entre-deux guerres en Belgique? Revue Belge de Philologie et d’Histoire, 80(4), 1275–1349. Graziano, P., Jacquot, S., & Palier, B. (Eds.). (2011). The EU and the Domestic Politics of Welfare Reform. Basingstoke: Palgrave. Hassenteufel, P., & Palier, B. (2007). Towards Neo-Bismarckian Health Care States? Comparing Health Insurance Reforms in Bismarckian Welfare Systems. Social Policy & Administration, 41(6), 574–596. Itçaina, X., Roger, A., & Smith, A. (2013). Institutionnalizations Compared: Implementing the EU’s 2008 Wine Reform. Comparative European Politics, 11(1), 119–142. Lewalle, H. (2006). Regard sur l’assurance maladie privée dans l’Union européenne. Revue française des affaires sociales, 2-3, 141–166. Ministère des Affaires sociales et de la Santé publique. (2009). Notification au Conseil des Ministres. Brussels, CMR Note du 13/07/09 (2009A70750.014). Ministère des Affaires sociales et de la Santé publique. (2010). Réponse à la procédure de l’article 226-Mise en demeure-dossier 2006-4293. Brussels, Cellule stratégique du ministère, LO/LB/FP/JMC/ch. Mossialos, E., & Thomson, S. (2004). Voluntary Health Insurance in the European Union. Brussels: World Health Organization and European Observatory on Health Systems and Policies. Naczyk, M. (2013). Agents of Privatization? Business Groups and the Rise of Pension Funds in Continental Europe. Socio-Economic Review, 11(3), 441–469. Perelman, J., & Closon, M.-C. (2004). Hospital Response to Prospective Financing of Inpatient Days: The Case of Belgium. Mimeo: UC Louvain. Pierson, P. (2000). Not Just What, but When: Timing and Sequence in Political Processes. Studies in American Political Development, 14(1), 72–92. Pierson, P. (2003). Big, Slow-moving and… Invisible: Macrosocial Processes in the Study of Comparative Politics. In J. Mahoney & D. Rueschemeyer (Eds.), Comparative Historical Analysis in the Social Sciences (pp. 177–208). New York: Cambridge University Press. Rezsohazy, R. (1957). Histoire du mouvement mutualiste chrétien en Belgique. Brussels: Editions Erasme.

6  PRIVATE HEALTH INSURANCE IN BELGIUM: MARKETIZATION CROWDED… 

189

Schokkaert, E., Guillaume, J., & Van de Voorde, C. (2018). Risk Adjustment in Belgium: Why and How to Introduce Socioeconomic Variables in Health Plan Payment. In T. McGuire & R. Van Kleef (Eds.), Risk adjustment, Risk Sharing and Premium Regulation in Health Insurance Markets. London: Academic Press. Schokkaert, E., & Van de Voorde, C. (2000). Risk Adjustment and the Fear of Markets: The Case of Belgium. Healthcare Management Science, 3, 121–130. Schokkaert, E., & Van de Voorde, C. (2005). Healthcare Reform in Belgium. Health Economics, 14, S25–S39. Schokkaert, E., Van Ourti, T., De Graeve, D., & al. (2010). Supplemental Health Insurance and Equality of Access in Belgium. Health Economics, 19, 377–395. Smith, A. (2016). The Politics of Economic Activity. Oxford: Oxford University Press. Stevens, Y., Van Rompaey, L., Huber, V., & Van Buggenhout, B. (1998). Issues in Complementary Health Insurance in Belgium. International Social Security Review, 51(4), 71–91. Streeck, W., & Thelen, K. (2005). Introduction: Institutional Change in Advanced Political Economies. In W.  Streeck & K.  Thelen (Eds.), Beyond Continuity. Institutional Change in Advanced Political Economies (pp.  1–39). Oxford University Press: Oxford. Thelen, K. (2003). How Institutions Evolve: Insights form ComparativeHistorical Analysis. In J. Mahoney & D. Rueschemeyer (Eds.), ComparativeHistorical Analysis: Innovations in Theory and Method (pp.  208–240). New York: Cambridge University Press. Thomson, S., Busse, R., Crivelli, L., van de Ven, W., & Van de Voorde, C. (2013). Statutory Health Insurance Competition in Europe: A Four-country Comparison. Health Policy, 109, 209–225. Woll, C., & Jacquot, S. (2010). Using Europe: Strategic Action in Multi-Level Politics. Comparative European Politics, 8(1), 110–126.

CHAPTER 7

Europeanized, Marketized but Still Governed by the State? Private Health Insurance in France Gaël Coron, Thomas Houssoy, and Cyril Benoît

1   Introduction The French healthcare system, including the delivery of health services, is financed by a mixture of public and private mechanisms. Coverage is provided to all legal residents through a National Health Insurance (hereafter NHI1) scheme. It is part-funded on an equally shared basis by employer  The Assurance maladie (literally, “Sickness Insurance”), more commonly known as Sécurité sociale (literally, “Social Security”). 1

G. Coron (*) Arènes (UMR CNRS 6051), EHESP, Rennes, France e-mail: [email protected] T. Houssoy Clersé (UMR CNRS 8019), Université de Lille, Lille, France e-mail: [email protected] C. Benoît Centre for European Studies and Comparative Politics, Sciences Po (Paris), CNRS, Paris, France e-mail: [email protected] © The Author(s) 2021 C. Benoît et al. (eds.), Private Health Insurance and the European Union, https://doi.org/10.1007/978-3-030-54355-6_7

191

192 

G. CORON ET AL.

and employee Social Security contributions which are deducted at source from salaries, as well as by other fiscal resources that have played a growing role in healthcare financing over the last 20 years. With an advisory board2 equally made up of employer representatives, employee representatives and a group of diverse stake-holders (such as government-appointed experts, representatives of other Social Security administrations, representatives of the mutual benefit societies and student trade unions), NHI funds (caisses) are quasi-public entities. Their activities are overseen by the Health Minister, in part thanks to the General Directorate appointed by the government whose power has increased since the 1990s at the expense of the advisory board. The range of health services covered is quite wide-­ ranging and the legal rate of reimbursement is around 70% for most routine care (e.g. general practitioners) and around 80% for hospital fees, with some exceptions for long-duration diseases that are almost fully funded by the NHI. In this context—and since the foundation of the NHI system—a share of health coverage has been left to private health insurance (hereafter PHI), subscribed by individuals or corporations (see Rodwin 2003). Private health insurance in France is thus qualified as complementary since, in most situations, both NHI and PHI share the reimbursement of healthcare costs. Recent developments have also contributed to certain changes: PHI now has an exceptionally high level of coverage and plays a significant role as a co-payer for treatments and services in the statutory healthcare system. It is also important to note that from the 1980s onwards, the share of health expenditure financed by NHI has ceased to grow. This movement coincides with more explicit institutional recognition of the PHI, a field that the State has tried to organize by assigning to it a range of public health missions. At the French government’s insistence, gradual institutionalization was consolidated by the creation of a single representative body grouping the various private health insurers operating in the country. Indeed, three different categories of players now cover the portion of health costs that is not reimbursed by the NHI. Traditionally, and crucially, the sector has been dominated by nonprofit actors (mutual benefit societies and provident institutions), but here too, recent developments have changed their place and function. As discussed throughout the chapter, most of these changes result from the co-evolution of European law and regulation and critical policy choices at the domestic level. 2

 The governing board became anadvisory board in 2004.

7  EUROPEANIZED, MARKETIZED BUT STILL GOVERNED BY THE STATE?… 

193

As such, and similarly to other country cases covered in this book, European regulations have had an important role in the changes that have occurred in the PHI sector in France. On the one hand, the so-called third generation insurance directives in the 1990s contributed to increased competition and to changes in how non-profit entities operate (see Chap. 2). On the other hand, the Solvency II directive led to a significant improvement in prudential standards and a transformation in insurers’ internal practices, with, again, greater implications for non-profit providers than for for-profit companies (see Chap. 3). However, the main thesis defended here is that the Europeanization of PHI in France was certainly not merely a mechanical adaptation of the sector to a clear and well-specified European set of constraints. These European directives have in fact come up against long-term transformative changes, affecting PHI, NHI and the overall relationship between them. Moreover, the different players’ respective strategies were instrumental in shaping this overall process. In sum, the European framework has led to a variety of “usages” (Graziano et  al. 2011) more than it forced private health insurers to respond. In reflecting upon the whole process, we are therefore particularly interested in the various interpretations of European provisions to wider institutional transformations that have affected the public-private mix in French healthcare. On this broad level, we are able to identify a twofold trend. On the one hand, what we can clearly identify is a gradual decline of the presence of historical actors in the PHI sector, namely mutual benefit societies, a decline that contrasts heavily with the mounting presence of for-profit entities. This partly results from the various forms of marketization introduced by Insurance, then Solvency II directives, but due consideration must be given to critical transformations of the public-private mix during the same time period. In effect, the other trend reported here is the seemingly paradoxical “statization” (Hassenteufel and Palier 2007) of PHI. The recent years have seen a number of initiatives carried out by successive French governments consisting in using regulation of the PHI sector to achieve a growing number of policy objectives. Even if these latter initiatives in France can be differentiated from the European directives, both trends show cross effects. As a result, a European-driven market now interacts closely with a state-­driven market (Gingrich 2011), which, in turn, has led to a number of unintended effects for the provision of healthcare and for individual welfare benefits.

194 

G. CORON ET AL.

To draw these inferences, we relied on a wide range of first- and second-­ hand sources, ranging from public statistics to interviews, while summarizing the recent historiographic and econometric contributions from both French and foreign experts on the state and dynamics of governance, policy-­making, finance or inequalities in the French healthcare system. This article is organized into three parts. In Sect. 2, we offer a description of the current state of the public-private mix in French healthcare, as well as a brief overview of the different players operating in the field of PHI. Section 3 evaluates the impact of EU law and regulation on the sector. We argue in Sect. 4 that these developments conflict with a gradual “statization” of the PHI market, a term used to describe the growing presence of (domestic) public policy and regulation in the sector. In particular, we demonstrate that politicians and decision-makers were not always aware of the effects EU law and regulation had on PHI actors when conducting their reforms. Overall, the opposing increases in both European and national requirements, constraints and policies appear to be a major source of unintended effects generated by the public-private mix in France healthcare system.

2   The Public-Private Mix and Private Health Insurance in French Healthcare Private health insurance in France is based on a so-called complementary regime, as public and private funders roughly co-finance the same healthcare for the same populations (see Sagan and Thomson 2016). Here we present how NHI and PHI are economically linked, giving the French public-­private mix its current shape. Then, we show that PHI is made up of a plurality of actors, each with its own legal forms, strategies and specificities. If this second dimension does not necessarily affect the distribution of expenditures, it strongly influences the mode of coverage, as well as the overall structure of the PHI market. 2.1  Private Health Insurance as a Minority Though Crucial Stakeholder in French Healthcare Both the level of PHI funding and the level of coverage of the population is high in France compared with most Northern and Western European health systems, including neighbors with comparable types of health

7  EUROPEANIZED, MARKETIZED BUT STILL GOVERNED BY THE STATE?… 

195

systems such as the Netherlands and Germany. This might be explained by the role private insurers play in financing health expenditures. Their role has clearly served to extend coverage beyond the legal rate of NHI reimbursement, set at a 70% level for most benefits. Since the 2004 DousteBlazy reform (named after the Health minister of the day), a basket of care reimbursed by PHI has been re-defined in order to both clarify its purpose and limit its coverage, thus fulfilling the reform’s aim to focus cost-­ containment policies on the demand-side. The de facto effect has been the increased integration of PHI within the implementation of health policies of the NHI (see further; Sorum 2005). However, taking a closer look at cost-­sharing by types of care, one observes that the overall mixed insurance organization can in some cases result in private insurance financing an equally important share or even prevailing over NHI interventions, notably in dental and optical care (40% and 71.5% in 2015, DREES 2018, p. 97). Such extreme divergences between the legal and actual cost-sharing between NHI, PHI and households are the result of enduring deficiencies in price regulation. Since the end of the nineteenth century, office-based practitioners have indeed represented a vocal and effective political force in defending freedom of price (Hassenteufel 2001). This particularity of the French political economy of health financing is deeply embedded in healthcare expenditure under the form of a “distribution rule of cost” between the three types of final payers (Fig. 7.1). Its foundational influence is that price-setting has operated mainly on a feefor-service basis for office-based practitioners (Element A) since the creation of NHI.  The main mechanism of price regulation for public

Fig. 7.1  The “distribution rule of cost” in the French health insurance system, its functioning and aggregate final distribution between financers. (Source: authors’ own calculations). Note: dimensions are not scaled for reasons of readability. Source for the aggregate final distribution is DREES (2018). Each letter relates to elements in the explanations below

196 

G. CORON ET AL.

authorities has been the national agreements on mandatory rates signed between the NHI and national physicians’ trade unions (Element B) ever since 1971. It is on this basis that the legal rate of NHI coverage determines the Social Security’s reimbursement (Element D), leaving co-payments fees to be covered by private insurers (Element E). Deductibles left aside (Element F) the main variable underlying the high variation in the actual rate of NHI coverage are the share of extra-fees (Element C) in the actual cost charged by health professionals. Consequently, the origins of the unequal NHI coverage between types of care can be found in two opposing concessions made to office-based practitioners in the price regulation mechanism. Firstly, freedom of prices was re-established de facto by the creation of a second-­tier sector during the 1980s agreements, removing most of the limitations put on extra-fees beyond the statutory rates. Secondly, there are critical exceptions to the prohibition of freely set extrafees beyond mandatory rates, the latter not being set at substantial levels and lacking adequate readjustments over time, as in the cases of dental and optical care and hearing aids. Beyond these micro-level situations, the mechanisms presented above are also the reflection of a macro-level reassignment of the role of the NHI and PHI which has drawn the French health insurance system further away from its complementary rationales. The latest dynamics of the aggregate cost-sharing can be traced back to the mid-1980s: after four decades of increasing NHI’s coverage, the socialized share of financing fell from a peak of 80% in 1980 to a level of 76–77% by 1990. Since then, NHI aggregate coverage has remained stable while the overall gain in shares of financing for PHI came from reducing the out-of-pocket amounts that households pay. However, this stability is a mistaken depiction of an unchanging landscape where (all) health expenditure seem to remain highly covered by the national insurance scheme. Described as a “silent metamorphosis” (Tabuteau 2010), the growing importance of PHI in France stems from this context of a growing segmentation of NHI coverage between (decreasing) ambulatory or primary care and (consistently high) chronic and hospital care. Its roots are found in the predominance of cost-containment policies starting in the early 1990s, that changed the nature of political trade-offs to offset the growth of health expenditure relative to the gross domestic product (GDP), shifting it from increasing NHI financing to adjusting its coverage parameters between types of care and sub-groups of beneficiaries (see Chap. 10). The most notable instrument of this “silent metamorphosis” is the Affections

7  EUROPEANIZED, MARKETIZED BUT STILL GOVERNED BY THE STATE?… 

197

Longue Durée program (ALD stands for “long term diseases”). Under ALD, beneficiaries are fully covered by NHI for a limited number of diseases and strictly for the care linked to this disease. They are exempted from co-payments. Quantitative evidence suggests that this programme covers 17% of NHI’s beneficiaries but represented 62.1% of its reimbursements in 2015 (Ministère des Affaires sociales et de la Santé 2017, p. 62). While more than 90% of hospital care costs are covered by NHI regardless of status, a stark difference exists regarding out-patient or primary care. In sum, the combination of a deficient price regulation with increasing segmentation among NHI beneficiaries results in a high dependency of French households on PHI to access the health system and bears a large responsibility in making France a country where very high health inequalities persist. 2.2  Variety of Private Health Insurers The private health insurance sector in France is divided among several actors, each with a different legal form and prior history in relation to NHI.  Traditionally, mutual benefit societies have played a decisive role and gradually emerged as privileged partners of public authorities (see below). These entities have specific characteristics such as the absence of shared capital, which leads to the election of directors by all members, and a different use of their profits than in insurance companies, as they are saved or reinjected to increase the benefits provided to their members. Governed by a specific legal code, the Code de la Mutualité, they are grouped within a national federation (Fédération Nationale de la Mutualité Française–FNMF). Though declining, one of their activities is also to manage compulsory health insurance for certain populations (such as civil servants) on behalf of the NHI. Health insurance is the core business of mutual benefit societie, and 85% of their sales revenue and 70% of contracts are generated through individually based schemes (DREES 2019). Other (minor in terms of revenues) products they offer include occupational welfare and funeral plans. A second category of private health insurers is formed by provident institutions (PIs, for Institutions de prévoyance). Like mutual benefit societies, provident institutions are also non-profit firms, but their governance differs as they are managed by trade unions and employer representatives. Represented by the Centre technique des institutions de prévoyance (CTIP), they have historically focused on the area of pension provision and, above all, in the management of collective

198 

G. CORON ET AL.

(complementary) pension regimes. In particular, they are in charge of managing the mandatory complementary pay-as-­ you-go retirement scheme. Through this activity and their governance, they have a privileged link with companies and are therefore well-positioned in corporate-level social protection. 88% of their health insurance’s contracts are collective and health insurance generates half of their sales revenue (DREES 2019). For their part, for-profit insurance companies are organized mainly as joint stock companies grouped together in the French insurance federation (Fédération Française de l’Assurance—FFA) and governed by the Insurance Code. In 2017, 103 insurance companies were engaged in health insurance. Nevertheless, this activity represents only 5% of their sale revenue, shared equally between individual and collective contracts (DREES 2019). Currently, the market shares in PHI are as follows: 51% by mutual benefit societies, 31% by for-profit insurers and 18% by provident institutions (ibid). This overall distribution was significantly different some 20 years ago, with an increase of the market share of for-profit insurance companies by 47% during this time period. Until recently, mutual benefit societies enjoyed a de facto monopoly, that was first affected by policy choices at the national level, which should also be noted, were adopted in anticipation of the future European framework. Active since the nineteenth century, mutual benefit societies are the product of the initial attempts made in 1930 to forge a compulsory health insurance system. However, this endeavor was critically transformed in 1945 with the creation of Social Security (NHI), whose management was mainly entrusted to trade unions. Alexandre Parodi, Minister of Labour, proclaimed explicitly that Social Security allow for “the succession of mutual benefit societies through trade unionism” (quoted in Siney-Lange 2018). A 1947 law (“Loi Morice”) tried to secure the position of mutual benefit societies in the new system by granting them a small share of the management of NHI for civil servants and large companies’ employees nationalized in 1945. “Co-payment” was also granted to them for benefits and services that were not fully covered by the new compulsory scheme. Initially intended to be temporary, the public-private mix that emerged following this critical period proved enduring. This initial choice has had long-term effects on the positioning of mutual benefit societies in France and on their relationship with NHIs. Soon after 1947, their representatives start presenting their organizations as the voice of healthcare users. They also become key veto players in

7  EUROPEANIZED, MARKETIZED BUT STILL GOVERNED BY THE STATE?… 

199

French healthcare, regularly joining or forming coalition against a number of reforms. This was demonstrated in 1978 when they built up an alliance with trade unions to defend the principle according to which health insurance should belong to the non-profit sector, and further still in 1979 when they led major protests against the introduction of a mandatory co-payment mechanism, thus preventing the mutual benefit societies from reimbursing individuals directly. Partly as a result of the constant political efforts, but also thanks to their long-standing links with the Socialist Party, mutual benefit societies representatives gained a seat on the Social Security Governing Board granted by the Socialist government in 1982. By then there was thus no market for PHI per se, in the sense that complementary health insurance contracts, though individual, were often acquired on a professional basis, each profession or sector being closer to a mutual benefit society or another. There was virtually no real competition between these non-profit actors, and a number of behaviours or instruments associated with a market “logic” (such as risk-selection) were considered illegitimate by mutual benefit societies or simply forbidden by law.

3   The Formation of a PHI “Market” (Partly) through Europeanization As in other countries, it is essentially through EU law and regulation that market instruments and principles were introduced in PHI from the late 1990s onwards. While 1992 Insurance directives changed a number of organizational features of mutual benefit societies and undermined their monopoly, Solvency II contributed to an alignment of their working practices with the rules and principles governing for-profit insurance activities. These two directives thus created a market for PHI on several levels, which entailed imposing formal provisions, introducing new values and practices. However, we maintain that it was far from being an automatic imposition for three main reasons. First, on account of the usage of Europe “by anticipation” by national authorities was arguably as important as the implementation of Insurance directives; second because Solvency II partly fits with the agenda of a number of (notably nonprofit) health insurers and thirdly, as argued in the following section, during the same period successive French governments tried to organize the PHI market nationwide.

200 

G. CORON ET AL.

3.1  Europeanization by Anticipation: How Non-profit Firms Lost Their De Facto Monopoly The provisions laid down in 1992 Insurance directives (see Chap. 2) have proven particularly critical for non-profit actors and especially for mutual benefit societies. Initially designed for “traditional” insurance companies, with no specific reference to health insurance, these texts require that the director or the controller of the company be “fit and proper”, a criterion which could potentially jeopardize the position of directly elected presidents of mutual benefit societies, often without particular actuary training or skills. This challenge to their democratic governance is more fundamentally reinforced by the overall idea that all entities operating in the single market now have to compete on an equal footage, an explicit threat to these actors that had enjoyed a rather privileged situation in the context of their established institutional relations with public authorities and, more pointedly, because certain rules were explicitly designed to favour them. And indeed, this is what happened. After a conflictual implementation phase where mutual benefit societies essentially mobilized to delay the process, a number of barriers protecting mutual benefit societies were eventually removed. Competition mechanisms were introduced within the whole PHI segment and a number of selective advantages, such as preventing competition in the complementary coverage of civil servants, (Benoît and Coron 2019), were removed. As a result, competition increased slightly, with the limited though significant development of forprofit insurance companies in the sector. For large insurance firms operating in France, healthcare did not necessarily become a major market after the 1992 Insurance directives, as it was far from being the most profitable segment of their activities. Yet, most insurance companies started to use health coverage to attract consumers to other products on offer (notably life insurance), notably for corporate-level contracts. The governance of mutual benefit societies was, however, preserved from this change. In the early 2000s, and after a mobilization of their representatives, the Mutuality Code was reformed to protect it, namely by separating their purely insurance activities from their charities or other benefits in kind that they provide (see below), and by adding some training requirements for their elected directors or controllers. Beyond these arguably expected effects, the question remains: why French authorities, who have cultivated strong links with mutual benefit societies (and vice versa), have more or less tacitly embraced Insurance

7  EUROPEANIZED, MARKETIZED BUT STILL GOVERNED BY THE STATE?… 

201

directives and implemented them? Part of the response lies in the way domestic actors have anticipated the shape of Insurance directives during the 1980s, in the light on what was happening in the same time within the field of PHI. Indeed, and over the course of the decade, a number of think tanks and representatives of insurance companies mobilized or simply published reports to advocate the possibility of “privatising” or at least, attenuating the “monopoly” of Social Security in social protection (this includes NHI but also pensions). In healthcare, these attempts never quite reached the political agenda and attracted relatively low public attention. Yet, they contributed to (re)framing a number of debates related to the country’s public-private mix (Benamouzig 2012), sometimes in connection with the debate on European integration. PHI was not immune to these developments, and a discussion gradually formed between a few numbers of politicians and senior civil servants around the possibility of either granting mutual benefit societies a legal monopoly in PHIs or developing a market by reinforcing competition, notably with insurance companies and provident institutions. During the early 1980s, the newly elected Socialist government, and particularly the Ministry of Social Affairs Pierre Bérégovoy, advocated strongly in favour of “prohibiting insurance companies from competing with mutual benefit societies” (Domin 2019) to reinforce the de facto monopoly on PHI. In fact, the first overhaul of the Mutuality Code in 1984 is more likely to strengthen the capacity of mutual benefit societies to act in other sectors (provident and savings) and thus pave the way for the end of health as the preserve of mutual benefit societies. However, this period is also marked by a number of debates within the governing Socialist Party around the compatibility of a number of its policies with European integration (Cole 2001). In 1985, an inter-ministerial group, led by a senior civil servant, Pierre Gisserot, was charged with identifying potential contradictions between domestic provisions in social protection and the regulations of the future single European market. The influential report that followed proposed a general prognosis: in the perspective of future integration, some prior choices related to PHI should be removed. In other terms, and several years before the Insurance directives came into force, the Gisserot report called for greater competition between health insurers, notably through the implementation of a more competitive framework and stricter prudential requirements. In 1989, a law was eventually adopted (“Loi Evin”) that set out the formal basis for a subsequent marketization of the sector through harmonizing the regulations

202 

G. CORON ET AL.

that applied to private health insurers. In stark contrast with the decisions that had been adopted in previous years, mutual benefit societies, insurance companies and provident institutions alike were also subjected to the same prudential framework. So the Gisserot report is France’s preparation for European openness to the single market. And the Evin law, it translates that in a way. So, there is indeed a relationship between the Evin law and European considerations. Interview, Paris, March 20, 2018, former managing director of a complementary pay-as-you scheme

As such, the Gisserot report was not driven by any explicit constraint “from Brussels”. It is best understood in the broader context of the debates that occurred during the Mitterrand presidency over economic policy (Machin and Wright 1985; Hall 1985). While some members of the Socialist Party wanted to take the opportunity of the party’s presence in government to pursue “truly” Socialist policies (potentially at the expense of European integration), a greater section was in favour of a more moderate, social democratic approach. These actors, who included a number of senior civil servants close to the party (such as Pierre Gisserot) often used the argument of European integration to legitimate their views by repeatedly arguing that European integration should trump all other political imperatives. Insurance directives thus had consequential effects in the French case not only as a result of their implementation, but also as they were referred to directly a priori in the context of domestic and partisan politicking. When the transposition eventually took place, the PHI market had therefore already been opened to competition, at least to a certain degree. Granted, several issues remained to be dealt with in order to comply with European regulations. Firstly, the taxation of insurance contracts ostensibly favoured non-profit players. In order to restore equality between them, since 2004, only differences in the content of contracts (such as the coverage or the range of benefits provided) led to differential taxation. Secondly, mutual benefit societies also possess health and social services or facilities and provide their members with benefits in kind. However, according to the principle of specialization included in the European framework, insurance operations have to be protected and given separate consideration in order to guarantee a fair allocation of prudential reserves (see Chap. 2). In 2001, the second reform of the Mutuality Code legally separated health

7  EUROPEANIZED, MARKETIZED BUT STILL GOVERNED BY THE STATE?… 

203

and social services from mutual benefit societies dedicated to health insurance, while maintaining the possibility for them to create holdings so that they could retain these various activities in the same structure (Benoît and Coron 2019). 3.2  The Solvency II Directive and the Organizational Side of Marketization As discussed in greater length in Chap. 3, Solvency II operates on a different basis to the prudential or regulatory frameworks which preceded it, in the sense that it confers an explicit power to a range of standards (Graz 2019), likely to change the daily lives of health insurers. It also conveys on the same basis a range of norms, metrics and newer ideas, while inducing a greater alignment of non-profit with for-profit companies. A careful observation of the consequences of Solvency II reveals that it indeed resulted in greater convergence in the market, as it entailed the formation of large non-profit entities, intending to compete and to challenge the growing presence of insurance companies in the sector. However, a deeper analysis of the motives that drove these transformations reveals that, if some of these behaviors are clearly attributable to Solvency II, it has also constituted a window of opportunity for certain strategies and thus served a posteriori to justify choices driven by other considerations. Before discussing this process in greater length, it is worth reviewing some of the most obvious implications that Solvency II had on the structure of the French insurance market. Under this directive, health insurance is considered a short-term risk. As such, it does not require important amounts of equity capital to meet Minimum and Solvency Capital Requirement (MCR and SCR, respectively). However, mutual benefit societies suffer more than any other category of health insurers under this framework, as it generates three structural inadequacies in their business model. First, the focus of mutual benefit societies on health insurance results in low levels of asset diversification, which by its very nature means that the amount of equity capital must be increased in order to meet solvency requirements. Following classical insurance economics, a higher diversification of assets lowers the default risk of insurance companies and therefore the amount of equity capital required. Secondly and as already discussed in Chap. 3, there are a number of incompatibilities between the mode of calculation of the SCR standard formula and mutual benefit societies’ organizational peculiarities (Abecassis et  al. 2014). In a nutshell,

204 

G. CORON ET AL.

these entities use a share of their savings to increase the benefits of their members. This is based on an overall conception of solidarity whereby future policyholders will pay for future needs. In stark contrast, Solvency II promotes a more capitalistic conception of solidarity, where sufficient savings have to be secured in order to cover current policyholders. This, in turn, inflates the capital burden, especially for small and medium-sized companies as is the case for most mutual benefit societies. A third and related issue concerns investment income which is relatively low for mutual benefit societies due to the structure of their activities. These are unlikely to rise within a broad context of low interest rates, thus putting additional pressure on solvency ratios (Rousseau and Elbarhdadi 2016). To cope with these various challenges, mutual benefit societies have gradually developed various strategies. They have notably created or reinforced actuarial functions within their own organizations. One might observe the formation of large risk management bodies, especially by the larger French mutual benefit societies. This also favoured an increase in staff turnover among actuaries who had first started their career in insurance companies and now work or lead departments within mutual benefit societies. Such transfers were quite rare before this period. Another series of changes relate to risk management per se. As they often developed on a professional basis, mutual benefit societies tend to have policyholders with similar socioeconomic backgrounds and that are consequently exposed to similar risks. Tighter risk management techniques were elaborated as a response, such as grouping policyholders in sub-populations targeted with dedicated preventive healthcare programs. Larger mutual benefit societies are also diversifying, sometimes outside their historical activities. At the sectoral level, it is essentially through mergers and acquisitions (M&As) that the various issues raised by stricter solvency requirements were tackled. Most M&As take the form of sophisticated configurations depending on the size (and therefore strategies) of the different actors. Abecassis et  al. (2014, p.  70) identifies two categories of convergence strategies between non-for-profit firms. On the one hand, soft convergences consist of agreements to share materials, technical skills and support activities. Small mutual benefit societies prefer such cost-saving strategies since they cannot afford the development of an internal model of SCR calculation. On the other hand, hard convergences consist of mergers and acquisitions, delegations of risk management (substitution) and union formations that allows medium and large mutual benefit societies to opt for

7  EUROPEANIZED, MARKETIZED BUT STILL GOVERNED BY THE STATE?… 

205

a strategy of developing a better risk management process and an internal model of SCR calculation. Mergers and acquisitions are the most noticeable form of hard convergence taking place in the French PHI market. Between 2001 and 2017, the number of insurance institutions have decreased four-fold (from 1702 to 474 today). Not-for-profit institutions were mostly affected by this transformation: there are more than four times fewer mutual benefit societies (1528 vs. 365) and half as many provident institutions (57 vs. 25) (DREES 2019, p. 47). A firm can also delegate to another mutual benefit society the actual activity of risk management (substitution) without being merged with it: in 2015, a quarter of the mutual benefit societies (88 over 411) registered have done so (Montaut 2017, p.  8) and are therefore exempt from most of the European insurance regulation. In 2017, of the 398 private health companies3 accounted by the Statistical Department of the French Ministry of Health (DREES), 41% are part of a union or group. Groups can merge insurance institutions of different types, sometimes supervising for-­profit and not-for-profit firms, which threat to blur each family’s: group’s boundaries, practices and distribution within the market. Overall, in 2017, groups held 84% of PHI market shares while independent institutions (in a form of soft convergence or not) only 16% (DREES 2019, p. 48). When we take a closer look at group composition, one can observe a limited hybridization of groups according to the kind of insurance company heading it. A notable exception is groups headed by provident institutions that shelter a greater number of mutual benefit societies. This is consistent with a strategy of delegating health insurance to mutual benefit societies created or absorbed by these firms in order to benefit from their historical position and their skills. Moreover, if some insurance companies sometimes shelter non-for-profit institutions, not-for-profit groups very rarely include for-profit institutions. These results are consistent with case studies of the hybridization process: group formation tends to have been a strategy protecting not-for-profit institutions’ tradition and practices in the face of regulations and market conditions, thereby pushing them to conform to classical for-profit insurance companies (Abecassis et al. 2018) (Table 7.1).

3   This figure includes groups and independent companies and excludes companies within groups.

Overall Mutualist groups Insurers groups Social protection groups 12

30

50 8

50

83

162 29

24

11

62 27

8

71

81 2

18

1

19 0

Overall Mutual institutions Insurance companies Provident institutions

Note: Substituted Mutual Institutions are not shown. Independent institutions include institutions that can be in a soft convergence with one another or a group

6

16

236 214

Number of independent institutions

Independent institutions categories

Groups categories

Number Number of institutions Included in this type of groups … of groups included in the groups Mutual Insurance Provident institutions companies institutions

Independent institutions

Groups

Source: DREES (2019, p. 49) based on data from the French supervisor (ACPR) and the CMU Funds, authors’ translation

Paritarian

Insurers

Overall Mutualists

Overall categories

Table 7.1  Number of independent firms and groups operating on French PHI market (2017)

206  G. CORON ET AL.

7  EUROPEANIZED, MARKETIZED BUT STILL GOVERNED BY THE STATE?… 

207

With a market share of 51%, not-for-profit institutions, mutual benefit societies still enjoy a dominant position. Yet, they have lost considerable market share (11 points over 16 years). Their grip has been loosened by group formations, which when taken into account, indicates a more accurate share of 39% (DREES 2019, pp.  48–49). Such a decrease is partly explained by these same M&As, as a significant number of mutual benefit societies are now merged by provident institutions. But it is also due to the rise in for-­profit insurance companies in the sector, a growing presence that is largely due to a parallel need for diversification induced by Solvency II. The essential paradox with this directive, at least in the French case, is that it has contributed to consolidating the presence of (heavily financed) insurance companies but also, it has engendered greater financialization of non-profit actors, whereas the original aim was, in several respects, to generate more stability and security on the insurance market. However, this overall process is not just about Solvency II and the constraints it entails for health insurers. Here also, the European framework has served to advance a variety of responses and strategies. Our own interview data even suggest that this holds true for the three categories of health insurers operating in France. For insurance companies, Solvency II has reinforced (and not initiated) the strategy consisting in using healthcare to attract policyholders to other products, especially at the corporate level, often in combination with other policies adopted during the same period (see below). Most M&As among provident institutions were decided several years before Solvency II came into force. In this period, labor unions and employee representatives tried to generate economies of scales (in part due to the growing competition they faced from insurance companies in the competitive context created by Insurance directives). For some mutual benefit societies, Solvency II even served as a window of opportunity. During the 1990s, the larger ones (though not always on account of the Insurance directives) increasingly focused on health insurance, thus limiting the charitable acts, care facilities or other benefits in kind which they historically provided to their members. Such specialization was not neutral, and led to fierce political battles within the FNMF, their representative body, as smaller entities argued that such change was rendering them too similar to traditional insurance companies. Solvency II helped larger firms to justify their increasing specialization in health insurance within the sector as it rendered this activity more demanding and intensive than before.

208 

G. CORON ET AL.

Honestly, if we look at the influence of the European dimension, it was not the most decisive. Of course it was very important. But, from within, it was in fact an opportunity to say: “what is important for us?” […] And I discovered in the board [people] dreaming of becoming insurers. It is a sort of… ­dominant-­dominated relationship. […] They wanted to show that they were as capable as the others of being insurers. Interview, Paris, Past FNMF director (Mutual benefit societies representative body), March 2018

In sum, European law and regulation has, one may argue, induced greater marketization of PHI in France, both at the sectoral, organizational and firms’ individual level. Yet, it has been constructed in different manners by the various actors involved. Crucially, some of their deep preferences in this context were also shaped by their response to critical policy choices that have been made at the domestic level, sometimes with no obvious connections to changes induced by Europeanization.

4   A State-Driven Market? Prior research has described the recent institutional transformations faced by the French healthcare system as an overall process of “statization” (Hassenteufel and Palier 2007). As in other Bismarckian countries, a number of prerogatives devolved to labour unions through sickness funds were gradually centralized and placed in the hands of ministerial, legislative or independent regulatory authorities with the aim of forging a “regulatory healthcare state”, often in combination with various cost-containment policies. This trend has also affected PHI.  From the beginning of the 2000s onwards, successive French governments have sought to increase State influence over this market. As such, and during the same time period, one might see both an increase in the role of European provisions and regulation on PHI accompanied by attempts to govern this market and to serve a variety of domestic purposes that essentially touch on universal health coverage and the provision of certain health benefits. As a result, a European-­driven market now increasingly interacts with attempts at building a State-­driven market (Gingrich 2011). Above all, the rationales promoted at the domestic level often conflict with the prior effects of Europeanization. This, in turn generated a number of side effects on the current form of the public-­private mix in French healthcare.

7  EUROPEANIZED, MARKETIZED BUT STILL GOVERNED BY THE STATE?… 

209

Prior to that, some critical decisions during the 1990s equipped the State for greater involvement in the management of NHIs, notably through a significant increase in the role of parliament in determining the yearly budget for Social Security or a gradual rise of fiscal resources in funding the system. A number of senior civil servants (labelled as the “Welfare elite” in the literature) were instrumental in pushing for this agenda by being able to influence governments of different partisan affiliations (Hassenteufel et al. 2010). In the early 2000s, a reform pushed this process a step further towards “technocratic recentralization” (Simonet 2019) by directly modifying institutionalized relationships between PHI and NHI. In 2003, a report authored by Jean-François Chadelat, a senior civil servant, called in effect for “new cooperation between compulsory and complementary health insurance, based on co-payment and co-regulation” through an “ad hoc structure” involving NHI, PHI and State representatives (Chadelat 2003, p.  23). This would require a further unification of sickness funds as well as a single representative body for PHI. Interestingly, mutual benefit societies played a role in driving that change. During their annual congress held the same year, they asked for their greater involvement in health policy. To a certain extent, this posture may be regarded as a response to increasing competition induced by Insurance directives, with mutual benefit societies trying to secure their position by developing partnerships with the State. The 2004 health insurance reform endorsed the majority of conclusions made in the Chadelat report. This law notably creates a single body in charge of the management of the sickness fund (UNCAM, for Union nationale des caisses d’assurance maladie) and crucially, a National Union of Complementary Health Insurance Organizations (UNOCAM, for Union Nationale des Organismes Complémentaires d’Assurance Maladie). For PHI, this reform has arguably two main implications. The first is an explicit recognition of the existence of a PHI market, with a diversity of health insurers; but simultaneously (and somewhat paradoxically), the formation of a single representative body as demanded by the government also signifies that all health insurers should now consider themselves as “partners” of NHI under the State’s auspices. On this broad institutional basis, several governments, while committed to limiting the increase of healthcare costs for NHI, will increasingly rely on PHI to serve their own policy objectives. As they cannot change the market’s structures, which now depends on European provisions, it is essentially through the form and the scope of insurance contracts that policy-makers are now trying to intervene.

210 

G. CORON ET AL.

4.1  Achieving Universal Health Coverage through the Private Sector As discussed in the first section, PHI is essential for French people to get a full or simply to get access to certain health benefits. In this context, achieving universal health coverage has constituted a major policy objective pursued by various French governments over the last decades (Nay et al. 2016). Yet, in an overall institutional context where parliamentary and ministerial actors are placing strict financial constraints on NHI funding, it is essentially through ensuring better access to PHI that governments try to fulfill such an objective. In turn, these initiatives conflict with the prior effects induced by Insurance and Solvency II directives. Access to PHI in France is quite broad, but a number of social categories such as the unemployed, young people, low-income retirees or precarious workers cannot afford the cost of health insurance and are overrepresented in the 5–6% of French that are deprived of PHI. By 1999, a law guaranteeing universal access to NHI had already existed (the Couverture maladie universelle, or CMU, literally translated as “Universal Sickness Coverage”). Then, a CMU-C (where C stands for “complementary”) was also created to allow people below the poverty line to have access to free complementary health care. But in so doing, it automatically generated a threshold effect: any excess over the CMU-C ceiling led to the payment of a full insurance premium. On the basis of this diagnosis, the Douste-Blazy law of 2004 created an Assistance for Complementary Health coverage (ACS, for Aide à la complémentaire santé) targeted at people above the CMU-C resource ceiling by 35% (this rate of overrun has been raised several times ever since). Operated as a voucher-based scheme, the ACS gives each beneficiary the opportunity to acquire the health insurance of his choice. In this way, a side market subsidized by the State was created, and competition between health insurers was reinforced on the domestic level. However, the effects of ACS have been mixed. Several years after its implementation, the reform has barely increased access to PHI for the most fragile and precarious groups. This overall diagnosis was endorsed by the candidate of the Socialist Party François Hollande during the 2012 presidential race. Hollande did not state in his manifesto that he would extend health coverage through NHI. Rather, he promised to increase regulation over PHI contracts and to cut taxes on these same contracts to make them more attractive. One might thus observe a light partisan effect if we compare his pledge with

7  EUROPEANIZED, MARKETIZED BUT STILL GOVERNED BY THE STATE?… 

211

the rationale behind ACS when implemented by a right-wing government, yet both policies are embedded in a very similar framework, in which the possibility of extending coverage by NHI is not envisaged. Mutual benefit societies played a significant role in Hollande’s proposal, again in an attempt at securing their position in an increasingly competitive environment, now compounded by the prospect of having to cope with Solvency II. One of the demands by mutual benefit societies demands related to the contracts targeted by tax cuts, namely that they be based on an individual-level, and not as corporate-level contracts which provident institutions are better placed to offer. Moreover, they urged that these contracts be “responsible”, meaning that they comply with a number of principles consistent with NHI coverage (contrary to several types of contracts already offered by insurance companies). Soon after his election as President, Hollande presented his project at the FNMF 40th annual meeting. This presidential project would have increased the State’s power over PHI and, potentially, contributed to the restoration of links between the Socialist Party and mutual benefit societies. Yet it never reached the political agenda. At the same time, employer and employee representatives negotiated a national agreement based on “competitiveness and employment security”. The goal of this text, which has initially nothing to do with PHI, was to negotiate a liberalization of the French Labor code, a demand that emerged from employers’ representatives. While several trade unions strongly opposed it, the Democratic French Confederation of Labor (CFDT, a union close to Christian and social-democracy) accepted some of these changes but imposed a compensation, according to which “all employees should have access to a corporate-level complementary health insurance to be funded at least 50% by employers” (Benoît and Coron 2019). On this basis, an agreement was reached in 2013 and translated into law as the Employment Security Act. Even if this text partly conflicts its own proposal, the government rapidly presented it as achieving the same goals as Hollande’s project, namely to universalize access to health coverage (ibid). Yet, it is hardly the case, since it focuses on reducing inequalities between employees, and crucially not between wage earners and outsiders (unemployed, civil servants, self-employed, retirees or students; see Franc and Pierre 2015, also Pierre and Jusot 2016). More fundamentally, the reform opened a new space for competition within the French PHI market, as it automatically raises a demand for corporatelevel, collective contracts. As such, it extends prior effects induced by

212 

G. CORON ET AL.

Europeanization while combining with it. Indeed, the French Mutuality sector could hardly oppose such reform as, between 2012 and 2013, a number of M&As or alliances between mutual benefit societies and provident institutions were already being formed in anticipation of Solvency II. 4.2  Who Should Care about Care Providers? The path towards universal health coverage in France reveals that the marketization of PHI is not solely induced by Europeanization. It is also not necessarily at odds with a parallel increase in State influence in the sector, as both revealed by the case of ACS and president Hollande’s proposal. Even if public authorities had not initiated the Employment Security Act of 2013, health insurance contracts remain heavily regulated (see below); more importantly, what this overall process suggests is that even when health policy towards PHI is not driven by Europeanization, the prior effects of Insurance and Solvency II directives are increasingly perceptible. A very similar logic applies to care provision and to the regulation of healthcare professionals, a second domain where governmental intervention on insurance contracts led to several transformations of the PHI market. As seen in the first section, one of the most salient traits of the French healthcare system is that price regulation in ambulatory care rests on a corporatist basis, as the price of benefits and services provided by healthcare professionals are based on national agreements negotiated between the latter’s representatives and public authorities. In this broad context, governmental intervention has been historically lower for segments where the share covered by NHI is weaker, typically in dental and optical care, as well as for hearing aids—three sets of benefits or services for which PHI largely exceeds NHI provision. In the above-mentioned context where successive governments have tried to improve access to care without increasing NHI intervention, one might observe attempts during the 2000s and the 2010s to put additional pressures on healthcare professionals operating in these domains in order to limit out-of-pocket expenditures and potentially, the concurrent rise in the cost of health insurance contracts (as the two are usually correlated). Yet, most of these initiatives have proven rather unsuccessful. During the same time period, private health insurers also led several initiatives to limit the share of the same costs they also support. Managed care organizations (MCOs) have been crucial in instrumenting such

7  EUROPEANIZED, MARKETIZED BUT STILL GOVERNED BY THE STATE?… 

213

strategy. MCOs rest on agreements passed between an insurer and several healthcare professional or facilities. Through the creation or the significant support of such networks, health insurers are able to limit the rise of expenditures, as healthcare professionals are committed to respecting fixed prices or rates, and as patients are financially incentivized to consult these professionals through the modulation of rates and levels of reimbursement. As NHI and the State in France retain significant powers when it comes to price regulation of healthcare benefits and services, MCOs have always played a more moderate role here than in other country, typically in the United States (Rodwin 2010). Yet, stagnation in NHI coverage and high out-of-pocket expenditures gradually emerged with the increasing competitive environment created by Insurance directives as well as stricter risk management brought about by Solvency II. In turn, this constituted a fertile ground for MCOs, which developed significantly during the 2000s and the 2010s. On the recent years, such a rise has been paralleled by the formation of large management platforms playing the role of brokers between insurers and healthcare professionals. As a result, a growing share of health services and benefits is now offered and reimbursed through large MCOs on behalf of health insurers. Such transformation has been partly driven by insurance companies, for which MCOs have been common tools for risk management since the 1980s. But it is a direct response to deep changes in the positioning of mutual benefit societies. For these entities too, managed care did not start some 20 years ago. Indeed, and since the very beginning of the Mutualist movement, most mutual benefit societies have developed their own network of professionals and facilities, initially as part of their overall commitment to improve access to care. Yet, these networks were not MCOs in the traditional sense of the term, essentially because mutual benefit societies always refused to engage in the modulation of rates and levels of reimbursement, and thus not to limit access to care, a peculiarity furthermore extended by a law forbidding mutual benefit societies of doing so. That said, these longstanding values and legal provisions have not prevented mutual benefit societies from developing MCOs on their own during the 2000s, especially since large groups formed after Solvency II now intend to use their size to engage in tighter bargaining with healthcare professionals (Benoît and Coron 2019). These sectoral transformations correspond to governmental agendas as of the mid 2010s, resulting in a more or less explicit (though increased) recognition and indirect use of MCOs to reduce out-of-pocket

214 

G. CORON ET AL.

expenditures. In 2014, mutual benefit societies obtained the right to modulate rates and levels of reimbursement, which allow them to pursue the development of MCOs. The Socialist government issued a law (Loi Le Roux) that explicitly recognizes such entities. By the same token, it also aligns provident institutions, insurance companies and mutual benefit societies with the same regulations, furthering their institutional integration. The same year, following an overall failure of previous administrations to limit price increases by healthcare professionals (including medical doctors, but also dentists or opticians), the Socialist government issued a decree that constrained health insurers not to reimburse services or benefits if these benefits exceed a certain amount of money. Health insurers thus cannot provide a full coverage to policyholders consulting health professionals charging extra fees, a very common practice in France, particularly in dental and optical out-patient care. As a result, and if they want to maintain broad-­based coverage, health insurers are incentivized to intervene directly on price charges by the same professionals, and it is essentially through MCOs that they can do so. One can thus view this policy as an implicit delegation of negotiating capacities to health insurers or, at least, as the creation of an additional source of pressure on healthcare professionals, complementing official negotiations under the auspices of NHI and the State. The governmental use of PHI (and indirectly MCOs) became more explicit a few years later. In his 2017 presidential manifesto, Emmanuel Macron proposed improving coverage for certain health benefits, ostensibly distancing himself from the project defended by his conservative opponent François Fillon who advocated for a partial privatization of Social Security (Benamouzig et al. 2017). President Macron’s project was finally adopted in 2019, and is expected to be fully implemented by 2022. It covers the benefits for which PHI and out-of-pocket expenditures are classified as high, namely optical care (22%), hearing aids (53%) and dental prostheses (43%). Despite some criticisms (especially by representative organizations of healthcare professionals), the government managed to gain sufficient support from health insurers and further, to negotiate implementation of the reform with their representative body, UNOCAM. It is based on a threefold compromise whereby the government accepts to increase health coverage slightly for some products while health professionals and care providers commit to lower prices for the same products. At the same time, health insurers are asked to maintain coverage without increasing their fees. Labelled 100% Santé (literally, “100% Health”), three

7  EUROPEANIZED, MARKETIZED BUT STILL GOVERNED BY THE STATE?… 

215

baskets of care are defined: the first is based on identification of where compromises shall apply; the second concerns products of possibly higher quality with more expensive goods or services but at a limited price (but where out-of-pocket expenditures may apply); and a third basket with free rates and the highest out-of-pocket expenditures. As NHI coverage remains weak in the areas of hearing aids, dental and optical care even after this reform, it is PHI that is likely to support a critical part of the negotiations with healthcare professionals. As a result, the role played by MCOs is expected to expand with this reform, as it constitutes more than ever the more appropriate tools that health insurers possess to maintain an offer at the lowest possible price. The 100% Santé may thus be considered as a further recognition of the existence of the PHI market by the State, again in combination with its growing intervention both on the products offered and on the relationship between its different participants.

5   Conclusion Since the end of the 1980s, European law and regulation has been a driving force in the changes that have affected PHI in France. The country arguably provides a case of misfit (Cowles et al. 2001), according to which the more the operating model at the national level diverges from the underlying logic of European law, the greater, it is expected, its impact in the country in question. In that respect, institutionalized relationships within the French public-private mix were likely to be problematic for its integration within the Single market, as its private side was essentially formed of a single group of dominant firms, widely supported through a number of legal provisions. Moreover, these same firms were non-profit actors that were not competing with each other, and a range of formal and informal rules governing their activity prevented the development of a market for PHI. Yet the marketization scenario was not written in advance, and the significant developments of competition within French PHI, the rise of for-profit insurance companies and the growing financialization of the sector were all, and in many respects, contingent developments. If they provided with the institutional basis for the PHI market, Insurance directives were preceded by an important reform that paved the way for such transformation, with no obvious demands “from Brussels”. A number of mutual benefit societies did not oppose Solvency II, as the reform helps them legitimize shifts they were currently undertaking. For-profit insurance companies, while incentivized to diversify with Solvency II, already had an agenda in healthcare.

216 

G. CORON ET AL.

Beside this variety of usages of the European framework, and the many effects they entailed at the sectoral, firm and organizational levels, this paper has argued that marketization of PHI was also intimately linked to another trend, namely what others as labelled a “statization” of French healthcare (Hassenteufel and Palier 2007). Under this overall process, the State extends its reach in healthcare without increasing the share of benefits covered through NHI. The result for the public-private mix is a growing reliance of public authorities on PHI to serve a number of policy objectives, as prominently exemplified by the case of “universal” health coverage and price regulation among healthcare professionals. While the European framework now increasingly shapes the market’s basic structure, governmental intervention focuses on the content of insurance contracts through various (and often discrete) instruments. We have discussed elsewhere the consequences of such co-evolution for French healthcare (Benoît and Coron 2019; see also Chap. 10). In a nutshell, the growing marketization of the public-private mix already seems to be detrimental to individual welfare as well as the scope of health coverage. This is largely due to the fact that domestic politicians (especially in the course of the most recent reforms discussed in Sect. 3) have been largely unaware of the effects Insurance and Solvency II directives had on PHI, and, one might add, in much the same way as the potential implications of these texts for the healthcare provision insurance sector in France have been poorly acknowledged at the European level.

References Abecassis, P., Coutinet, N., & Domin, J.-P. (2014). Les principes mutualistes confrontés aux modalités de regroupements des organismes complémentaires d’assurance-­maladie. Revue internationale de l’économie sociale (RECMA), 331, 60–74. Abecassis, P., Coutinet, N., & Domin, J.-P. (2018). Les transformations de l’assurance maladie complémentaire à la lumière de la démutualisation/hybridation des banques coopératives. Revue d’économie industrielle, 161, 9–38. Benamouzig, D. (2012). Du grand soir au clair-obscur. Expertise économique et privatisation bureaucratique de l’Assurance maladie. Actes de la recherche en sciences sociales, 93, 56–73. Benamouzig, D., Chêne, G., & Jusot, F. (2017, May). President Macron and his Health Policies. BMJ Opinion, Online. Benoît, C., & Coron, G. (2019). Private Health Insurance in France: Between Europeanization and Collectivization. The Milbank Quarterly, 97(4), 1108–1150.

7  EUROPEANIZED, MARKETIZED BUT STILL GOVERNED BY THE STATE?… 

217

Chadelat, J.-F. (2003). La répartition des interventions entre les assurances maladie obligatoires et les complémentaires en matière de dépenses de santé. Paris: La Documentation Française. Cole, A. (2001). National and Partisan Contexts of Europeanization: The Case of the French Socialists. Journal of Common Market Studies, 39(1), 15–36. Cowles, M.  G., Caporaso, J., & Risse, T. (2001). Transforming Europe: Europeanization and Domestic Change. Ithaca: Cornell University Press. Domin, J.-P. (2019, July 4). Marché de la prévoyance complémentaire ou monopole mutualiste. Retour sur un débat fondateur (1983–1985). Paper presented at the Association française d’économie politique, Annual meeting, Lille. DREES. (2018). Les dépenses de santé en 2017. Panoramas de la DREES, S 18-132. Paris: Direction de la Recherche, des Études, de l’Évaluation et des Statistiques (DREES). DREES. (2019). Les complémentaires santé. Acteurs, bénéficiaires, garanties. Édition 2019. Panoramas de la DREES, S 19-135. Paris: Direction de la Recherche, des Études, de l’Évaluation et des Statistiques (DREES). Franc, C., & Pierre, A. (2015). Compulsory Private Complementary Health Insurance Offered by Employers in France: Implications and Current Debate. Health Policy, 119, 111–116. Gingrich, J. (2011). Making Markets in the Welfare State: The Politics of Varying Market Reforms. Cambridge: Cambridge University Press. Graz, J.-C. (2019). The Power of Standards: Hybrid Authority and the Globalisation of Services. Cambridge: Cambridge University Press. Graziano, P., Jacquot, S., & Palier, B. (Eds.). (2011). The EU and the Domestic Politics of Welfare State Reforms. Basingstoke: Palgrave. Hall, P. (1985). Socialism in One Country: Mitterrand and the Struggle to Define New Economic Policy for France. In P. Cerny & M. Schain (Eds.), Socialism, the State and Public Policy in France (pp. 81–107). London: Frances Pinter. Hassenteufel, P. (2001). Liberalization through the State: Why is the French Health Insurance System Becoming so British? Public Policy and Administration, 16(4), 84–95. Hassenteufel, P., & Palier, B. (2007). Towards Neo-Bismarckian Healthcare States? Comparing Health Insurance Reforms in Bismarckian Welfare Systems. Social Policy and Administration, 41(6), 574–596. Hassenteufel, P., Smyrl, M., Genieys, W., & Moreno-Fuentes, J. (2010). Programmatic Actors and the Transformation of European Healthcare States. Journal of Health Politics Policy and Law, 35(4), 517–538. Machin, H., & Wright, V. (1985). Economic Policy and Policymaking under the Mitterrand Presidency, 1981–1984. London: Pinter. Ministère des Affaires sociales et de la Santé. (2017). Programme de qualité et d’efficience maladie. In Projet de loi de financement de la Sécurité sociale pour l’année 2018. Paris: Ministère des Affaires sociales et de la Santé.

218 

G. CORON ET AL.

Montaut, A. (2017). La situation financière des organismes complémentaires assurant une couverture santé. Rapport 2016. Paris: Direction de la Recherche, des Études, de l’Évaluation et des Statistiques (DREES). Nay, O., Béjean, S., Benamouzig, D., et al. (2016). Achieving Universal Health Coverage in France: Policy Reforms and the Challenges of Inequalities. Lancet, 28(387), 2236–2249. Pierre, A., & Jusot, F. (2016). The Likely Effects of Employer-mandated Complementary Health Insurance on Health Coverage in France. Health Policy, 121, 321–328. Rodwin, M. (2003). The Healthcare System Under French National Health Insurance: Lessons for Health Reform in the United States. American Journal of Public Health, 93(1), 31–37. Rodwin, V. (2010). The Metamorphosis of Managed Care: Implications of Health Reform Internationally. Journal of Law and Medical Ethics, 38(2), 352–364. Rousseau, M., & Elbarhdadi, L. (2016, January 5). French Mutual Insurers are Feeling the Many forces of Consolidation. Standard and Poor’s Rating Services. Sagan A., & Thomson, S. (ed.) (2016), Voluntary health insurance in Europe. Country experience. WHO Regional Office for Europe, European Observatory on Health Systems and Policies. Simonet, D. (2019). Technocratic Recentralization in the French Healthcare System: A Critical Evaluation. Journal of Health Planning and Management, 34, 825–835. Siney-Lange, C. (2018). La Mutualité, grande semeuse de progrès social. Paris: La Martinière. Sorum, P. (2005). France Tries to Save its Ailing National Health Insurance System. Journal of Public Health Policy, 26(2), 231–245. Tabuteau, D. (2010). La métamorphose silencieuse des assurances maladie. Droit Social, 85–92.

CHAPTER 8

Ireland: The Ambiguous Role of the Health Insurance Market Philippe Martin and Pascale Turquet

1   Introduction The configuration of Irish Healthcare is rather singular in Europe. Market mechanisms and forces combine with the public healthcare system and, strictly speaking, the term privatization rather than marketization should be used in order to describe this configuration. The fact is that private health insurances companies (PHIs) play a significant part in national health policy and, for that reason, are subject to tight regulations that aim to correct adverse market effects. This blurry situation has challenged the European legal framework, but also poses domestic problems, as will be outlined in this chapter. Private health insurance companies play an important role in Ireland. They currently account for around 12% of total spending on health, which

P. Martin (*) Centre for Comparative Labour Law and Social Security, Université de Bordeaux, CNRS, Pessac, France e-mail: [email protected] P. Turquet LiRIS, Université de Rennes 2, Rennes, France e-mail: [email protected] © The Author(s) 2021 C. Benoît et al. (eds.), Private Health Insurance and the European Union, https://doi.org/10.1007/978-3-030-54355-6_8

219

220 

P. MARTIN AND P. TURQUET

means that Ireland has one of the three largest private health insurance (PHI) markets in Europe (Turner 2016). Around 45% of the population is covered by a voluntary health insurance, allowing them access to private hospitals or to private beds in public hospitals and shorter waiting lists compared to the public system. Private health insurance mainly plays a duplicative role (as well as a complementary one) as it provides insured people with an alternative coverage to the public system. The point is that the Irish public healthcare system is not truly universal: as a National Health Service, it covers the whole population and is funded through taxation, but actually, many people have to pay fees for each visit to a doctor or hospital. Only a third of the population (37%) has access to free healthcare: those who are entitled to the Medical Card—that is, people on low incomes. The others have to pay for out-patient and in-patient care, as well as for drugs and other medical services, although these are subsidized.1 However, some types of healthcare are free for all: in case of long-­ term illness, or maternity and infant care, medical expenses are borne by national schemes—the Long-term Illness Scheme and the Maternity and Infant Care Scheme—assuming that people have registered for these schemes. Before 2009—that is, before the financial and economic crisis— people aged more than 70 years were automatically entitled to a medical card, but this measure has since been abolished.2 A number of companies offer voluntary private health insurance coverage in Ireland, but the current market is quite concentrated: VHI Healthcare is the original semi-public, not-for-profit insurer and its market share is currently 50%; Laya Healthcare (formerly British United Provident Association(BUPA) Ireland) has 26% market share and Irish Life Health— which has acquired Aviva Health and GloHealth—has 20% (HIA 2017). Another insurer, Hospital Saturday Fund (HSF) Health Plan, only provides cash benefits (not in-patient health insurance). Some entities only cover the employees and retirees of certain companies: these are restricted membership schemes. Most companies have agreements with hospitals so that the insured do not have to pay upfront costs. 1  As regards drug prescriptions, people who are not entitled to the Medical Card would pay a monthly capped amount (€144 per month maximum) if they sign up for the Drugs Payment Scheme, which is a kind a public insurance; above this amount, they are reimbursed by the scheme. 2  The Medical Card for people more than 70 years of age is currently means-tested, though the threshold for this category is higher than the usual one.

8  IRELAND: THE AMBIGUOUS ROLE OF THE HEALTH INSURANCE MARKET 

221

Why and to what extent the PHI market plays an ambiguous role in the Irish Healthcare system? The system is usually presented as “two-tiered,” which basically means that it offers advantages to those who can afford private health insurance: PHI duplicates public coverage but provides differential treatment; moreover, there are direct and indirect public subsidies for PHI and the incentive structures in place serve to distort resource allocation in favor of people with PHI (Turner 2016). Actually, many countries have a two-tier healthcare system and this is not necessarily an issue when the public scheme offers real universal coverage and adequate primary care. In these cases, such as in Spain for example, the PHI market plays its duplicative or complementary role in an autonomous way: sales and purchases of health insurance policies are regulated by economic and consumer law and insurers’ activity is not regarded as a public or social service. However, in Ireland, because the public statutory system is not universal, the legislation strongly regulates the PHI market to ensure that it plays a general interest role in healthcare coverage. In a certain way, the PHI market is seen as a tool for social protection and welfare policy. This was achieved through an original and creative transposition of the 1992 EU Insurance directive, whereby the Irish legislation mixed market rules and the interests of the common good. Nevertheless, it involved serious legal issues that led to EU law and national public policy being accommodated to PHI.  Moreover, the effects of the economic and financial crisis (2009–10 and 2015–16) called the two-tier Irish Healthcare system into question, as the proportion of people faced with unaffordable private health expenditure significantly increased. The Irish Government is currently reconsidering universality, and this might lead to a fundamental transformation of the role of PHI.

2   An Original Transposition of the 3rd EU Directive: Mixing Market Rules and the Interests of the Common Good 2.1  The Roots of the Irish System: Voluntary Insurance and Solidarity Principles For historical and cultural reasons, Ireland was reluctant to set up universal compulsory/public health insurance. Private and voluntary arrangements had always played an important role. Even before Irish independence, the Catholic Church and the medical profession opposed

222 

P. MARTIN AND P. TURQUET

the extension to Ireland of the medical benefit system funded by social insurance to provide free GP care and medications which was introduced in the rest of the UK (Wren and Connolly 2019). In 1947, the Irish Department of Health was established and the Irish government planned to offer a National Health Service for all citizens, free at the point of use. This proposal also failed because of opposition from the medical profession, the Catholic Church3 and the Department of Finance. Under the 1953 Health Act, the role of voluntary and community organizations was recognized, and the government was permitted to fund those organizations for health provisions. This approach was consolidated by the 1957 Voluntary Health Act, which introduced a state-sponsored health insurance scheme: the Voluntary Insurance Board, which subsequently became VHI Healthcare. This body offered private health insurance to the wealthiest 15% of the population who were not eligible for public hospital coverage (Colombo and Tapay 2004). This contributed to the development of a two-tier system, as people who did not qualify for free visits to dispensary doctors were induced to purchase a voluntary health insurance policy. The system was rather twisted in the sense that people who could not afford to pay the fees for a General Practitioner (GP) visit had a strong incentive to head to the hospital for admission, where all these services were free, resulting in the Irish tradition of visiting hospitals for minor treatments. This was also true for VHI patients, because their insurance covered in-patient, but not out-patient treatment.4 Between 1957 and 1997, the semi-public body VHI Healthcare was the only voluntary health insurer, so there was no real market for private health insurance. It can be considered that the Irish Healthcare system was partially private—VHI playing an important role in access among the population to healthcare—but not really “marketized.” Moreover, it has to be kept in mind that the provision of private health insurance has functioned pursuant to solidarity principles (Colombo and Tapay 2004). Under the 1957 Act, VHI Healthcare applied a set of rules such as community-­rated 3  Since the nineteenth century, Ireland had a tradition of Catholic Church-founded and controlled hospitals. They were formerly funded by charitable endowments augmented by grants from parliament and, from 1930, by a national lottery. In the modern era, they have been funded by taxation and private health insurance (Wren and Connolly 2019). 4  This is no longer the case, as nowadays, VHI comes mainly in the form of hospital plans providing access to semi-private or private rooms in public and private hospitals (Turner 2016).

8  IRELAND: THE AMBIGUOUS ROLE OF THE HEALTH INSURANCE MARKET 

223

premiums whereby all insured people paid the same premium for a given plan, irrespective of their health status, open enrollment—the insurer must accept all applicants—and lifetime coverage. 2.2  Opening Up the PHI Market: The “Irish Mix” Following the introduction of the EU’s Third Non-Life Insurance Directive in 1992, the PHI market was opened up to competition by the Health Insurance Act 1994 and the 1996 Health Insurance Regulations. BUPA Ireland, a branch of the British United Provident Association, started operations in 1997. However, during the first years of this opening, there were only two competitors—VHI Healthcare and BUPA—and in 2002, VHI Healthcare still retained a dominant share of the market with over 80%. Since then, other insurers have entered (but also left) the market, which is quite concentrated, as noted above. The Irish legislation complies with EU law as regards the health insurance market, but it also strongly regulates this market. A willingness to regulate private cover and to maintain the established solidarity-based system is clearly expressed in the 1994 Act (revised in 2012). The legislation imposes general interest obligations upon all health insurers whose activity is controlled by public authorities. It also implemented a risk equalization scheme in order to support community-rated health insurance. 2.2.1 The Objectives of the Irish Legislation These objectives are expressed by the 1994 Health Insurance Act as amended in 2012. According to the letter of the law, the main objective is to ensure broad access to health insurance, with no risk selection. The health insurance market is thus a medium for accessing healthcare5 and its functioning has to comply with “the interests of the common good,” which implies regulation. As noted by Colombo and Tapay (2004), “Regulation of private cover continued to be considered necessary to ensure the long-term sustainability of the established solidarity-based system. For this reason, Ireland obtained legal approval from the EU to apply open enrolment, community rating and lifetime cover requirements to the 5  In Ireland, the purchase of VHI is supported by the Government through a system of tax relief: the Health Insurance Tax Credit reduces the amount of the health insurance premium paid by people. In general, the lower the household income, the higher the credit.

224 

P. MARTIN AND P. TURQUET

expanded private health insurance market in order to protect the ‘general good.’” Section 1A of the Health Insurance Act

“The principal objective of this Act is to ensure that, in the interests of the common good and across the health insurance market, access to health insurance cover is available to consumers of health services with no differentiation made between them (whether effected by risk equalisation credits or stamp duty measures or other measures, or any combination thereof), in particular as regards the costs of health services, based in whole or in part on the health risk status, age or sex of, or frequency of provision of health services to, any such consumers or any class of such consumers (…)” For the purpose of that objective, the law states that some points must be taken more particularly into account, especially “the fact that health needs of consumers of health services increase as they become less healthy, including as they approach and enter old age.” The terms “interests of societal and intergenerational solidarity” are included in the text of the law in order to justify the fact that “the burden of the costs of health services (shall) be shared by insured persons by providing for a cost subsidy between the healthier and the less healthy, including between the young and the old.” This implies a system of risk equalization credits. The legislation also wishes to discourage insurance companies from engaging in practices or offering health insurance policies which, by segmentation of the health insurance market (or through whatever means), favor the coverage of the health insurance risk of the more healthy, including the young, over the coverage of the less healthy, including the old. It should be noted that this point was added into the 2012 Act in reference to the BUPA decision by the Court of First Instance of the EU (2018) in which the notion of “active risk selection” was analyzed and discussed as regards the Irish VHI market. 2.2.2 Private Health Insurance Obligations and Monitoring System The Health Insurance Act 1994 sets out four private health insurance obligations, which are designed to support the objective of the law.

8  IRELAND: THE AMBIGUOUS ROLE OF THE HEALTH INSURANCE MARKET 

225

Community rating: Insured persons pay the same level of premium for a given level of benefit, regardless of their health profile (age, gender or health status). Lifetime community rating was implemented in the Irish health insurance market on May 1, 2015. It is a modified version of community rating whereby the premium does not vary in respect of an individual’s age but involves premium loadings that apply to people aged 35 and above who had no health insurance and are entering the health insurance market. This system is expected to improve the sustainability of the market as a whole, but without undermining the fundamental principle of intergenerational solidarity. Open enrollment: Health insurers must accept all applications, regardless of age or health status. Some legislative measures aim to prevent adverse selection risk which could result in people taking out insurance just as they need it. Lifetime cover: An insurance policy cannot be terminated or fail to be renewed by the insurer without the consent of the insured person, even as the insured person ages or his physical condition declines. Minimum benefits: Insurers must provide a certain minimum level of benefits prescribed by the legislation for all insurance products. The minimum benefits requirements are designed to protect consumers from purchasing a product that does not provide at least a minimum level of cover. The health insurance market is monitored and regulated by a national independent authority, the Health Insurance Authority (HIA) created by the Health Insurance (Amendment) Act 2001. All health insurers operating in the Irish market have to be registered with the Authority; they also have to comply with EU (insurance directives) and national legislation, especially with regard to prudential rules. The HIA advises the Minister for Health, either at his or her request or on its own initiative, on matters relating to health insurance. It can issue enforcement notices to enforce compliance with the legislation (the Health Insurance Act) or engage in prosecutions. One of the main missions of the HIA is to administrate the risk equalization scheme initially introduced in 1996. 2.2.3 A Risk Equalization Scheme Risk equalization is a process that aims to equitably neutralize differences in insurers’ costs that arise due to variations in the age profile of their members. It involves transfer payments between health insurers to spread some of the claim costs of high-risk members amongst all the private health insurers in the market, in proportion to their market share. The

226 

P. MARTIN AND P. TURQUET

1994 Act provided the government with the authority to implement a risk equalization scheme (hereinafter RES) to support community rating. An initial scheme was established in 1996 but was revoked in 19996 without any payments having been introduced. A new scheme was introduced by the Health Insurance (Amendment) Act 2001 and implemented in July 2003. Under this legislation, the Health Insurance Authority was entitled to make recommendations on whether the scheme would begin to require payments to be transferred from one insurer to another if the market equalization percentage, which measures the difference in risk profiles between insurers, was between 2 and 10%.7 The 2003 scheme provided for the risk equalization of in-­ patient costs on the basis of age and gender, that is, the factors used in assessing the risk of an insurer’s population were age and gender.8 These provisions were seriously challenged by BUPA Ireland and were neutralized by the Irish Supreme Court which quashed the scheme in 2008.9 Due to this decision by the Supreme Court and during the period 2009–2012, interim measures were implemented: a system of age-based tax credits in order to support community rating. The 2009 Act provides that insurers receive higher premiums for insuring older people, but that older people receive tax credits equal to the amount of the additional premium so that all people continue to pay the same amount for their health insurance. In this way community rating is maintained but insurers receive 6  The RES was discussed through a consultation process, and an independent advisory group on risk equalization was tasked with preparing a White Paper on Private Health Insurance which was adopted by the Department of Health and Children in 1999. In this White Paper, it was decided to preserve the principle of community rating and the advisory group also concluded that risk equalization was necessary in a community-rated market. Nevertheless, the Government considered that the RES as it was designed could lead to uncontrolled costs and price increases and did not reward efficient insurers. The Irish Government thought that “additional competition has the potential to mitigate price increase pressure by, in particular, encouraging insurers to adopt cost containment measures.” It finally decided that the RES should be amended. 7  Insurers were required to file returns every six months. Upon the basis of these returns, the HIA was supposed to assess the risk profiles of insurers and recommend to the Minister for Health and Children whether payments should be made under the scheme. New entrants were provided a limited opportunity to opt out of the scheme for three years, together with a phase-in period during which only 50% of any payments would be required. 8  In 2001, VHI Healthcare’s market share was 85% of which 9% for people more than 65 years of age (and 23% between 49 and 64 years); BUPA’s market share was 8% of which 4% for people more than 65 years of age (and 6% between 49 and 64 years). 9  About the legal issues, see Sect. 2.

8  IRELAND: THE AMBIGUOUS ROLE OF THE HEALTH INSURANCE MARKET 

227

higher premiums for older people to partly compensate for the higher level of claims. In order to fund the system, insurers paid a Community Rating Levy in respect of all individuals covered for health insurance.10 Finally, and after a real “legal saga,” a RES was reintroduced in 2013.11 It is rather similar to the interim measures. Like the interim system, the 2013 RES involves insurers receiving higher premiums for insuring members of less healthy groups of the population. Again, like the interim system, the upper part of the premium for these individuals is not paid by the individual, rather it is paid in the form of risk equalization credits. It follows that, while each individual pays the same net premium for the same product, the insurer receives higher premiums for insuring members of less healthy groups. Another similarity with the interim system is that the credits are funded by a stamp duty payable by insurers for each person that they insure. The main differences with the interim system are that risk equalization credits are paid from a fund operated by the Health Insurance Authority rather than in the form of tax credits. Risk equalization credits payable in respect of premiums vary on the basis of age, gender, and level of cover, rather than just on the basis of age. Stamp duty rates vary between children and adults and between two levels of cover. Risk equalization credits are now also payable for claims, with a fixed amount payable from the Risk Equalization Fund for each night an insured person spends in private hospital accommodation. This reduces the cost to the insurer of insuring less healthy individuals. Stamp duty levies and risk equalization credits are revised every year on the basis of a report made by the HIA and provided to the Minister for Health. It should be noted that the system set up in 2013 was designed to comply with EU competition law requirements. In this regard, the HIA carries out an overcompensation (ex post) test in accordance with the notion of “reasonable profit” established by the ECJ in Altmark12 and with the 2012 Service of General Economic Interest (SGEI) Framework.13 Thus, all insurers are required to maintain and to provide to the HIA yearly profit and loss statements as well as certified balance sheets in respect of  In 2009, this levy was €160 for adults and €53 for children.  In 2012, the Irish authorities notified the Commission of the new 2013 RES.  The Commission authorized the 2013 RES by decision of February 20, 2013. 12  ECJ July 24, 2003, Case C-280/00, Altmark Trans GmbH and Regierungspräsidium Magdeburg/Nahverkehersgesellschaft Altmark GmbH. 13  Communication from the Commission: European Framework for State aid in the form of public service compensation, OJ C 8, 11.1.2012, p. 15–22. 10 11

228 

P. MARTIN AND P. TURQUET

Table 8.1  Risk equalization premium credits Contract type

65–69 70–74 75–79 80–84 85+

Non-advanced

Advanced

Male

Female

Male

Female

€400 €725 €1075 €1450 €2175

€300 €550 €850 €1150 €1450

€1000 €1750 €2550 €3450 €4975

€650 €1250 €1925 €2700 €3350

Source: The HIA (2018)

Table 8.2  Community rating stamp duties Community rating stamp duties (contract type)

Non-advanced

Advanced

Adult Child

€240 €80

€399 €135

Source: The HIA (2018)

their health insurance business, and any other information the HIA may deem necessary. Tables 8.1, 8.2 and 8.3 show the rates applied for contracts commencing and/or renewing from April 1, 2018, onward.

3   Legal Issues: Accommodating EU Law and National Public Policy to PHI The Irish health coverage system based on a strongly regulated PHI market posed problems as regards EU law, especially EU competition law and insurance law in the internal market. The risk equalization scheme was challenged in Irish and European Courts as breaching European rules on state aid. Furthermore, the legal status of VHI Healthcare was challenged in respect of the “three generations” of EU Insurance directives. The results of this confrontation between EU law and national public policy towards PHI show a sort of accommodation, or a reconciliation of the two rationales: the Irish legislation had to adapt to the requirements of EU law as regards the legal status of health insurers operating in the market; but EU competition rules were also interpreted in a flexible way that allowed Member States such as Ireland to implement solidarity mechanisms in the health insurance market.

8  IRELAND: THE AMBIGUOUS ROLE OF THE HEALTH INSURANCE MARKET 

229

Table 8.3  Profitability of insurers on the Irish Private Medical Insurance (PMI) market in 2014 €m

Aviva Health

Great Lakes Elips (Laya Quinna VHI (GloHealth) Healthcare) Healthcare

Earned premiums before reinsurance and risk equalization credits Claims incurred before reinsurance Expenses and reinsurance Underwriting result (gross of RES) Impact of RES Underwriting result (net of RES) Impact of investments Profit before tax (gross of RES) Profit before tax (net of RES) Profit as a % earned premiums (gross of RES) Profit as a % earned premiums (net of RES)

327.2

64.8

465.8

0

1446.7

(239.1)

(40.4)

(344.5)

1.4

(1385.0)

(45.7) 42.4

(4.7) 19.7

(64.2) 57.1

0.5 1.8

(104.2) (42.5)

(31.5) 10.9

(17) 2.7

(48.1) 9

3.2 5.0

71.8 29.3

0.5 42.9

0 19.7

0.2 57.3

1.8

18.6 (23.9)

11.4

2.7

9.2

5.0

47.9

13.1%

30.4%

12.3%

n.a.

(1.7%)

3.5%

4.2%

2.0%

n.a.

3.3%

Source: The HIA (2018) a Quinn Insurance Ltd (trading as Quinn Healthcare) closed to new business in May 2012, and existing policyholders were invited to renew their policies with Elips Insurance Ltd (trading as Laya Healthcare)

3.1  Competition Law and the Irish Risk Equalization Scheme As soon as BUPA Ireland entered the PHI market, this insurer questioned the proper existence and the purpose of the RES. On March 2, 1999, BUPA lodged a complaint with the European Commission against the implementation of a new RES planned by the Irish Government. BUPA considered that the RES would breach article 87§1 of the Treaty establishing the European Community which prohibits state aids. Between March 1999 and April 2003, BUPA Ireland added memorials, submissions and reports in order to supplement its plea, but these legal proceedings did not succeed. When the RES was reintroduced in 2003, BUPA immediately opposed it, considering among other points that the purpose of the scheme was to

230 

P. MARTIN AND P. TURQUET

grant money to the formerly monopolistic organization, VHI Healthcare, which usually covered the older and less profitable sector of the population.14 In 2005, according to the recommendations of the HIA, the Minister of Health communicated to BUPA Ireland the payments it had to make pursuant to the new rules from January 1st 2006. BUPA then commenced legal action before the Irish Courts and suspended its payments, waiting for a definitive decision. In November 2006, the High Court judged in favor of the application of the RES legislation. BUPA appealed this decision at the Supreme Court, which granted BUPA’s appeal and ruled that the RES was invalid because it was based on an incorrect interpretation of the terms “community rating across the market for health insurance” in the provisions of the Health Insurance Act 1994.15 The RES was considered by the Supreme Court as having a strong impact on BUPA’s market position and chances of profit. However, while the case was still pending before the Irish Supreme Court, BUPA brought a complaint before the Court of First Instance of the European Communities against a decision by the European Commission which had decided not to raise an objection to the establishment of the RES in Ireland.16 BUPA claimed that the scheme was a State-­ enforced subsidy for VHI Healthcare, and was anti-competitive. In the challenged decision, the Commission had considered that “the RES involves payments which are limited to the minimum necessary to compensate private medical insurers for SGEI obligations and therefore does not involve State aids in the sense of Article 87(1) EC.” Indeed, according to the Commission, health insurers in Ireland were in charge of performing a Service of General Economic Interest (SGEI) which can justify an exception to the application of competition rules.17 14  At this time, it was estimated that VHI healthcare covered 20 times more people aged over 80 years than BUPA. 15  Irish Supreme Court, BUPA Ireland v Health Insurance Authority, July 16, 2008, IESC 42. 16  Decision C (2003) 1322 final, of May 13, 2003, State aid N 46/2003-Ireland. 17  The notion of SGEI was introduced into the European Treaties at article 86 of the EC Treaty (currently article 106§2 of the Treaty on the Functioning of the European Union (TFUE)), which states that “Undertakings entrusted with the operation of services of general economic interest or having the character of a revenue-producing monopoly shall be subject to the rules contained in the Treaties, in particular to the rules on competition, in so far as the application of such rules does not obstruct the performance, in law or in fact, of the particular tasks assigned to them. The development of trade must not be affected to such an extent as would be contrary to the interests of the Union.”

8  IRELAND: THE AMBIGUOUS ROLE OF THE HEALTH INSURANCE MARKET 

231

The BUPA case as judged in 2008 by the Court of First Instance of the European Communities18 was abundantly commented upon, and it is not the intention of this chapter to give a redundant commentary.19 Basically, the legal debate was about state aid versus the compensation approach (Sauter 2008) and the Court of First Instance of the European Communities finally dismissed BUPA’s claim and recognized the validity of the compensation system set up by the Irish Government. Some relevant points of the judgment can be highlighted. Firstly, the legal background taken into account by the Court is quite meaningful, referring as it does to the EC Treaty’s provisions about competition law (art. 86 and art. 87), but also to article 16 which recognizes the place occupied by SGEI in the shared values of the Union, and article 152 which provides that “A high level of human health protection shall be ensured in the definition and implementation of all Community policies and activities” and states that “Community action in the field of public health shall fully respect the responsibilities of the Member States for the organisation and delivery of health services and medical care.” Next, regarding the responses of the Court to the numerous arguments of the applicants (BUPA and others), it can be observed that they are globally determined and justified by the use of the legal theory of the “manifest error of assessment” which clearly limits the scope and depth of legal control by the Commission (and therefore by the Court) upon national measures such as the implementation of a RES. This led to a flexible approach to the notion of SGEI. The main arguments of the claimants were based on incorrect interpretation and enforcement of articles 87§1 EC (prohibition of State aids) and 86§2 EC (the SGEI exception) by the Commission. According to BUPA, the conditions for financial compensation for SGEI obligations, as recognized by the Court of Justice (in the Ferring and Altmark rulings),20 were not satisfied. BUPA notably considers that a public service is universal and compulsory as regards the service provision, and the fact that an activity is authorized, controlled and 18  Court of First Instance of the European Communities, February 12, 2008, Case T-289:03, BUPA and others v Commission. 19  For an in-depth study of the BUPA case, see in particular Sauter (2008); also Biondi (2008). 20  ECJ, November 22, 2001, Case C-297/15, Ferring SA versus Agence centrale des organismes de sécurité sociale (ACOSS) ; ECJ, July 24, 2003, Case C-280/00, Altmark Trans GmbH and Regierungspräsidium Magdeburg versus Nahverkehrsgesellschaft Altmark GmbH.

232 

P. MARTIN AND P. TURQUET

regulated is not enough to conclude that it is a service of general economic interest. In addition, according to BUPA, the notion of SGEI is an autonomous concept of Community Law and a Member State cannot decide on its own whether there is a SGEI or not. BUPA considered there were no real SGEI obligations in the Irish system and no genuine act of public authority establishing SGEI missions. The claimant also questioned the necessity and the proportionality of the RES. The Court considered that, due to the special nature of the SGEI mission invoked in this case, the criteria established in Altmark had to be adapted, which led some legal scholars to wonder whether BUPA’s case overruled or expanded the Altmark ruling (Sauter 2008). It also considered that there was no clear and precise definition of the notion of SGEI in European law and therefore, that the Member States had wide discretionary power in this respect. For this reason, control by the Commission should be limited, and this was also justified by the fact that the debate took place in the field of health policies, which falls within the competence of national law. Regarding the existence of a SGEI mission in this case, the Court considered that the Irish legislation not only authorized health insurers to operate and regulate their activity, but that it constituted a true “act of public authority” involving the creation and definition of a particular mission. The objective of the legislation was clearly to serve the general interest by offering alternative health coverage to half of the Irish population. The PHI obligations limit insurers’ commercial freedom to an extent that goes beyond the ordinary conditions of authorization to operate in a specific sector. It should be observed that, for the Court, the limitation of commercial freedom was the criterion that served to recognize that there was actually an act of public authority, and therefore to conclude upon the existence of a SGEI. Concerning the universal nature of a public service, according to the Court, this did not mean that the service should cover the whole population. The Court considered the Irish Health Insurance legislation (open enrollment, community rating and lifetime cover) as aiming to guarantee broad access to health coverage for the Irish population, so that PHI services could be seen as universal under EU law. Regarding the compulsory nature of a public service, the Court stated that it had to be understood in the sense that the operators in charge of the SGEI mission had a duty to provide the (health insurance) service in the market in compliance with the SGEI obligations. The fact that the insurance was voluntary did not matter.

8  IRELAND: THE AMBIGUOUS ROLE OF THE HEALTH INSURANCE MARKET 

233

As regards the necessity and the proportionality of the RES, the Court limited its appreciation to the control of “manifest error,” which means that the Court relied on the appraisal by the European Commission of complex economic facts and that the claimant had to provide evidence in order to prove that this appraisal was not plausible. The Commission itself limits its control of a Member State’s appraisal of the context in which the PHI market evolves. The Commission only has to check whether the risk of dysfunction of the PHI market as assessed by the Member State is plausible, and the Court only judged whether the Commission could reasonably estimate that the compensation system (the RES) was in this context necessary and appropriate. It recognized in this case that the compensation system, that is, the Irish RES, was necessary and proportionate: in the absence of the RES, there would be a risk of “active risk selection” in a community rating market, in the sense that an insurer could attract young customers with “good risk profile” by offering coverages and policies specially adapted to their needs and with attractive premiums. Therefore, to the Court, there was a plausible risk of market instability in the absence of RES. Finally, it estimated that the compensation system was appropriate and proportionate as the claimants had not provided sufficient evidence that the measure of the difference in risk profiles between insurers (between 2 and 10%) by the Irish HIA in order to trigger the RES was manifestly disproportionate. Ultimately, the Court’s ruling and the control by the European Commission tend to show that there is no “legal orthodoxy” in this matter, in the sense that pure market rules and competition law are not considered as the absolute and necessary condition to govern access to healthcare in countries such as Ireland, in which PHI is de facto a vehicle for social protection. 3.2  The Status of VIH Healthcare Despite the opening of the health insurance market to competition in 1994, VHI Healthcare, the initial not-for-profit organization, continued to be exempted from the application of EU Insurance directives. As a result, VHI Healthcare did not have to satisfy prudential financial solvency requirements as an insurance company. VHI Healthcare’s competitors claimed that its exemption from the regulations of the Insurance Acts gave it an unfair advantage because it was not required to hold a minimum level of reserves to guarantee solvency and did not need to establish subsidiaries

234 

P. MARTIN AND P. TURQUET

to engage in other business activities. Changing the regulatory status of VHI Healthcare to bring it in line with its competitors was initially proposed in 1999, and again in 2007 by the HIA which recommended that “VHI Healthcare should be obliged to operate in the provision of non-­ insurance services in the same manner as if it was regulated as an authorized non-life insurance company. The Minister for Health and Children should bring forward legislation to amend the Voluntary Health Insurance Acts including a requirement that VHI Healthcare be allowed and obliged to establish associated companies (or subsidiaries) to carry out non-­ insurance activities.”21 It should be noted that the “doctrine” of the HIA tends to mix public interest and competition (it is actually rather market-­ oriented). In its report, the HIA asserts that “a particular model of PHI market has evolved in Ireland with a small number of competitors as a result of the history of Irish policy decisions and EU Insurance legislation. In this report, the different policy and legal principles of competition, insurance, regulation, Irish healthcare and health insurance are considered together in a single consistent manner. The overall objective of the Report’s recommendations is to encourage competition in the Irish PHI market in order to best serve society and health insurance consumers” (HIA 2017). However, nothing was done and a complaint was brought before the European Commission. In November 2007, the European Commission issued a reasoned opinion to Ireland requesting that it take the measures necessary to ensure compliance with the EU non-life insurance directives. The case was brought before the European Court of Justice (ECJ) which rendered a judgment in 2011.22 The ECJ held that Ireland had an obligation under Article 3 of the Third Non-life Insurance Directive to ensure that monopolies in the non-life insurance market within their territories were abolished by July 1, 1994, and that VHI Healthcare was not immune to this provision. The exemption was thus unlawful. VHI Healthcare had to submit an authorization application to the Central Bank, like all health insurers, with a deadline of May 2014. VHI Healthcare finally met the solvency requirements in January 2016.

21  HIA, Competition in the Irish Private Health Insurance Market, Report to the Minister for Health and Children, January 2007. 22  ECJ, case C-82/10, Commission versus Ireland, September 29, 2011.

8  IRELAND: THE AMBIGUOUS ROLE OF THE HEALTH INSURANCE MARKET 

235

4   The Irish Healthcare System in Crisis Private Health Insurance (PHI) plays a key role in access to healthcare. As explained above, it is highly regulated because it is considered to be a tool for social protection and welfare policy. However, PHI is not universal and has always been controversial, especially regarding its subsidization by the Irish state through tax reliefs and privately insured patients’ access to public hospital beds. This became a critical issue during the last economic crisis because the Irish state started to reduce public spending on health in 2009, initiating a cost transfer from the state onto people and causing financial hardship for people with lower incomes. In 2011, the new government decided to end the unfair, two-tier system of access to healthcare by charging privately insured patients for the use of all beds in public hospitals and capping the tax relief on premiums. The latter measure resulted in a drastic increase in the premiums as well as a significant decrease in the number of insured persons. The universal healthcare insurance project was eventually abandoned for financial reasons, but the two-tier health insurance system is nevertheless still under pressure. Market regulation alone is not enough to guarantee universal access to health and PHI obviously creates inequalities. The recent Sláintecare report provides for a separation of the private and public elements in the healthcare system, removing private practice from public hospitals and questioning the role of PHI.  The implementation of this universal single-tier service would mean a substantial overturning of the system given the current role of private insurance and the sheer importance of this market. 4.1  Public Expenditure and Access to Healthcare in Ireland: A Long-Standing Issue The Irish health system has long been characterized by significant underfunding (Turner 2018a, b). The government started to tackle this extreme under-spending about twenty years ago, reflecting the rise in national income. As a result, from the late 1990s until the economic crisis, overall levels of public expenditure rose rapidly, albeit from a very low base.23 Per 23  Public spending grew by 8.4% p.a. from 1995 to 2000 and by 11.2% p.a. from 2000 to 2008 (Lynch 2018). Between 1997 and 2007, the health budget quadrupled, making up for decades of under-spending in health (Burke et al. 2014).

236 

P. MARTIN AND P. TURQUET

capita expenditure, which was below the EU average till the 2000s, is now in line with expenditure in other countries24 (Nolan et  al. 2014). This trend of significantly increased spending reversed in 2009 (Burke et  al. 2014). The economic downturn started in August 2008 but was not reflected in the public finance policy until 2009 because the Irish government adopted countercyclical policies regarding healthcare expenditure between 2008 and 2009. It is for this reason that, despite the pressures on the public budget, healthcare expenditure as a share of gross domestic product (GDP) continued to rise during this period. This countercyclical policy combined with the sharp decline in income resulted in a significant increase in the percentage of the Irish population covered with free medical cards. The uptake of these cards, which enable low-income individuals to get free access to healthcare, rose from 30.1 to 40.3% between 2008 and 2013 (Mazeikaite et al. 2018). However, from 2009 and following the 2010 European bailout, the Irish government started to reduce public spending on health, which had increased significantly in the previous decade (Lynch 2018). In addition to the cost transfer from the state onto the people, the Irish health system experienced continuous cuts to staff and budgets alongside an increasing demand for care from 2009 to 2014. These measures also implemented increases in user charges, including people with medical cards (Burke et al. 2016). Due to this cost-containment policy, co-payments significantly increased,25 thereby reducing the depth of cover. The restructuring process of the hospital sector included a reduction of hospital employment,26 a decrease in salaries for health professionals and reduced payments for healthcare providers. The number of acute hospital beds also continued to decline despite being already well below the EU15 average in 2008 (Burke et al. 2014). Nevertheless, it seems that “Despite the declining budget and 24  In 2017, Ireland devoted 7.1% of its GDP to healthcare, below the EU28 average of 9.6%. However, Ireland was in the 5th place in 2017  in terms of health expenditure per capita, between Austria and the Netherlands and well above the EU average (OECD/ European Union 2018). 25  The drug reimbursement deductible was raised. In-patient, out-patient and emergency charges in hospital were also increased (OECD 2017). The statutory coverage of primary care for wealthy individuals more than 70 years of age was also removed (Mazeikaite et al. 2018). 26  There were over 12,000 fewer Health Service Executive (HSE) staff in December 2013 than there were at the height of public health sector employment in 2007 (Burke et  al. 2014). The total number of whole time equivalent (WTE) staff employed dropped almost 10% between 2009 and 2013 (Department of Health 2018).

8  IRELAND: THE AMBIGUOUS ROLE OF THE HEALTH INSURANCE MARKET 

237

staff numbers, public hospitals managed to do ‘more with less’ in the period 2008 to 2012, indicating increased efficiencies during this time” (Burke et al. 2014). The capacity constraints resulted in longer hospital waiting lists from 2011 to 2015, which clearly lead to healthcare rationing. These waiting times and the growing demand for emergency care admissions indicated system deficiencies by 2013 (Johnston et al. 2019). In 2012, the increase in health expenditure became an area of concern for the European Commission. The short term priority highlighted in the Economic Adjustment Programme for Ireland Winter 2012 Review “was to ensure that the measures envisaged in this year's Budget are fully implemented, especially as regards cost-containing measures in the health sector” (European Commission 2013b, p. 15). In the Autumn 2012 Review, the existence of a €700 million estimated structural gap of the health vote was emphasized. The report pointed out that significant scope remained, including by implementing key reforms announced in the 2012 annual output statement for health expenditures.27 It also suggested that “Government could review the fees and salaries paid to medical professionals (…), increase co-payments for products and services, and tackle the unsustainable growth in medical cards, including through greater use of GP Visit Cards to substitute for more expensive medical cards” (European Commission 2013a, p. 22). Following these recommendations, qualifying criteria for medical cards were tightened in October 2013. Their number, which had peaked in September 2013, dropped significantly by 65,000 in six months (Burke et al. 2014).28 The economy started to recover in 2013 and public spending on health services has been rising in real terms since 2015 (Lynch 2018).

27  Among them, “the introduction of a unique patient identifier, the enhancement of primary care reimbursement services, the expansion of general practitioners’ (GPs) after-hours services, fuller cost recovery of the costs of treating private patients in public hospitals, and the introduction of a financing mechanism based on the ‘money-follows-the-patient (MFTP) (…) The rationalisation of the hospital network can be speeded up and the drugs bill reduced further’” (European Commission 2013a, p. 21). 28  33% of the population had a medical card in December 2017, compared to 40% in 2012. The percentage of people with a medical card has decreased among the older and younger age groups. The decrease among young people could be partly due to the introduction of free GP visit cards for children less than 6 years of age from 2015 (Department of Health 2018).

238 

P. MARTIN AND P. TURQUET

Nevertheless, voluntary health insurance and out-of-pocket payments29 still make up a significant portion of health expenditure (OECD 2017). As previously noted, Ireland has one of the highest ratios of private funding in the EU-15, exceeded only by Spain, Portugal and Greece. The share of public spending dedicated to health accounted for 72% of total health funding in 2016 while that of PHI was 12% (OECD/European Union 2018). The same year, out-of-pocket payments represented 13% of the total funding (OECD/European Union 2018).30 Due to the high level of co-payments and limitations in the scope of coverage, Ireland has the second highest rate of private health insurance spending as a proportion of health financing in the EU28. In terms of healthcare resources, Ireland had 2.9 practising physicians per 1000 population in 2016, compared with an EU average of 3.6. The country had 2.9 hospital beds per 1000 population in 2015, compared with an EU average of 5.1 (OECD/ European Union 2018). Furthermore, the percentage of patients waiting more than three months for cataract surgery was 45 in 2005 and 77 in 2016. This number was respectively 55% and 63% in 2005 and 2016 for hip replacement (OECD/European Union 2018). According to this data, healthcare funding and organization remain critical issues in Ireland despite the increase in public spending in the recent years. Connolly and Wren (2017) used the 2013 Survey of Income and Living Conditions to examine the extent and causes of unmet demands for healthcare services in Ireland and found that factors associated with the healthcare system—cost and waiting lists—accounted for the majority of unmet needs. Johnston and his co-authors (2019) brought up the fact that poorer households, most of whom had medical cards, were disproportionately affected by cost-sharing policies introduced during the economic downturn. This means that the intended safety net of the medical card may no longer play its role. If this were the case, there should be no need to purchase PHI.  These researchers demonstrate that private payments contribute to financial hardship, particularly for people on low incomes. Moreover, despite signs of economic recovery, the proportion of households experiencing unaffordable private health expenditure increased from 15% to 18% between 2009–10 and 2015–16. According to this 29  Out-of-pocket payments are high because co-payments are applied on a broad range of services, including primary care. 30  Due to the economic crisis, Irish health spending coming from private sources grew from 21% in 2008 to 30% in 2015 (Turner 2018a).

8  IRELAND: THE AMBIGUOUS ROLE OF THE HEALTH INSURANCE MARKET 

239

analysis, the largest component of unaffordable spending for poorer households is PHI (Johnston et al. 2019).31 These studies highlight the regressive character of both co-payments and PHI which partly explain the need for a reform to improve access to healthcare in Ireland. 4.2  Toward a Universal Healthcare System? 4.2.1 The 2011 Universal Health Insurance Project Abandonment Confronted with the inadequacies of the health care system, the 2011 newly elected government promised reforms to the health system. Included among the reforms were a commitment to the introduction of free care by general practitioners for everyone by 2015 and the implementation of a universal, single-tier health service through universal insurance in 2016 (Nolan et  al. 2014). The government’s stated objective was to end the unfair two-tier system of access to healthcare, which favors privately insured patients at the hospital level (Turner 2015). Two measures which affected the PHI market were announced in 2013 and quickly implemented; namely the introduction of charges for the use of all beds in public hospitals by privately insured patients and the capping of tax relief on premiums (Turner 2015). As previously mentioned, private health insurance is a decisive factor regarding access to healthcare in Ireland (Turner 2016).32 Private insurers are not integrated with providers; people with private health insurance can be treated in private hospitals and as private patients in public hospitals, whereby they obtain faster access to diagnostics and treatments. Public hospitals and specialist doctors face strong financial incentives to treat privately insured people in public hospitals33 and until recently, insurers were 31  There has been an increase in the number of products on the PHI market, among them contracts including excesses and co-payments and special high co-payments for knee and hip replacements as well as for cataract removal. The latter is considered as an attempt to dissuade older people from purchasing those particular products. Excesses and co-payments are not allowed by law for private treatment in public hospitals (Lynch 2018). 32  Only a small number of policies provide partial reimbursement of certain primary care expenses, such as GP visits and physiotherapy. 33  The different payment methods for public and private patients incentivise “two-tier” access to hospital care. Hospitals are funded by a block grant and consultants receive a salary for treating public patients. For the treatment of private patients, hospitals receive per diem charges and consultants a fee-for-service, providing an incentive for hospitals to treat more private patients whom waiting time is shorter (Connolly and Wren 2017).

240 

P. MARTIN AND P. TURQUET

not charged the full cost of using beds in public hospitals (Turner 2016). Until the end of 2013, 20% of beds in public hospitals were designated as private beds and patients were charged when occupying these beds, whereas the insurers were not charged for the use of public beds by private patients, at the exception of low statutory charges. The new charging structure for privately insured patients being accommodated in public hospitals34 came into effect on January 1, 2014, and led to a significant increase in claims cost for the insurers (Lynch 2018; Turner 2015; Turner 2018b). In addition, the tax relief on PHI premiums was capped in October 2013. As highlighted by Turner, “These measures were designed to reduce the State subsidy of privately insured patients, who tend to be concentrated in higher social classes, thus making such subsidies regressive” (Turner 2015, p. 1355). Following the capping of tax reliefs, PHI premiums rose sharply (Turner 2018a).35 It is for this reason that there was a drop in the number of privately insured individuals with 245,000 fewer people covered in December 2013 than in December 2008 (Burke et al. 2014). As a consequence, PHI coverage, which had peaked in December 2008 at 50.9% of the population, declined to 43.4% in December 2014 (The Health Insurance Authority 2018). As explained in part 1, lifetime community rating was introduced in the Irish PHI market from May 1, 2015 (Lynch 2018; Turner 2015). Such a measure incentivized insurers to innovate and succeeded in attracting people. It partially reversed the decline in the number insured although prior to its introduction, the number of people with PHI had started to stabilize (Turner 2015). There were 2,247,000 people insured with in-patient health insurance plans at the end of June 2019,36 a slight decrease from 34  The significant change was that people who decided to waive their right to be treated publicly (and thus be private patients) would be liable to a minimum daily charge of €813 for each night in most acute public hospitals, irrespective of the designation of the bed that they occupied (€407 for a day case and €1000 for a single room per night). The effect of this change on claims appears to have occurred with a time lag with most of the increasing effect on claims occurring in 2015 and some continuing effect in 2016 (The Health Insurance Authority 2018). 35  The average premium increased by 10% in 2013 and 4% in 2014 (The Health Insurance Authority 2017). 36  In addition to those insured with in-patient plans, there were 106,000 insured with products solely providing out-patient benefits or health insurance cash plans (The Health Insurance Authority 2018).

8  IRELAND: THE AMBIGUOUS ROLE OF THE HEALTH INSURANCE MARKET 

241

the end of 2008.37 The average amount paid for a health insurance premium for in-patient cover in 2018 was €1210. This figure is based on the gross premium level, but child premiums and young adult discounts have a lowering effect on the average figure38 (The Health Insurance Authority 2018). The combined effect of targeted product features and the difference in premiums for different products means that those above the age of 60 paid, on average in 2018, premiums that were 30% higher than the premiums paid by those below the age of 60 for the most popular levels of cover (The Health Insurance Authority 2018). Based on the Dutch case, that is, compulsory PHI with for-profit insurers operating in competition, the universal health insurance was supposed to be launched in 2016 after the introduction of universal free GP care by 2015 (Burke et  al. 2016). It would have been obligatory for people to purchase health insurance, which would have provided hospital and some primary care cover. The state would have paid or subsidized premiums for those on low incomes (Daly 2018). However, following the poor election results in 2014, a new health Minister was appointed (Burke et al. 2016). In 2015, he concluded that the universal health insurance model would be too expensive and not affordable, as an analysis suggested it would increase healthcare expenditure in Ireland by between 3.5% and 10.7% per annum (Wren et al. 2015). The project also encountered opposition from a whole range of interest groups, among which were privately insured persons with faster access to care, insurers fearing the rise in managing costs and medical professionals involved in private care (Daly 2018). Eventually, free GP care was only introduced on an age basis in 2015 (below 6 and above 70 years old) (Burke et al. 2016). 4.2.2

S láintecare: The Imminent End of the Two-tier Healthcare System? In 2016, the new government established a cross-party parliamentary committee to develop a reform agenda for the next ten years. In its final report, known as “The Sláintecare report,” the Committee suggested a radical re-orientation of the Irish health system. The report is comprised 37  In December 2008, the market peaked at almost 2.3 million; it decreased to 2,025,000 million in December 2014 (The Health Insurance Authority 2018). 38  The net premiums are reduced by income tax relief, which is 20% of the gross premium up to a maximum of €200 and which the insurers receive directly from the Revenue Commissioners (The Health Insurance Authority, 2018).

242 

P. MARTIN AND P. TURQUET

of five sections: Population Health Profile, Entitlements and Access, Integrated Care, Funding and Implementation (Houses of the Oireachtas 2017). Recognizing the fact that “the Irish health service as it stands at the moment is not providing the population with fair or equitable medical care” (Houses of the Oireachtas 2017, p. 12), Sláintecare aims to “establish a universal single tier service where patients are treated on the basis of health need rather than on ability to pay” (Houses of the Oireachtas 2017, p. 12), notably by ensuring a clear separation of the public and private elements of the health system. Among the report’s recommendations were the removal or reduction of out-of-pocket charges for accessing healthcare services, the expansion of free-at the-point-of-use GP care to all and the removal of private practice from public hospitals (Turner 2018b). People will be incentivized to access care outside of hospitals as the care should be shifted when appropriate to the lowest level of complexity and from hospitals to the community (Burke et  al. 2018).39 The Committee recommended that clear entitlements to universal healthcare be provided to all, underpinned by legislation and proposes the introduction of a health card scheme, the Cárta Sláinte (health card), which will entitle all residents to access a comprehensive range of services based on need. The shift toward primary and community-based care will therefore involve the expansion of entitlements to primary and social care services, as well as expansion of capacity within the system to deliver better access to primary care and general practice, and to public hospital care as well. Additional negotiations with general practitioners’ representatives will be required to increase capacity of primary care centres. However, this will be a costly process as capacity problems are pretty serious, and “renegotiated capitation rates would need to reflect existing tight market conditions for medical labour supply, as well as the cost to private GPs of the foregone out-of-pocket payments due to a shift to universal coverage” (European Commission 2018, p. 39).40 Better access to public hospitals entails the phased elimination of private care from public hospitals, leading to an improvement in the public 39  “This shift away from the current hospital-centric model will enable our system to better respond to the challenge of chronic disease management, to provide care closer to home for patients, to deliver better value-for-money and to maintain a strong focus on health promotion and public health” (Sláintecare Report 2017, p. 12). 40  The cost of universal GP care is estimated at €455 million over five years. The cost of universal primary care is estimated at €265.6 million over the first five years of the plan (Sláintecare Report 2017).

8  IRELAND: THE AMBIGUOUS ROLE OF THE HEALTH INSURANCE MARKET 

243

system’s ability to provide affordable public care. As highlighted by Turner (2018a, b), the phasing out of private practice in public hospitals will not be straightforward. Untangling the public-private links requires Ireland to confront powerful vested interests groups, including healthcare professionals and private insurers. As already mentioned, many privately insured patients are treated in public hospitals, and many public hospital consultants have contracts that permit them to undertake private practice. Private insurance creates perverse incentives in publicly funded hospitals, where preferential treatment of privately insured patients adds to doctors’ private revenues (European Commission 2019). This is the reason why, when the reforms are implemented, “hospital consultants would lose what they currently earn in private income from treating private patients in public hospitals” (Burke and Thomas 2018). Consultants will have to choose between private and public work contracts, the latter having to be renegotiated. The cost of expanding public activity in public hospitals is estimated at €649 million41 while the cost of increasing numbers of public hospital consultants should be €119 million between Years 4 and 10 (Houses of the Oireachtas 2017). The restriction of privately insured hospital treatment to private hospitals would also reduce the choice of facilities that private health insurers can offer their customers and put additional pressure on the private hospital system (Turner 2018b). Untangling the public-private links could disrupt the private insurance industry and will represent a core challenge for Irish government as assumed by the European Commission in its latest Country report: “it is the Sláintecare vision to ultimately remove private practice from public hospitals, but Ireland has so far proven unable to confront the powerful stakes of the enormous insurance market” (European Commission 2019, p. 41). Sláintecare proposed a phased approach, expanding entitlements over a ten-year period. The additional costs associated with introduction of the single-tier system are estimated at €2.844 million (average expansion each year) over this ten-year period (Houses of the Oireachtas 2017, p. 133). In addition, in order to implement system change, there is a need for one-­ off transitional funding which is estimated at €3 billion (Houses of the Oireachtas 2017, p.  131). According to the European Commission (2018), it is unclear whether the costing exercise underpinning key reform measures is sufficiently robust. However, in the absence of reform, “there 41  The private income currently accruing to public hospitals is to be replaced by public money.

244 

P. MARTIN AND P. TURQUET

are likely to be losses of effectiveness, accessibility, resilience and fiscal sustainability” (European Commission 2018, p. 38) yet, “the costs of not implementing the Sláintecare vision would be much higher in the long run” (European Commission 2018, p.41). The 2019 Sláintecare Action Plan was published in March 2019 and acknowledges 2019 as the first full year of Sláintecare’s implementation. According to Burke (2019), the plan is weaker on universalism than the 2017 Oireachtas report and this was already the case in the Implementation Strategy. The implementation of the Sláintecare vision would undoubtedly be a welcome development for Ireland, and it is desirable that the Irish Government could succeed in implementing the universal single-tier service proposed in the 2017 report. To achieve this, it will have to confront vested interests and also overcome the Irish predisposition toward a “fundamentally libertarian perception of healthcare as a marketable commodity rather than a social right” (Wren and Connolly 2019).

References Biondi, A. (2008). BUPA V.  Commission: Case T-289/03 British United Provident Association Ltd (BUPA) and others v. Commission of the European Communities, Judgment of the 12 of February 2008. European State Aid Law Quarterly, 7(2), 401–407. Burke, S. (2019). Sláintecare Action Plan 2019, HSPM. Retreived from https:// www.hspm.org/countries/ireland18092013/livinghit.aspx?Section=7.2%20 Future%20developments&Type=Section#21Sl%C3%A1intecareActionP lan2019. Burke, S., Barry, S., Siersbaek, R., Johnston, B., Ní Fhallúin, M., & Thomas, S. (2018). Sláintecare—A Ten-year Plan to Achieve Universal Healthcare in Ireland. Health Policy, 122(12), 1278–1282. Burke, S., Normand, C., Barry, S., & Thomas, S. (2016). From Universal Health Insurance to Universal Healthcare? The Shifting Health Policy Landscape in Ireland Since the Economic Crisis. Health Policy, 120, 235–240. Burke, S., & Thomas, S. (2018). The publication of the Sláintecare Implementation Strategy. HSPM.  Retreived from https://www.hspm.org/countries/ireland18092013/livinghit.aspx?Section=8.1%20The%20stated%20objectives%20of%20the%20health%20system&Type=Section Burke, S., Thomas, S., Barry, S., & Keegan, C. (2014). Indicators of Health System Coverage and Activity in Ireland during the Economic Crisis 2008–2014—From ‘More Withless’ to ‘Less With Less’. Health Policy, 117, 275–278.

8  IRELAND: THE AMBIGUOUS ROLE OF THE HEALTH INSURANCE MARKET 

245

Colombo, F., & Tapay, N. (2004). Private Health Insurance in Ireland: A Case Study. OECD. Connolly, S., & Wren, M.-A. (2017). Unmet Healthcare Needs in Ireland: Analysis Using the EU-SILC Survey. Health Policy, 121, 434–441. Daly, M. (2018). ESPN Thematic Report on Inequalities in Access to Healthcare Ireland. European Social Policy Network (ESPN). Department of Health. (2018). Health in Ireland. Key Trends 2018, Government of Ireland. European Commission. (2013a, January). Economic Adjustment Programme for Ireland Autumn 2012 Review. European Economy Occasional Papers, 127. European Commission. (2013b, April). Economic Adjustment Programme for Ireland Winter 2012 Review. European Economy Occasional Papers, 131. European Commission. (2018). Country Report Ireland. European Commission. (2019). Country Report Ireland. Houses of the Oireachtas. (2017, May). Committee on the Future of Healthcare Sláintecare Report. Dublin: Houses of the Oireachtas. Johnston, B.-M., Burke, S., Barry, S., Normand, C., Ní Fhallúin, M., & Thomas, S. (2019). Private Health Expenditure in Ireland: Assessing the Affordability of Private Financing of Health Care. Health Policy. Lynch, B. (2018). Significant Developments in Irish Health Insurance and Healthcare since 1950. In The Irish Healthcare System. An Historical and Comparative Review, A report commissioned by the Health Insurance Authority. Mazeikaite, G., O’Donoghue, C., & Sologon, D. (2018). The Great Recession, Financial Strain and Self-assessed Health in Ireland. The European Journal of Health Economics, 20, 579–596. Nolan, A., Barry, S., Burke, S., & Thomas, S. (2014). The Impact of the Financial Crisis on the Health System and Health in Ireland. European Observatory on Health Systems and Policies. OECD. (2017). Ireland: Country Health Profile 2017. In series: State of Health in the EU, OECD. OECD/European Union. (2018). Health at a Glance: Europe 2018. State of Health in the EU Cycle. Sauter, W. (2008). Risk Equalisation in Health Insurance and the New Standard for Public Service Compensation in the Context of State Aid and Services of General Economic Interest Under UE Law. TILEC, no. 11. The Health Insurance Authority. (2017). Annual Report and Accounts 2017. Retreived from https://www.hia.ie/publication/annual-reports-accounts. The Health Insurance Authority. (2018). Annual Report and Accounts. The HIA. (2018). Guide to Risk Equalisation. Retrieved from https://www.hia. ie/regulation/risk-equalisation

246 

P. MARTIN AND P. TURQUET

Turner, B. (2015). Unwinding the State Subsidisation of Private Health Insurance in Ireland. Health Policy, 119, 1349–1357. Turner, B. (2016). Voluntary Health Insurance in Europe: Country Experience, Ireland. In Sagan, A., & Thomson, S. (Eds.), Observatory Study Series (vol. 42, pp. 77–82). World Health Organization and European Observatory on Health Systems and Policies. Turner, B. (2018a, March). Putting Ireland’s Health Spending into Perspective. The Lancet, 391. Turner, B. (2018b, September). International Benchmarking of the Structure of Irish Healthcare. In The Irish Healthcare System. An Historical and Comparative Review. A report commissioned by the Health Insurance Authority. Wren, M.-A., & Connolly, S. (2019). A European Late Starter: Lessons from the History of Reform in Irish Health Care. Health Economics, Policy and Law, 14(3), 355–373. Wren, M.-A., Connolly, S., & Cunningham, N. (2015). An Examination of the Potential Costs of Universal Health Insurance in Ireland. N°RS45 in Research Series from Economic and Social Research Institute (ESRI).

CHAPTER 9

The Dutch Way: Experimenting with Competition in the Healthcare System Pascale Turquet and Philippe Martin

1   Introduction In the Netherlands, the 2006 comprehensive health reform introduced new mandatory health insurance for curative medicine and removed the distinction between public and private funds, the latter of which were thereafter put into competition. All residents and non-residents paying income tax on their earnings are required to purchase basic health insurance. They are free to choose their own insurer, who is obliged to accept these private individuals under their policy. The 2006 Health Insurance Act (HIA) had four objectives: accessibility, quality, efficiency and affordability of health care.

P. Turquet (*) LiRIS, Université de Rennes 2, Rennes, France e-mail: [email protected] P. Martin Centre for Comparative Labour Law and Social Security, Université de Bordeaux, CNRS, Pessac, France e-mail: [email protected] © The Author(s) 2021 C. Benoît et al. (eds.), Private Health Insurance and the European Union, https://doi.org/10.1007/978-3-030-54355-6_9

247

248 

P. TURQUET AND P. MARTIN

Under HIA, health insurance revenues comprise: –– fixed nominal premiums paid directly by those above 18 years of age company (about EURO 1300 a year (Ministry of Health, Welfare and Sport, 2018));1 –– income-dependent contributions which can be considered as employers’ contributions.2 These are paid into an equalization fund, which is supplemented by a state subsidy (to cover the cost of health care for children under 18 years old). The resources provided by this fund are then divided between insurers according to a risk-adjusted compensation system. In addition, all individuals aged 18 and above also pay a mandatory policy excess (amount of expenses that must be paid out-of-pocket before an insurer will pay any expenses) of EUR 385 (in 2019). Individuals may also choose to voluntarily increase this deductible by a maximum of EUR 500, which causes the nominal premium to decrease. Given that the nominal premium is regressive in character, the tax administration provides a monthly allowance for individuals whose income is below the ceiling defined by the state.3 In 2019, monthly payments can be up to EUR 99 for individuals and EUR 192 for couples, according to their level of income. The Dutch healthcare system is rooted in the Bismarckian tradition, with predominantly public financing and a private delivery of health care (Helderman et al. 2012). In the post-war period, the Netherlands combined both corporatist and etatist policy arrangements (Helderman et al. 2005). The governance was shared among the government, professional organizations and insurers. The balance of responsibilities in the Dutch healthcare sector has shifted repeatedly over time (Kroneman et al. 2016). The state started to implement price and volume control and containment in the late 1970s and 1980s (Helderman et al. 2012). The introduction of regulated competition goes back to the early 1990s and eventually gave rise to the HIA in 2006. With this Act, the Netherlands clearly opted for managed competition. Such competition is regulated by the public  They have increased significantly since 2006.  As of 1 January 2019, the income-related contributions for health insurance under the Health Insurance Act payable by the employer will be increased from 6.90% to 6.95%. The maximum contribution base for the Health Insurance Act is EUR 55,923 as of 1 January 2019. 3  Annual income may not exceed EUR 29.562 for singles or EUR 37.885 for families (2019). 1 2

9  THE DUTCH WAY: EXPERIMENTING WITH COMPETITION… 

249

authorities who are in charge of the contents of the minimum medical care package and prohibit risk selection. Insurers are free to set the value of premiums for each level of cover; however, these premiums will be equal for all policyholders, regardless of their health condition, age or background. In this competition framework, insurers enter into a contractual relationship with care providers in order to contain cost increases (Enthoven 1993). Selective contracting should improve the efficiency and the quality of healthcare delivery. In the Netherlands, the basic health insurance package defined by the government covers medical care provided by general practitioners (GPs), medical specialists and midwives, and pharmaceutical care.4 By purchasing the basic health insurance package, residents of the Netherlands can choose between different policies: a contracted care policy (in kind policy), a non-contracted care policy (restitution policy) or combinations of the two. In addition to the compulsory insurance package, policyholders also have the option of choosing fully private supplementary insurance for care that is not reimbursed by the basic health insurance, that is, dental care, alternative medicine/homoeopathy, glasses and contact lenses and additional cover for physiotherapy. The central government is responsible for the overall healthcare system but has adopted a more distant role of supervisor of the market since 2006. The Dutch Healthcare Authority (NZa) regulates insurers and providers and the Netherlands Authority for Consumers and Markets supervises competition. Managed competition failed to prove effective in curbing healthcare costs after its introduction: total health expenditure per capita increased by 4% in 2006 and 2007 (van de Ven and Schut 2009). Growth has levelled off since 2012 after reverting to a system of sector agreements on spending and shifting costs from public to private sources, for example, by increasing the level of the compulsory deductible, that is, out-of-pocket

4  More specifically: “medical care provided by GPs, medical specialists (consultant physicians) and obstetricians; district nursing; hospitalization; mental health services, including hospital care (mental health-related) up to a maximum of three years; medications; dental care up to age 18 years; services provided by various types of therapists, including physical therapists, remedial therapists, speech therapists and occupational therapists; nutritional/ dietary care; medical aids; ambulance support/sedentary medical transport; physiotherapy for people with chronic illnesses” (Ministry of Health, Welfare and Sport 2018, p. 11). This is about 60% of the healthcare budget.

250 

P. TURQUET AND P. MARTIN

expenses for patients (Jeurissen and van Ginneken 2019).5 According to OECD (2018), the annual average growth rate (in real terms) in health spending per capita was zero between 2013 and 2017. In 2017, the Netherlands ranked sixth among the EU member states both in terms of health expenditure per capita and health spending as a share of gross domestic product (GDP).6 Whether or not this latest trend is wholly or partially attributable to managed competition remains to be determined. The rest of the chapter is organized as follows. Section 2 shows that two logics have always been at work in the Dutch health insurance system, which has been characterized by both public and private interactions since its establishment. Corporatist arrangements dominated during the postwar boom. The Dutch government first became concerned about increases in healthcare costs in the late 1960s. The initial measures to control prices were taken in the mid-1970s. Following the 1986 draft Dekker reform, elements of competition were introduced in the 1990s, giving rise to crucial changes in the health insurance market. These measures made the 2006 reform feasible and indeed palatable (Cohu et  al. 2006). They brought about new actors and the government has since adopted the more distant role of a supervisor and facilitator of the healthcare market (Kroneman et al. 2016). Nevertheless, the implementation of elements of competition is accompanied by an extensive set of public regulations. Section 3 discusses the idea of the congruence of the Dutch legal framework with EU law. The privatization and marketization of the Dutch healthcare system have been supported by a national, legal framework that is formally suitable for the full application of European Economic Law. However, the functioning of the new system demonstrates an innovative way of governing and regulating competition in the healthcare and health insurance markets, combining EU law and national legislation. Finally, it

5  Following the economic crisis, the compulsory deductible increased from EUR 150 in 2008, via EUR 220 in 2012 and EUR 350 in 2013, to EUR 385 yearly in 2016 (Van Esch et al. 2017). Since 2012, cost containment has also focused on: shifting costs between various statutory sources in combination with major cuts in budgets (most notably the current long-term care reform); substitution between different types of care: institutional care with home care, and secondary care with primary care (as visible in mental and long-term care); increased focus on improving efficiency (e.g. tendering of generics) and eliminating fraud (OECD/European Observatory on health systems and policies 2017). 6  The Dutch health spending to GDP ratio was 10.1% in 2017 while the European average was 9.6% the same year (OECD/EU 2018).

9  THE DUTCH WAY: EXPERIMENTING WITH COMPETITION… 

251

can be debated whether EU competition law is adequate in regulating the national health system based on managed competition. Section 4 discusses the effects of competition in a context where accessibility, affordability and solidarity are becoming increasingly important topics. In both the insurance and healthcare markets, competitive conditions can hardly be met, and it seems that the improvement of these competitive conditions is unlikely to eliminate market failures. Tensions have arisen between market and state regulation. Despite the 2006 reform, the state has not reduced its presence. We are going to try to understand why it has increased its role in monitoring, controlling and where necessary, intervening.

2   The 2006 Reform as a Result of a Long-term Process over Decades During the twentieth century, the Dutch healthcare system was characterized by both corporatist and etatist approaches, with complex interdependency between public and private actors such as healthcare providers, insurers, trade unions and employers (Helderman et al. 2005). In order to better understand this model, it is necessary to briefly analyse the various steps in its implementation as well as the reasons put forth in support of reforms. 2.1  Dutch Health Insurance Over the Twentieth Century The first attempt to create a legal health insurance framework was the Kuyper proposal in 1904. As a result of its failure, the Dutch healthcare system remained dominated by private (non-profit and commercial) bodies till 1940. This is the reason why the social health insurance emerged relatively late in the Netherlands compared to other European States. The first Dutch mandatory health insurance scheme for low-wage workers was established by the German occupying Authorities in 1941 (Okma 1997; van de Ven and Schut 2008). These workers had to join officially licensed sickness funds, which were obliged to accept all eligible applicants irrespective of their health status. The Decree issued by German Authorities thus created a mixed public and private system (Vonk and Schut 2018). This social health insurance was reformed in 1964 with the Sickness Fund Act (ZFW). Until 1985, this social health insurance was made up of

252 

P. TURQUET AND P. MARTIN

three schemes: the compulsory scheme, the voluntary scheme (for the self-­ employed with an income under the threshold) and the elderly scheme (for pensioners). These schemes were financed mainly from income-­ related proportional contributions. High earning employees and the selfemployed with an income above the threshold were either uninsured or took out a private plan (Götze 2010). In 1968, the Exceptional Medical Expenses Act (AWBZ) was passed. This law provided mandatory insurance for “high-risk” patients such as the elderly, the handicapped and those suffering from mental disorders. Following the implementation of this law, the Dutch system was usually described as a tripartite system. These three compartments were: –– universal public insurance for high-risk patients (AWBZ), –– “general curative medicine” insurance (ZFW and voluntary private insurance for those whose income was above the threshold set by the law), –– supplementary voluntary private health insurance. In 1986, the government abolished the subsidized voluntary and elderly ZFW schemes for financial reasons, which included a disproportionately high number of persons with unfavourable risks. Their memberships were divided between mandatory social insurance and private insurance according to income. In fact, most of them were moved to the remaining mandatory ZFW scheme (Okma et al. 2010). As a consequence, the government introduced the Joint Funding Act of Elderly Members (MOOZ) in the same year, implementing a risk equalization system between private and social insurance. Private insurers had to pay MOOZ surcharges into a central pool if their share of elderly clients to insure remained lower than average (Cohu et al. 2006; Götze 2010; Thomson and Mossialos 2006). In the same year, the Health Insurance Access Act (WTZ) required the private companies to issue standardized and cross-subsidized health insurance policies for the non-insured (people over the age of 65 years and other specific high-risk groups). The bill forced these health insurance companies to accept every eligible applicant for these WTZ plans (Götze 2010; Vonk and Schut 2018). In 1991, the state extended the eligibility for these plans to every privately insured person paying a higher premium than that of the standard policy. The WTZ goods and services, as well as its tariffs, were set by local authorities (Cohu et al. 2006). As a result, the

9  THE DUTCH WAY: EXPERIMENTING WITH COMPETITION… 

253

share of the standardized insurance policies expanded rapidly. The deficits were retrospectively compensated from a pool that was filled by mandatory surcharges paid by the privately insured (Vonk and Schut 2018). On the one hand, government interventions, based on the WTZ bill, thus instigated a gradual socialization of the private health insurance sector as of 1986. On the other hand, the idea of competition between insurers was introduced in the 1987 Dekker plan which proposed to put in place a unitary system thereby removing the distinction between private and public insurance. This plan recommended allowing patients to freely choose their insurers and setting up care networks managed by insurers. The plan met with opposition and was eventually not implemented. Nevertheless, many elements of the Dekker plan were maintained by the Simmons plan (1990) and the following years saw crucial changes (Turquet 2012), which included: –– the abolition of regional monopolies of the sickness funds in 1992, by allowing consumers a free choice of sickness fund (Helderman et al. 2005);7 –– the introduction of a community-rated premium in the ZFW scheme in 1989. As sickness funds were free to determine their own prices, this was the beginning of price competition;8 –– the introduction of an age-gender risk equalization scheme in 1993, which was subsequently improved by including morbidity parameters (Douven et al. 2006; Götze 2010). 2.2  What Was the Role of Health Insurers? These state interventions went hand in hand with changes in the strategy and practice of the insurers. After World War II (WWII), sickness funds started to establish their own private insurance companies (foundations). These non-profit bodies aimed to attract those that were not eligible for social insurance. Their close links with sickness funds allowed them to achieve significant savings and increase their market share. During this period, commercial insurers were competing with them on the same 7  However, there was relatively little mobility among the insured at that time and insurers failed to make use of the opportunity for selective contracting that was offered to them. 8  However, community rating prevented health insurers from calculating premiums on the basis of individual risk factors and health status.

254 

P. TURQUET AND P. MARTIN

market segment by charging community-rated premiums instead of implementing risk-appropriate pricing. By doing so, insurers clearly contributed to the worsening of their adverse selection problem. In the 1970s, when healthcare expenditure started to increase and economic growth started to slow, they introduced high deductibles. After this attempt failed, they had little choice but to start to introduce age-related premiums in the 1980s. They were eventually joined by the foundations which adopted the same pricing policies (Vonk and Schut 2018). As explained, the situation post WWII was quite complex in the second compartment (Götze 2010) wherein non-profit and profit insurers were competing for those with income above the social insurance ceiling. Due to state regulation which favoured the “socialization” of the private market (Götze 2010) and for fear of further state intervention, commercial insurers started to behave, at least in some ways, like social insurers, thereby exposing themselves to adverse selection. That said, in the face of the growth in health expenditure, they finally chose to modify their pricing models, and even non-profit insurers followed suit. In the 1990s, governmental reforms headed in the opposite direction and led to the “marketization” of social insurance (Götze 2010), thereby inducing gradual change in the practices of the actors. Private health insurers were able to establish their own sickness funds. Development of competition led to a number of mergers between health insurers, including profit and non-profit bodies (Helderman et al. 2012). In any case, the controlled market was not the one in which the commercial private insurers wished to operate. In the mid-1990s, they started to withdraw from it. During the 1990s, sickness funds and commercial insurers were operating more or less on the same terms.9 When the HIA was introduced in 2006, most commercial companies had already left the market and the few remaining sold their health insurance portfolios (Vonk and Schut 2018).

9  During the 1970s and 1980s, only private insurers operating in the second compartment offered sold supplementary coverage in the third one. Since the early 1990s, sickness funds started to offer supplementary health plans on their own. As this was formally prohibited, sickness funds had to cooperate with private insurers (ZBW).

9  THE DUTCH WAY: EXPERIMENTING WITH COMPETITION… 

255

2.3  The Role of State Regulation As will be seen below, competition in the social insurance sector did not bring government control over healthcare provisions and prices to an end. The Dutch government first became concerned about increases in healthcare costs in the late 1960s. In the mid-1970s, the need for cost containment resulted in a government emphasis on measures to control tariffs and quantities. While the share of GDP devoted to health expenditure fell significantly between 1995 and 2000, owing to the very strict nature of the measures adopted, waiting lists also began to appear (Helderman et al. 2012). Such a policy was very unpopular, and the need for reform arose in the early 2000s, especially to relieve these bottlenecks. Several characteristics of the Dutch health system, such as the prior existence of competition in the second compartment, risk equalization system and the presence of the private sector in health care, made this reform possible (Cohu et al. 2006). New mandatory health insurance for routine care was introduced on 1 January 2006. It removed the distinction between mandatory social insurance and voluntary private insurance, as envisaged by the Dekker and Simmons plans. Since that date, all insurers competing in the second compartment are allowed to make profits and pay dividends to their shareholders (van Ginneken et  al. 2008). All residents and non-residents paying income tax must subscribe to health insurance. In addition, the state finances health coverage for those younger than 18 years of age. The main aim of the reform was to promote competition between insurers (Turquet, 2013). Competition is regulated by the public authorities, who define a minimum medical care package and prohibit risk selection. Insurers are free to set the value of premiums for each level of cover, provided that it is the same for all those insured within any given group. In addition, insured people are free to choose their level of deductible. Nevertheless, the new insurance mandate did not replace government planning and control and, according to some researchers, the state has not reduced its presence: “it has extended its role in different ways. Under the responsibility of the Ministry of Health (MoH) and the Ministry of Finance, a new health competition authority (Zorgauthoriteit) monitors the functioning of insurance and healthcare markets” (Okma et al. 2010, p. 25).

256 

P. TURQUET AND P. MARTIN

The Dutch Health Care Authority (NZa) was created in 2006 to supervise the compliance of actors with the Health Insurance Act (Zvw) and the Health Care Market Regulation Act (Wmg). Furthermore, this Authority establishes tariffs on healthcare services—as long as these are not subject to free negotiations—and promotes transparency in terms of price and quality (Kroneman et al. 2016). However, the creation of semi-­independent regulatory agencies does not seem to “imply the end of the deeply rooted practice of formal and informal consultations and negotiations between governmental actors and organised interest groups in Dutch health care” (Helderman et  al. 2012, p.  116). This issue will be addressed in more detail below.

3   The Compatibility of the Dutch Legal Framework with EU Law As revealed in the second section of this chapter, the Dutch political pathway for health care has been market-oriented for years, even before the 2006 reform. This orientation might be seen as a result of the influence of the legal and political background of the EU. However, legally speaking, healthcare policies—just like social policies—remain a national competence of the Member States, and in the Dutch case, there is no evidence of willingness or a decision to comply with European obligations when elaborating the new health insurance scheme. What is to be seen is rather a sort of congruence between the Dutch healthcare policy and the European economic law, as far as privatizing and “marketizing” the Dutch healthcare system made it “suitable” for the application of EU legal concepts and rules, especially internal market law (non-life insurance directives) and competition law.10 In this sense, the “Dutch way” appears to be a rather distinctive one, as it combines and mixes EU law and national legislation through the action of independent authorities in charge of the regulation of this very particular market. Finally, one can wonder if EU competition law is adequate to regulate the national health system based on “managed competition”.

10  EU competition rules are set out in the Treaty on the functioning of the European Union in articles 101 (anticompetitive agreements), 102 (dominance abuse), 106 and 107 (State aid).

9  THE DUTCH WAY: EXPERIMENTING WITH COMPETITION… 

257

3.1  A Legal Framework Formally Suitable for the Full Application of European Economic Law 3.1.1

 ealth Insurers Operating Under Private H and Commercial Law Before the 2006 reform, the statutory health insurance scheme was based on multiple sickness funds (not-for-profit bodies) which covered employees earning less than a certain income level (mandatory system), and private health insurers covering other categories and high-income employees (voluntary system). Hence, insurances that mainly offer and provide the statutory health coverage are either former sickness funds converted into private companies, or private insurers that offered sickness benefits and health coverage in the former system. Since 2006, all health insurers have operated under private law that would allow them to make profits and pay shareholders. However, the market currently seems to be dominated by insurers operating on a not-for-profit basis.11 As private insurers, health insurers are no longer subject to substantial state control. They are required to obtain the authorization of the Dutch Central Bank and must comply with the regulation elaborated by the Dutch Competition Authority: the Authority for Consumers and Markets (ACM). Legally speaking, private health insurers who provide the mandatory health coverage operate in a competitive market. The residents have freedom of choice of their health insurer and receive transparent information about the market as consumers: various websites—including a governmental website—give an overview of the whole offer in health insurances so that consumers can compare policies and prices.12 Health insurance activity as a whole falls within the scope of commercial law, insurance regulation (including solvency rules) and competition law. However, as noted above, due to the fact that health insurers are in charge of the statutory health coverage, public interest rules are imposed upon them that limit their freedom of trade. This is the meaning of the notion of “managed competition” (Maarse et al. 2016). 11  Some of them clustered into an umbrella organization (currently ten health insurers) called Health Insurers Netherlands (Zorgverzekeraars Nederland, ZN) which defines its members as “social entrepreneurs”. 12  Actually, there are not huge differences, as health insurers have to propose and cover the statutory medical care package and are not allowed to apply risk selection when setting premiums. Health insurers differentiate themselves through their supplementary health insurance offer.

258 

P. TURQUET AND P. MARTIN

3.1.2

 ealth Insurances as Undertakings for the Purpose of EU H Competition Law Before the 2006 Act fully privatized statutory health insurance, the Dutch Competition Authority endeavoured to comply with the EU competition law, especially regarding the legal status of private health insurance. In 1998, the Mededingingswet (Competition Act) entered into force and the Netherlands aligned its competition law with EU rules (Van de Gronden and Szyszczak 2014). The sickness funds were initially considered as public bodies and were thus subject to strict State control. However, regarding the European Court of Justice (ECJ) rulings about the notion of economic activity as a criterion for the application of EU competition law,13 the Dutch Competition Authority considered those funds as undertakings in cases that emerged in the late 1990s. At this time, the Dutch Competition Authority “assigned great value to the freedom of the sickness funds to set the level of premiums and to the room of manoeuvre they had on markets for purchasing services and products from health care providers” (Van de Gronden and Szyszczak 2014). Nevertheless, the Dutch Authority changed its interpretation in light of the German public health insurance company AOK ruling by the European Court of Justice in 2004, which dealt with the German sickness funds in charge of statutory health insurance, as the ECJ stated that “the concept of an undertaking in Community competition law does not cover bodies entrusted with the management of statutory health insurance and old-age insurance schemes which pursue an exclusively social objective and do not engage in economic activity”.14 The fluctuations in the Dutch Competition Authority doctrine reveal the willingness to comply with the EU law or rather to be congruent with it. The legal landscape has changed since the 2006 reform. As already noted, the Dutch legislation has established a market-oriented system and the statutory healthcare schemes are no longer managed by sickness funds, but by private insurance companies. According to legal scholars and experts, there is no doubt that the Dutch private insurances currently 13  The Dutch competition authority referred to case law of the Court of Justice of the European Union (CJEU), such as Joined Cases C-159/91 and C-160/91, Poucet and Pistre [1993], and Case C-224/94, FFSA [1995]. About the notion of economic activity according to CJEU case law, see Chap. 5 of this book. 14   ECJ, Joined Cases C-264/01, C-306/01, C-354/01 and C-355/01 AOK Bundesverband and Others versus Ichthyol-Gesellschaft Cordes, Hermani & Co. and Others, 18 March 2004.

9  THE DUTCH WAY: EXPERIMENTING WITH COMPETITION… 

259

operating within the statutory health insurance scheme are undertakings for the purpose of EU competition and antitrust law. Tamara K. Hervey (Hervey 2011) notes: “The Netherlands Competition Authority considers the private insurers providing social healthcare insurance in the Netherlands to be ‘undertakings’, subject to competition law” and refers to other scholars’ opinion about it (Paolucci et al. 2006). Similarly, Van de Gronden and Szyszczak (2014) observe that “they have the freedom to influence the level of benefits the affiliated persons are entitled to,15 and they are for profit. Another argument is that they are not subject to substantial state control”. Furthermore, in the Zorgverzekeringswet state aid case, the European Commission put forward the argument that the Dutch private health insurance companies were undertakings.16 Moreover, and perhaps due to the fluctuant character of the ECJ rulings, the Dutch government added a provision to the Health Insurance Act stating that, irrespective of whether private health insurers might qualify as undertakings under EU law, they must be regarded as such within the meaning of Dutch competition law. In doing so, the Dutch government highlighted the path followed since the 1990s, namely the market orientation of the National Health Insurance. This point is particularly interesting as it shows how a national legislation, such as that of the Netherlands, can be convergent and congruent with EU law whilst also remaining autonomous. 3.2  A Distinctive Way of Governing and Regulating Competition in the Healthcare and Health Insurance Markets The governance of the Dutch healthcare market is based on a mix of EU law, national legislation and the role of national bodies and authorities. In fact, there are three kinds of markets: the healthcare provision market (providers and consumers), the healthcare purchasing market (insurers and providers) and the health insurance market (insurers and consumers). As will be observed further on, the main competition law issues are in relation to the health provision market. Indeed, “the health insurance market 15  One could argue that this freedom is rather theoretical or virtual, because health insurers tend to offer the same basic package of medical care and benefits; selective contracting (with care providers) may however improve the quality and reduce the cost of healthcare for the beneficiaries. 16  Decision of the Commission of 22 December 2005 on the introduction of a risk equalization system in the Dutch Health Insurance, N541/2004 and N542/2004-C (2005) 1329 fin.

260 

P. TURQUET AND P. MARTIN

on the whole seems to function better than the health provision market. Individuals mostly choose their health plan based on easy-to compare price information” (Van Ginneken et al. 2011).17 The new healthcare system is designed to be consumer-oriented: the legislation gives an impetus on consumer choice and insurers have incentives to selectively contract providers. Competition law and policy are supposed to correct market failures, especially those related to market power, that is, situations in which a firm or a group of businesses is able to raise the market prices over marginal costs. The Dutch Competition Act of 1997 (Mededingingswet), which entered into force in 1998, established a new body—the Nederlandse Mededingingsautoriteit (The Netherlands Competition Authority/ NMa)—which was in charge of the enforcement of the new legislation. This authority, which was created “in spontaneous harmonisation with the EU system of competition law” (Sauter 2011), merged with the Dutch authorities for telecommunications and consumer affairs on 1 April 2013 and became the Autoriteit Consument en Markt (Consumer & Markets Authority /ACM). A few years later, the 2006 Health Insurance Act created a sector-specific authority, the Dutch Healthcare Authority (NZa), which is exclusively tasked with the enforcement of the sector-specific competition policy for health care. The NZa is an independent body, although it is funded by the Ministry of Health. Thus, since 2006, there has been a dual policy towards competition enforcement in the healthcare sector: on the one hand, general competition rules enforced by the ACM; and on the other hand, sector-specific competition policy enforced by the NZa. In a way, despite the fact that though these two regulation bodies are supposed to be coordinated,18 they could be described as concurrent powers (Sauter 2011).

17  Nevertheless, these authors observe that there are not much evident outcomes of competition as the service level of the insurer is largely similar with only four major insurers left and the quality of purchased care, which is difficult to assess for individuals, plays a smaller role. Furthermore, 64% of individuals are covered by collective contracts, and since most group contracts are negotiated on the premium level, not on the basis of the quality of the contracted care, quality choices in this market do not yet influence the quality of purchased care. 18  Both collaborate through a mutual protocol.

9  THE DUTCH WAY: EXPERIMENTING WITH COMPETITION… 

261

3.2.1 The Role of the Authority for Consumers and Markets The Authority for Consumers and Markets (ACM) has a general mission to ensure fair competition in all sectors of the economy and to protect the position of consumers. This includes preventing cartels and dominant market positions. The legal regime is EU inspired, based on the distinction between anticompetitive agreements and the abuse of dominance (the ACM is responsible for the enforcement of EU competition law in the Dutch market, based on article 101 and 102 TFEU-Treaty on the Functioning of the European Union). The ACM’s control system is mainly ex-post (except for mergers for which the ACM makes ex-ante reviews), and infringements of the rules on dominance abuse or anticompetitive agreements are sanctioned by fines. The ACM supervises health insurers and healthcare providers as market players. In the healthcare sector, the main current activity of the ACM consists of reviewing hospital mergers. Actually, it seems that before 2004, the Dutch Competition Authority did not review any mergers in the healthcare sector. However, over 100 healthcare mergers have been reviewed since 2004 and none were blocked (Sauter 2011). More recent studies show the same policy by the ACM: hospital mergers are reviewed and usually cleared (OECD 2018, 2019). The first decision by the ACM to block hospital mergers occurred in July 2015. In coming to this decision, the ACM explored the possibilities for patients and health insurers to discipline the merged hospital about prices and quality, but finally found these options were insufficient (OECD 2016). No cases of dominance abuse in the healthcare sector have been dealt with by the ACM; those issues are left for the Healthcare Authority (NZa). As far as anticompetitive agreements are concerned, the ACM imposed sanctions on long-term care providers, local pharmacists, organizations of psychologists and psychotherapists, and even on the Dutch practitioners’ association. However, those decisions are subject to appeal before the Dutch Courts, and some of them were judged unjustified.19 So far, there have not been decisions regarding dominance abuse or anticompetitive agreements in the health insurance market, although an issue with the relationship between insurers and some liberalized 19  For instance, in its judgment of 17 December 2015, the Rotterdam District Court decided that ACM’s fine against the Dutch Practitioners’ Association was unjustified. Although the association had put recommendations on its website concerning the establishment of new clinics, this did not constitute a competition restriction

262 

P. TURQUET AND P. MARTIN

professions (physical therapists and psychotherapists) arose a few years ago. These professions accused health insurers of abusing buying power, as they usually refuse to negotiate with individual practitioners and offer standard contracts on a “take it or leave it basis”. The case was not reviewed by the ACM, but the issue was taken into account in its 2010 guidelines, which focus on the issue of buying power and selling power in the context of vertical agreements (Van Ginneken et  al. 2011). In 2015, the ACM launched a study aiming to gain an insight into the functioning of the health insurance market, and, where possible, to come up with recommendations on how to strengthen competition (ACM 2016). However, this study does not deal with legal issues; it examines the reasons why competition in the health insurance market is relatively ineffective. 3.2.2 The Role of the Dutch Healthcare Authority The NZa supervises the core principles of the Health Insurance Act 2006, and especially the obligation of insurers to accept any applicant and offer proper care, and to respect the prohibition on differentiation of insurance premiums. In addition, it supervises the health insurance and the healthcare market, and it can impose tariff and performance regulations. The NZa also controls mergers in the healthcare sector, thus there is a double merger test (by the NZa and by the ACM). The NZa strives for dynamic competition within the market structure, in particular to guarantee market entry. The kind of control exercised by the NZa is ex ante and the authority can impose specific obligations on individual players who have significant market power (SMP) and can also impose general obligations on all market parties, for instance, by intervening in contract terms. The concept of significant market power originated in the electronic communications sector20 and was then adopted into the healthcare sector. The key element for a finding of SMP is dominance,21 which is slightly different from the legal notion of dominance abuse in EU competition law. Proof of abuse is not necessary for an SMP finding, because it does not mean that any legal rule has been breached. The action of the NZa is more preventive than punitive as it imposes remedies such as the 20  The European Commission has adopted guidelines about SMP in telecoms in 2002, revised in 2018. 21  Must be assessed the market share of the reviewed undertaking and its competitors. The core question is to find if the reviewed undertaking has the ability to determine its behaviour independently from other market participants (customers, suppliers and competitors).

9  THE DUTCH WAY: EXPERIMENTING WITH COMPETITION… 

263

obligation to be more transparent, not to discriminate, or to accept a price regulation.22 In practice, the NZa will give priority to controlling the selling power of healthcare providers over buying power of health insurance companies, and will focus on the exclusion of competitors, rather than the exploitation of consumers and providers. The NZa receives about a hundred complaints relating to SMP every year, but only a portion are registered, and the authority has taken an average of only two decisions per year since 2011. The cases reviewed by the NZa very rarely regard health insurers as being in a position of significant market power. This might be because the NZa considers the significant market power of insurers to be desirable (De Boer and Geilmann 2017). In the Apotheek Breskens case (decision of 18 November 2009), a pharmacy decided to boycott insurers’ preference schemes for cheaper generic drugs by refusing to contract with them, so the pharmacy forced the consumers to pay for their drugs directly and claim partial restitution from their insurers. In this case, the NZa imposed an obligation to deal pending further investigations for the first time. This does not mean that health insurers are exempted from competition law control. For instance, in 2017, the NZa fined a healthcare insurer, De Friesland Zorgverzekeraar, in response to an enforcement request on the ground of breach of the NZa Transparency in the Care Procurement Process Regulations. According to some scholars, the twofold mission of the NZa—simultaneously enforcing competition policy and making sure that private insurers comply with public interest rules—created unresolved tensions between a universalist model of health services with equal access and equal treatment, and a market-driven model emphasizing efficiency, innovation and patient choice (Prosser 2005). 3.2.3 Different Approaches to Competition Law There are some common factors in the legal approach of both authorities. Consequences on the healthcare and health insurance markets are considered more significant than formal (or “per se”) restrictions, which is consistent with the EU Commission approach of applying the provisions of the Treaties on competition law. However, it seems that there are more differences than similarities between the approaches of the ACM and the NZa to competition law: the ACM is “prohibition based” and aims to police functioning markets. It derives part of its power and most of its  Punitive sanctions such as fines can be ordered when the obligations are not met.

22

264 

P. TURQUET AND P. MARTIN

principles from the EU. The NZa’s action is purely based on national law and regulation and is more subject to political pressures. It is more concerned with creating markets, which is coherent with its ex ante control (Sauter 2011). As noted above, the system of sanction also differs: fines versus price constraints. As regards the interpretation of competition concepts (dominance, foreclosure, etc.), the ACM apparently has precedence over the NZa. However, the Healthcare Market regulation Act 2006 provides for the priority of the sector-specific over the general competition rules (the legal principle specialia generalibus derogant may apply here). Nevertheless, according to Van de Gronden and Szyszczak (2014), the NZa “has not been able to take the lead in matters concerning competition and health care. This finding should be explained by the role of the ACM which was at that time already established in the Dutch health care sector”. Incidentally, in 2016, a legislative proposal regarding the transfer of tasks—notably merger tests and the SMP instrument23—by the Nza to the ACM was presented to the Lower House of the Dutch Parliament. This proposal was not adopted. 3.3  Is EU Competition Law Adequate to Regulate the National Health System Based on “Managed Competition”? As previously highlighted, the Dutch system is based on managed competition, which means that the sponsor—the government—establishes rules of equity, selects participating plans, manages the enrolment process, creates price-elastic demand and manages risk selection (Enthoven 1993). In the Netherlands, private insurers in charge of statutory health insurance operate under public interest legislation that guarantees open enrolment, lifetime cover, government-set income-based contributions deducted at source, additional community-rated premiums, a package of minimum benefits defined by the government and a risk equalization scheme. It should be noted that before the 2006 reform was passed, the Dutch government asked the European Commission to clarify whether or not art. 54 of the third non-life insurance directive which regards the possible

23  It was proposed to impose new SMP measures on healthcare providers only, and not on health insurers. This proposal was not favoured by the business community and was criticized by Maverick, a leading competition law firm in the Netherlands.

9  THE DUTCH WAY: EXPERIMENTING WITH COMPETITION… 

265

existence of substitutive private coverage24 could be relied on to justify the extensive regulation of private health insurance. The Commission answered (through the so-called “Bolkestein letter” in 2003) that the projected reform of the Dutch healthcare system showed that there was room for intervention at the national level. It stated that the Dutch regime characterized by open enrolment, mandatory coverage, uniform rates (at a level freely determined by the insurer) and a risk equalization fund could be justified under article 54 of the third non-life insurance directive as these principles appeared necessary to ensure the legitimate objective of the Dutch government (Thomson and Mossialos 2010). Nevertheless, the combination of market forces and public intervention questions the adequacy of EU competition law to govern and regulate this kind of arrangement. In this discussion, it has to be borne in mind that EU competition law is double sided. On the one side, the general prohibition of monopoles, dominance abuse, cartels and state aid is intended to enhance market efficiency. Technically, it supposes that the competition authorities define the market they are reviewing, in order to distinguish between the different areas in which companies compete with each other. But it is not that easy to cope with the usual “market definition” as regards the healthcare sector which is actually very specific. Care is not simply a product, and the standard legal tests in order to assess the effects of monopolistic situations do not comply with the healthcare sector, especially because of the prevalence of the “third party pays” system: people are not very sensitive to price changes because of the presence of health insurance. On the other side, EU competition law contains exceptions that would allow national legislation to regulate the market and, to some extent, to limit the scope of competition. The most important is the Service of General Economic Interest (SGEI) exception expressed by article 106(2) TFEU,25 but also defined within the Dutch Competition Act 24  Article 54 of the third non-life insurance Directive states that a Member State in which insurance contracts “may serve as a partial or complete alternative to health cover provided by the statutory social security system may require that those contracts comply with the specific legal provisions adopted by that Member State to protect the general good in that class of insurance, and that the general and special conditions of that insurance be communicated to the competent authorities of that Member State before use”. 25  The notion of SGEI was introduced in the European Treaties, at article 86 TEC (Treaty Establishing the European Community) that became article 106§2 TFUE, which states that “Undertakings entrusted with the operation of services of general economic interest or having the character of a revenue-producing monopoly shall be subject to the rules contained in

266 

P. TURQUET AND P. MARTIN

(e.g. article 11 Mw). However, it appears that national competition authorities find it difficult to appropriate the SGEI exception, especially concerning the functioning of the risk equalization scheme. Before the European Court of First Instance ruling in the BUPA case (2008),26 the ACM doctrine was rather “competition-oriented”. In its guidelines, the ACM stresses that the SGEI exception, which is capable of justifying competition restrictions, should be interpreted narrowly (Van de Gronden and Szyszczak 2014). This looks rather paradoxical, as the European Commission itself seemed more “SGEI-oriented” towards the Dutch healthcare reform: in 2005, the European Commission decided that the Dutch basic insurance could fall within the scope of the SGEI exception. Moreover, the Dutch risk equalization scheme was considered as a proportional restriction of competition in order to guarantee the basic insurance, even though there was no explicit indication of the SGEI exception.27 Despite (or because of) this friendly approach to regulated competition by the European Commission, there still are domestic debates in the Netherlands about the risk equalization scheme. Sauter’s commentary on the EU Commission’s decision is rather critical: he observes that the Dutch risk equalization scheme functions with both ex ante and ex post compensation. In his opinion, “ex ante compensation is desirable in order to retain incentives for efficiency and any ex post compensation should be limited to the necessary minimum”. Here instead, ex post compensation was seen by the European Commission as part of a system of double solidarity: among the insured population and persons with various income levels (Sauter 2011). The proper Dutch Ministry of Health seems to agree the Treaties, in particular to the rules on competition, in so far as the application of such rules does not obstruct the performance, in law or in fact, of the particular tasks assigned to them. The development of trade must not be affected to such an extent as would be contrary to the interests of the Union”. 26  Court of First Instance, Case T-289/03, British United Provident Association Ltd (BUPA), BUPA Insurance Ltd and BUPA Ireland Ltd v Commission of the European Communities, 12 February 2008. In this case, the Court set a new standard for assessing whether public compensation—through a risk equalization system between providers of private health insurance in Ireland—constitutes a state aid or is covered by the SGEI exception. The BUPA ruling is notably less rigid than the standards set in Altmark (24 July 2003, Case C 280-00) and gives room for public intervention in the health insurance market. 27  Decision of the Commission of 22 December 2005 on the introduction of a risk equalization system in the Dutch Health Insurance, N541/2004 and N542/2004 -C (2005) 1329 fin.

9  THE DUTCH WAY: EXPERIMENTING WITH COMPETITION… 

267

that ex post compensation (refund of costs) is not adequate as it can compromise the incentive structure of the scheme which is supposed to promote competition on its merits, that is, efficiency and quality.28 Moreover, the Azivo case brought before the EU Commission in 2006 shows that some actors are unsatisfied with the risk equalization scheme and with the Commission’s decision to approve it in 2005. The claimant, an independent insurance company with around 150,000 policyholders requiring more extensive health care than that of the average person insured in the Netherlands, complained about shortcomings in the equalization system: the compensation provided for some healthcare insurers is higher than necessary to cover the costs of meeting their public service obligation, whereas the position for a number of other healthcare insurers is that they are inadequately compensated. According to the claimant, the Commission committed errors of appraisal regarding the operation of the equalization system and did not investigate the matter adequately. The claimant stated that the decision is “incomprehensible, or at the very least inadequately reasoned”. Unfortunately, the Azivo case was withdrawn from the register of the Court29 so there has not been a final judgment on the accuracy of the Commission’s approach. It must be taken into account that Azivo then referred to the criteria for SGEI set in Altmark decision. Finally, with the BUPA ruling, it can be concluded that the Member States gained more control of the regulation and financing of SGEI, especially in the health insurance market. As observed by Van de Gronden and Szyszczak (2014) the Dutch case shows that a new form of Euro-national competition law is emerging for the healthcare sector with national authorities taking the lead in shaping the contours of this law. This can be explained by the relative lack of coherence of the EU competition law as enforced by the EU Commission and the ECJ (a case-by-case approach). Actually, very few cases regarding the players in the national healthcare systems have been dealt with by the ECJ; national bodies, authorities and governments therefore do not have much legal material to which they can refer. It remains the role of governments to build a legal framework that effectively balances public interest regarding health care (this includes social and demographic issues) and the business interests of market players.  Ministry of Health, Welfare and Sport Risk Insurance under the Health Insurance Act in the Netherlands, Summary of 7 July 2008. 29  Order of the Court of First Instance of 13 October 2008, Azivo Algemeen Ziekenfonds De Volharding v Commission (Case T-84/06) (2008/C 327/73). 28

268 

P. TURQUET AND P. MARTIN

4   Market Efficiency: To What Extent? On Competition-related Issues The Dutch health system can be characterized as a system of managed competition. In such a context, consumers freely choose among health insurers and health insurers are supposed to contract selectively with healthcare providers. The state regulates competition within both insurer and provider markets to ensure universal access to high-quality, affordable care (Krabbe-Alkemade et al. 2017). Essentially, health insurers compete for subscribers whilst healthcare providers compete for contracts with health insurers. These processes may lead to greater efficiency by lowering prices and increasing quality of care. In the case of the Netherlands, such results remain unachieved as of yet. Competition failures are still numerous both on the insurance market and on the healthcare providers market. 4.1  The Health Insurance Market Far from being atomistic, the health insurance market is rather concentrated.30 According to the most recent survey by KPMG (2018), 9 health insurance groups were active in 2017, consisting of 24 insurers in total.31 However, 90% of health insurance was in the hands of just 4 large players in 2017. These top four health insurance groups consist of multiple insurers, and a single insurer within a group can be responsible for several insurance brands. Most insurers stem from the former sickness funds founded by medical associations, local communities and labour unions. As a result, insurers are still not-for-profit and organized as a cooperative (Stolper et al. 2019). This is not the case of the Swiss IptiQ, which entered the Dutch health insurance market in 2018. The first new entrant since 2006, it has gone on to acquire three existing brands. Despite the low number of insurance groups, numerous basic health insurance packages are available on the market. Due to this lack of transparency, it is difficult for consumers to compare different options, thus making it difficult to choose. A study conducted by the Netherlands Authority for Consumers and Markets (ACM) and the Dutch Healthcare  About market concentration in the health insurance sector, see Chap. 4 of this book.  According to the Dutch Health Care Authority, they were 118 en 1990, 33 in 2006. Well before 2006, the healthcare sector anticipated the reform with scale enlargement. Private insurers and sickness funds merged into large companies to strengthen their competitive position. 30 31

9  THE DUTCH WAY: EXPERIMENTING WITH COMPETITION… 

269

Authority (NZa) into the differences between basic health insurance packages in 2018 showed that approximately 9.8 million insured individuals have basic health insurance packages for which a cheaper alternative exists that is almost identical or very similar to their current package. According to this analysis, several insurance groups offer policies that are presented as different, but which, are the same or very similar. Overemphasizing such small differences makes it difficult for the insured to choose wisely (ACM 2018a). This artificial differentiation limits consumers’ ability to make informed choices and does not encourage them to switch insurers. The relatively weak switching rates on the basic health insurance market only confirm this. The 2006 reform resulted in a very high switching rate of 18% whereas during the 2007–2015 period, switching rates returned to lower levels, between 4% and 8% per year (Douven et al. 2017). This fact is consistent with the conclusions of the 2018 Health Care Monitor (ACM 2018b):32 Dutch consumers are unaware of the opportunities they have when it comes to selecting the health insurance that best meets their needs. More than half of all consumers do not seek out other health insurances and the majority of consumers have no idea how much they could save each year by switching to a similar health insurance with another insurer (ACM 2018a). It is easily understood why competition between health insurers is suboptimal according to the ACM. One of the reasons for this conclusion is the existence of obstacles to entry and expansion (ACM 2016). According to a 2017 ACM study, the most significant obstacles for new entrants are the following (ACM 2017a): –– difficulties in meeting the European capital requirement (Solvency II);33 –– problems in obtaining a license from the Dutch Central Bank; –– regulatory uncertainty experienced by market participants. Although the Dutch health insurance market can be regarded as an oligopoly, it does not generate excess earnings for the insurers, whose gross profit margins were negative in both 2016 and 2017 (KPMG 2018). 32  The Health Care Monitor is an annual survey among a representative sample of consumers into the health insurance market. 33  See Chap. 3 of this book.

270 

P. TURQUET AND P. MARTIN

In spite of the fact that an increase in premium income of health insurance companies in these two years can be observed, these companies are indeed trying to reduce this increase to keep the policies affordable (KPMG 2018). They do this by giving back their equity to the insurers and selling basic health insurance policies below cost price (KPMG 2018). But how long will these companies be able34 and willing to avoid a major increase in the premiums in the long run? To date, some insurers still appear to think that they have a social mission to fill, or a kind of a “moral obligation to act upon the public goals of the system” (Stolper et al. 2019, p. 297).35 A possible explanation for this may be their non-profit nature. Nonetheless, they may adapt their behaviour in the future as underlined by these authors: what will happen if new CEOs, with different perspectives on the insurer’s mission, take over the wheel and/or insurers would become more profit-oriented? This question is particularly interesting given that in 2018 the first foreign, for-profit insurer entered the Dutch market. (Stolper et al. 2019, p. 298)

Managed competition is also suboptimal if insurers are incentivized to select their clients to protect themselves against adverse selection. This is the reason why risk selection is one of the first issues to be resolved when implementing a competition-based health insurance. A compensation system has to be implemented to discourage risk selection. However, due to imperfections in the risk equalization systems, groups of high-risk consumers (such as chronically ill people) are substantially undercompensated, resulting in incentives for risk selection in all countries (van de Ven et al. 2017). The Dutch risk equalization system includes about 160 risk-classes, half of which are based on diagnoses or prior utilization related to chronic illness. Although it is reputed as the most sophisticated in the world, it does not always succeed in compensating insurers sufficiently for the predictably high costs of relatively small groups of individuals (Eijkenaar and van Vliet 2017). As insurers are undercompensated for these individuals, there 34  The Solvency II directive, implemented in 2016, requires them to have higher reserves. However, the average Solvency II ratio for health insurers was 188% in 2017 relative to 271% in 2016 (KPMG 2018). 35  They conducted interviews with CEOs and organized separate focus groups with purchasers and marketers of five Dutch health insurers to examine the incentives insurers face for enhancing quality of care.

9  THE DUTCH WAY: EXPERIMENTING WITH COMPETITION… 

271

is room for risk selection. That is the reason why, according to Eijkenaar and van Vliet (2017), particularly for individuals with rare diseases, improving risk equalization remains “challenging and dependent on the outcomes of the ongoing societal debate regarding coverage and reimbursement of expensive drugs” (Eijkenaar and van Vliet 2017, p. 1175). In the basic health insurance, selection is prohibited but can be circumvented. In this regard, supplementary insurance can be an effective tool for risk selection as insurers commonly offer basic insurance and supplementary insurance as a joint product (van de Ven et al. 2017). The supplementary insurance market is an unregulated market without risk equalization and accounts for 10% of the total health insurance market (KPMG 2018). On this supplementary insurance market, insurers are able to apply risk-rating and selective underwriting which, to date, remains undone so as to protect their reputation (van Winssen et  al. 2017). Similarly, it seems that only a minority of health insurers use health questionnaires when people apply for supplementary coverage. WillemseDuijmelinck et  al. (2017) have shown that the percentage of insurers using at least one tool to protect themselves against adverse selection, that is, selective underwriting, risk-rating and/or product differentiation, even decreased between 2009 and 2015. Nevertheless, insurers are able to use less visible instruments to protect themselves against adverse selection. According to Willemse-Duijmelinck et al. (2017), insurers may promote their product to targeted consumer groups by using selective marketing, increasing the premiums of high-risk preferred products, targeting groups insurance36 and so forth. That is why the government has frequently been asked to regulate supplementary health insurance; however, they judged it to be infeasible in light of the European regulation. That said this less visible selection is consistent with the fact that the number of consumers without supplementary health insurance rose from 7% in 2006 to 16% in 2016. These figures can be expected to increase sharply in the years to come (see van Winssen et al. 2017). 36  About two-thirds of the Dutch population are covered by a group health insurance plan (55% en 2006). Such groups can be organized by any legal entity (e.g. employers, shops, sports clubs, patient organizations and private initiatives). Three-quarters of those with collective insurance are insured through their employer. Whereas insurers have to respect open enrolment, groups are free to reject applicants (Groenewegen and de Jong 2007; van de Ven et al. 2017). These enrollees benefit from a 10% rebate. Allowing such a discount encourages risk selection.

272 

P. TURQUET AND P. MARTIN

A potential for self-selection also exists. A few years ago, Roos and Schut found that an increasing proportion of high-risk individuals “believe that insurers would not be willing to offer them another supplementary insurance contract” (Roos and Schut 2012, p. 51). High-risk individuals perceive supplementary insurance as a switching cost, believing they are “locked-in their current supplementary insurance contract” (Roos and Schut 2012, p. 60). Such a self-selection limits their freedom of choice on the basic health insurance market and therefore reduces the effectiveness of managed competition. Insurers are not encouraged to lower the premium or improve the quality of care in response to preferences of the high-risk consumers. In addition, as the propensity to switch insurers decreases with age and increases with education and health status, this “may provide insurers with incentives to focus on accommodating the preferences of young, well-educated and healthy people” (Boonen et  al. 2016, p. 349). The Health Care Monitor 2019 (ACM 2019) confirms that higher-risk individuals are considerably less likely to switch health insurers than individuals who indicate they do not need any health care. It shows that people with a chronic disease, a condition or a lengthy recovery period are much more worried about being refused by another health insurer (18%) or that they will encounter hidden costs (14%). Among the individuals who claim not to need any health care, those percentages are respectively 2% and 8% (ACM 2019). 4.2  The Healthcare Providers Market and the Difficulties of Selective Contracting In a managed competition system, health insurers negotiate terms relating to the delivery, price, quality and volume of service with healthcare providers (Croes et al. 2018). In the Dutch system, insurers are allowed to contract selectively with providers and, in doing so, form provider networks. To incentivize hospitals to control cost, improve quality, transparency and to facilitate contractual negotiations at a product level as well, a competitive prospective payment system called the Diagnosis Treatment Combination (DTC) system was introduced in 2000. From then on, market competition has been implemented incrementally: the average share of freely negotiable DTCs was gradually expanded, from about 10% of hospital revenue in 2005 to 70% in 2012. For the remaining 30% (consisting

9  THE DUTCH WAY: EXPERIMENTING WITH COMPETITION… 

273

of the most complex hospital products), prices are still regulated (Douven et al. 2019; Krabbe-Alkemade et al. 2017). To date, selective contracting remains relatively uncommon. There are several reasons for this, one being that hospital markets have become highly concentrated: the number of general hospitals steadily declined from 101 in 2000 to 90 in 2004 and 81 in 2015 (Schut and Varkevisser 2017). This had been an ongoing process since the 1960s, which intensified as a result of the marketing strategy during the 2000s (Kroneman et al. 2016). An ACM study into the effects of hospital mergers on prices and volumes has shown that merger leads to increased prices. It had previously been revealed already that healthcare quality does not significantly improve after mergers (ACM 2017b, c). State regulation may also reduce incentives for competition and selective contracting. In 2011 the government introduced a macro budget instrument to ensure that annual hospital expenses do not exceed the available budget. Any budget overrun is compensated by a levy on the hospital. This threat affects contractual agreements between insurers and providers. Following this, fixed or maximum lump sum payments were introduced, which dampened competition (Douven et al. 2019). In addition, insurers are only allowed to limit the reimbursement of non-contracted care to such an extent that it is still affordable for patients to visit any provider they desire. Indeed, the Health Insurance Act stipulates that insurers must pay a reimbursement when patients make use of a non-contracted provider. The Dutch Supreme Court ruled that this reimbursement may not be so low as to act as a deterrent for patients to use this provider, therefore weakening the selective contracting instrument (Stolper et al. 2019). Although health insurers can selectively contract with GPs as well, they have little incentive to do so. Owing to the level of organization which GPs enjoy via their powerful national association, they are able to remain independent from the insurers. They manage to benefit from the long-­ standing trust relationship with their patients and so succeed in mobilizing public support against unfavourable contracting practices of health insurers and enforcement of the competition act (Schut and Varkevisser 2017). Another reason for the relative lack of success of selective contracting is the reluctance of insured persons to renounce their freedom of choice. Patients also wonder whether insurers with restrictive networks are really committed to the provision of good quality care. Indeed, quality of care

274 

P. TURQUET AND P. MARTIN

remains ill-defined and there is no evidence that selective contracting is able to improve it: A major challenge in any healthcare system is to provide third party purchasers with the right incentives to steer on quality of care (…). Currently, the incentives for price competition seem to be much stronger than those for quality competition. (Stolper et al. 2019, p. 298)

For example, in 2009, a large regional health insurer designated preferred providers for two different medical procedures. As permitted by law, patients were exempted from paying their deductible when they went to a preferred provider (Van der Geest and Varkevisser 2016). The same year, this insurer decided to stop using the differential deductible to channel patients, as a majority of those enrolled reacted negatively towards the differential deductible, considering it to be an infringement of their freedom to choose their own provider. In the end, the insurer admitted the absence of reliable quality indicators (Van der Geest and Varkevisser 2016). As discussed above, the conditions for exercising free competition do not appear to be met in the Netherlands. The trend towards concentration, between insurers and hospitals respectively, hinders negotiation and reflects the struggle to strengthen their position and obtain market power between insurers and providers. The option to switch insurers is rarely used by consumers due to fear and lack of information, with the exception of collectives representing large groups of insured persons. As highlighted by Maarse et al. (2016), the imperfect structure of risk equalization, the market segmentation and the zero-sum relation between individual and collective supplementary schemes also contribute to a rise in concern “on the consequences of the reform for solidarity” (Maarse et  al. 2016, p.  5). Furthermore, quality of care does not appear to be the main concern in the negotiations, the focus of which is dominated by price and volume. Patient’s influence on quality is still rather weak (Kroneman et al. 2016). The key assumption underpinning the managed competition model is that insurers are the customers’ representatives. In fact, there are various indications of a trust problem in the Dutch health insurance market (Maarse and Jeurissen 2019). Groenewegen et al. (2019) addressed the issue and found strong and mutual trust relationships between patients and healthcare providers. However, this is not the case between healthcare providers and insurers, nor between insured and insurance companies. Weak trust relationships in both cases have been revealed by the same

9  THE DUTCH WAY: EXPERIMENTING WITH COMPETITION… 

275

research. If people insurers are perceived as money-driven agents, this restricts their intended role as selective purchasers of health care (Groenewegen et al. 2019). This lack of legitimacy of the health insurers vis-à-vis the customers and the providers is one of the reasons that selective agreement has not developed. This topic is now becoming an important issue and several recent research studies focus on trust in the healthcare system.

5   Conclusion Competition in the health sector does not appear to receive a lot of societal support. Opposition is growing against competition as the driving force of the health market. Health was one of the most contentious issues in the 2017 election campaign and several political parties even ran on a platform of reducing market mechanisms in health care.37 Accessibility, affordability and solidarity are difficult to meet in a competitive private market. This is why tensions have arisen between the competitive market and need for control. Some experts believe that conditions of competition should only be improved to mitigate market failures due to adverse selection or concentration. However, this is highly debatable. Nevertheless, the health insurance mandate, that is, the fact that for-profit companies are allowed to compete for minimum coverage insurance plans has not replaced government planning and control to date (Okma and Crivelli, 2013). The multiplicity of regulations to safeguard accessibility38 and contain costs, the sectoral agreements between stakeholders and the Minister of Health39 and the ruling of the Supreme Court to ensure that the reimbursement obligation be upheld all demonstrate the necessity for the state to guarantee and protect collective interest. Following decades of reforms, the Dutch government seems to be rediscovering the need for consensus-­ building and, as a result, the “traditional Dutch way” of dealing with societal problems. 37  The Socialist party proposed replacing the current array of around 30 insurers with a single national scheme for the universal basic package. 38  In 2008, a hospital was close to bankruptcy and the NZa decided to give it financial support. The Ministry of Health was heavily involved as he said he could be held accountable for the continuity of hospital care in the region (Maarse and Paulus 2011). 39  Deemed necessary to curb the increase in healthcare expenditure in the early 2010s (see above).

276 

P. TURQUET AND P. MARTIN

References Authority for Consumers and Markets (ACM). (2016). Competition in the Dutch Health Insurance Market. Interim Report. Retrieved March 18, 2018, from https://www.acm.nl/sites/default/files/old_publication/publicaties/16129_competition-in-the-dutch-health-insurance-market.pdf. Authority for Consumers and Markets (ACM) (2017a). Barriers to Entry and Expansion in the Dutch Health Insurance Market. Public Report. Retrieved July 18, 2019, from https://www.acm.nl/sites/default/files/old_publication/ publicaties/16984_executive-summary-barriers-to-entry-and-expansion-inthe-dutch-health-insurance-market.pdf. Authority for Consumers and Markets (ACM). (2017b). Intensified Attention for Anticompetitive Risks of Hospital Mergers. Explanatory Notes. Retrieved July 30, 2019, from https://www.acm.nl/sites/default/files/documents/ 2018-01/explanatory-notes-merger-control-hospitals_1.pdf. Authority for Consumers and Markets (ACM). (2017c). Price and Volume Effects of Hospital Mergers. Investigation into Effects of Hospital Mergers 2007–2014. Public Report. Retrieved July 27, 2019, from https://www.acm.nl/sites/ default/files/documents/2018-01/report-price-and-volume-effects-of-hospital-mergers.pdf Authority for Consumers and Markets (ACM). (2018a). Choosing Health Insurance: More Transparency Alone is Not Enough. News. Retreived July 27, 2019, from https://www.acm.nl/en/publications/choosing-healthinsurance-more-transparency-alone-not-enough. Authority for Consumers and Markets (ACM). (2018b). Zorgmonitor 2018: Minder verschil in basispolissen waargenomen. Consumentenonderzoek zorgverzekeringsmarkt. Retreived July 21, 2019, from https://www.acm.nl/sites/ default/files/documents/2018-05/rapport-acm-zorgmonitor-2018.pdf Authority for Consumers and Markets (ACM). (2019). Zorgmonitor 2019: Onderzoek naar overstappen bij zorgverzekeringen. Retreived July 21, 2019, from https://www.acm.nl/sites/default/files/documents/2019-05/zorgmonitor-2019.pdf. Boonen, L.  H. H.  M., Laske-Aldershof, T., & Schut, F.  T. (2016). Switching Health Insurers: The Role of Price, Quality and Consumer Information Search. European Journal of Health Economics, 17, 339–353. Cohu, S., Lequet-Slama, D., & Volovitch, P. (2006). Pays-Bas, une réforme du système de santé fondée sur la concurrence et la privatisation. Revue française des affaires sociales, 2-3, 217–238. Croes, R. R., Krabbe-Alkemade, Y. J. F. M., & Mikkers, M. C. (2018). Competition and Quality Indicators in the Health Care Sector: Empirical Evidence from the Dutch Hospital Sector. European Journal of Health Economics, 18, 5–19.

9  THE DUTCH WAY: EXPERIMENTING WITH COMPETITION… 

277

De Boer, C. J., & Geilmann, M. (2017). The Netherlands. In P. Kobel, P. Këllezi, & B.  Kilpatrick (Eds.), Antitrust in Pharmaceutical Markets and Geographic Rules of Origin (pp. 263–288). Cham: LIDC Springer. Douven, R., Burger, M., & Schut, F. (2019). Does Managed Competition Constrain Hospitals’ Contract Prices? Evidence from the Netherlands. Health Economics, Policy and Law. https://doi.org/10.1017/S1744133119000215. Douven, R., Katona, K., Schut, F. T., & Shest, V. (2017). Switching Gains and Health Plan Price Elasticities: 20 Years of Managed Competition Reforms in The Netherlands. European Journal of Health Economics, 18(8), 1047–1064. https://doi.org/10.1007/s10198-017-0876-8. Douven, R., Mot, E., & Pomp, M. (2006). Health Care Reform in the Netherlands. The Hague: CPB Netherlands Bureau for Economic Policy Analysis. Eijkenaar, F., & van Vliet, R. C. J. A. (2017). Improving Risk Equalization for Individuals with Persistently Highcosts: Experiences from the Netherlands. Health Policy, 121, 1169–1176. Enthoven, A. C. (1993). The History and Principles of Managed Competition. Health Affairs, 12(Suppl. 1), 24–48. Götze, R. (2010). The Changing Role of the State in the Dutch Healthcare System, TranState Working Papers 141. Resource Document. Retreived July 14, 2019, from https://www.econstor.eu/handle/10419/43883. Groenewegen, P. P., & de Jong, J. D. (2007). Dutch Health Insurance Reform: The New Role of Collectives. Eurohealth, 13(2), 10–13. Groenewegen, P. P., Hansen, J., & de Jong, J. D. (2019). Trust in Times of Health Reform. Health Policy, 123, 281–287. Helderman, J. K., Bevan, G., & France, G. (2012). The Rise of the Regulatory State in Health Care: A Comparative Analysis of the Netherlands, England and Italy. Health Economics, Policy and Law, 7(01), 103–124. Helderman, J. K., Schut, T. F., van der Grinten, T. E. D., & van de Ven, W. P. M. M. (2005). Market-Oriented Health Care Reforms and Policy Learning in the Netherlands. Journal of Health Politics, Policy and Law, 30(1–2), 189–209. Hervey, T. K. (2011). If Only It Were So Simple: Public Health Services and EU Law. In M.  Cremona (Ed.), Market Integration and Public services in the European Union. New  York: Oxford, Collected Courses of the Academy of European Law, XIX/2, 179-310Collected Courses of the Academy of European Law. Retreived January 14, 2019, from http://hdl.handle.net/1814/17621. Jeurissen, P., & van Ginneken, E. (2019). The Dutch Health System In 2019: Many Major Reforms, But Still Work In Progress. Health Policy, 123(3), 249–251. Kieke, O. G. H., & Crivelli, L. (2013). Swiss and Dutch “Consumer-driven Health Care”: Ideal Model or Reality? Health Policy, 109(2), 105–112. KPMG. (2018). Overview of the Dutch Insurance Market in 2017. Resource Document. KPMG. Retreived February 15, 2018, from https://assets.kpmg/ content/dam/kpmg/nl/pdf/2018/sector/verzekeraars/the-dutch-insurance-market-2017.pdf.

278 

P. TURQUET AND P. MARTIN

Krabbe-Alkemade, Y.  J. F.  M., Groot, T.  L. C.  M., & Lindeboom, M. (2017). Competition in the Dutch Hospital Sector: An Analysis of Health Care Volume and Cost. European Journal of Health Economics, 18, 139–153. Kroneman, M., Boerma, W., van den Berg, M., Groenewegen, P., de Jong, J., & van Ginneken, E. (2016). The Netherlands: Health System Review. Health Systems in Transition, 18(2). Resource Document. European Observatory on Health Systems and Policies. Retrieved from https://apps.who.int/iris/bitstream/handle/10665/330244/1817-6127-eng.pdf. Maarse, H., & Jeurissen, P. (2019). Low Institutional Trust in Health Insurers in Dutch Health Care. Health Policy, 123, 288–292. Maarse, H., Jeurissen, P., & Ruwaard, D. (2016). Results of the Market-oriented Reform in the Netherlands: A Review. Health Economics, Policy and Law, 11(2), 161–178. Maarse, H., & Paulus, A. (2011). The Politics of Health-care Reform in the Netherlands since 2006. Health Economics, Policy and Law, 6, 125–134. Ministry of Health, Welfare and Sport. (2018). Healthcare in the Netherlands. Resource Document. Ministry of Health, Welfare and Sport. Retrieved from https://www.government.nl/ministries/ministry-of-health-welfare-and-sport OECD. (2016). Annual Report on Competition Policy in the Netherlands 2015, DAF/COMP/AR, 34. Resource Document. OECD. Retreived June 2, 2019, from https://one.oecd.org/document/DAF/COMP/AR(2016)34/en/pdf. OECD. (2018). Annual Report on Competition Policy in the Netherlands 2017, DAF/COMP/AR, 34. Resource Document. OECD. Retreived June 2, 2019, from https://one.oecd.org/document/DAF/COMP/AR(2018)43/en/pdf OECD. (2019). Annual Report on Competition Policy in the Netherlands 2018, DAF/COMP/AR, 34. Resource Document. OECD. Retreived June 2, 2019, from https://one.oecd.org/document/DAF/COMP/AR(2019)24/en/pdf OECD/EU. (2018). Health at a Glance: Europe 2018: State of Health in the EU Cycle. Paris/EU, Brussels: OECD Publishing. OECD/European Observatory on Health Systems and Policies. (2017). Netherlands: Country Health Profile 2017. State of Health in the EU. Paris/ European Observatory on Health Systems and Policies, Brussels: OECD Publishing. Okma K. G.H. (1997). Studies in Dutch Health Politics, Policies and Law. University Utrecht: Medical Faculty. Okma, K. G. H., Cheng, T. M., Chinitz, D., Crivelli, L., Lim, M. K., Maarse, H., & Labra, M.  E. (2010). Six Countries, Six Health Reform Models? Health Care Reform in Chile, Israel, Singapore, Switzerland, Taiwan and The Netherlands. Journal of Comparative Policy Analysis, 12(1–2), 75–113. https://doi.org/10.1080/13876980903076237. Okma K. G. H., & Crivelli, L. (2013). Swiss and Dutch “Consumer-driven Health Care”: Ideal Model or Reality? Health Policy, 109 (2), 105–112.

9  THE DUTCH WAY: EXPERIMENTING WITH COMPETITION… 

279

Paolucci, F., Den Exter, A., & Van de Ven, W. (2006). Solidarity in Competitive Health Insurance Markets: Analyzing the Relevant EC Legal Framework. Health Economics, Policy and Law, 1, 107–126. Prosser, T. (2005). The Limits of Competition Law. Markets and Public Services. Oxford University Press. Roos, A. F., & Schut, F. T. (2012). Spillover Effects of Supplementary on Basic Health Insurance: Evidence from the Netherlands. European Journal of Health Economics, 13, 51–62. Sauter, W. (2011). The Role of Competition Rules in the Context of Healthcare Reform in the Netherlands. In S.  Van de Gronden & K.  Neergaard (Eds.), Health care and EU Law (pp. 338–358). The Hague: TMC Asser Press. Schut, F.  T., & Varkevisser, M. (2017). Competition Policy for Health Care Provision in the Netherlands. Health Policy, 121, 126–133. Stolper, K. C. F., Boonen, L. H. H. M., Schut, F. T., & Varkevisser, M. (2019). Managed Competition in the Netherlands: Do Insurers have Incentives to Steer on Quality? Health Policy, 123, 293–299. Thomson, S., & Mossialos, E. (2006). Choice of Public or Private Health Insurance: Learning from the Experience of Germany and the Netherland. Journal of European Social Policy, 16(4), 315–327. Thomson, S., & Mossialos, E. (2010). Private Health Insurance and the Internal Market. In E. Mossialos (Ed.), Health Systems Governance in Europe. The Role of EU Law and Policy (pp. 419–460). Cambridge: Cambridge University Press. Turquet, P. (2012). Health Insurance System Financing Reforms in the Netherlands, Germany and France: Repercussions for Coverage and Redistribution? International Social Security Review, 65(1), 29–51. Turquet, P. (2013). Les réformes du financement de la protection sociale en Europe : l’exemple de l’assurance maladie. In J.-C.  Defraigne, J.-L. de Meulemeester, D.  Duez, & Y.  Vandenborght (Eds.), Les modèles sociaux en Europe. Quel avenir face à la crise? (pp.  129–146). Bruxelles: Idées d’Europe, Bruylant. Van de Gronden, J., & Szyszczak, E. (2014). Introducing Competition Principles into the Healthcare through EU Law and Policy: A Case Study of the Netherlands. Medical Law Review, 22(2), 238–254. van de Ven, W.  P. M.  M., & Schut, F.  T. (2008). Universal Mandatory Health Insurance in the Netherlands: A Model for the United States? Health Affairs, 27(3), 771–781. van de Ven, W. P. M. M., & Schut, F. T. (2009). Managed Competition in the Netherlands: Still Work-in-progress. Health Economics, 18(3), 253–255. van de Ven, W. P. M. M., van Vliet, R. C. J. A., & van Kleef, R. C. (2017). How can the Regulator Show Evidence of (No) Risk Selection in Health Insurance Markets? Conceptual Framework and Empirical Evidence. European Journal of Health Economics, 18, 167–180.

280 

P. TURQUET AND P. MARTIN

van der Geest, S. A., & Varkevisser, M. (2016). Using the Deductible for Patient channeling: Did Preferred Providers Gain Patient Volume? European Journal of Health Economics, 17, 645–652. Van Esch, T. E., Brabers, A. E., van Dijk, C. E., Gusdorf, L., Groenewegen, P. P., & de Jong, J. D. (2017). Increased Cost Sharing and Changes in Noncompliance with Specialty Referrals in The Netherlands. Health Policy, 121(2), 180–188. Van Ginneken, E., Busse, R., & Gericke, C. A. (2008). Universal Private Health Insurance in the Netherlands: The First Year. Journal of Management Marketing in Healthcare, 1(2), 139–153. Van Ginneken, E., Schäfer, W., & Kroneman, M. (2011). Managed Competition in the Netherlands: An Example for Others? Eurohealth, 16(4), 23–26. van Winssen, K. P. M., van Kleef, R. C., & van de Ven, W. P. M. M. (2017). A Voluntary Deductible in Health Insurance: The More Years You Opt for it, the Lower your Premium? European Journal of Health Economics, 18, 209–226. Vonk, R., & Schut, F. (2018). Can Universal Access be Achieved in a Voluntary Private Health Insurance Market? Dutch Private Insurers Caught between Competing Logics. Health Economics, Policy and Law. https://doi. org/10.1017/S1744133118000142. Willemse-Duijmelinck, D. M. I. D., van de Ven, W. P. M. M., & Mosca, I. (2017). Supplementary Insurance as a Switching Cost for Basic Health Insurance: Empirical Results from the Netherlands. Health Policy, 121, 1085–1092.

CHAPTER 10

In Between the Market and Public Health Insurance: A Place for Occupational Welfare in Europe? Thomas Houssoy, Marion Del Sol, and Philippe Martin

1   Introduction There are several approaches in  the literature to describe welfare state models. The most common one is the scheme-based approach Economists also employ the insurance-based approach,1 while some studies (including those presented in this volume) are based on the notion of public-private 1  In this case, the Welfare State is designed to be one broad insurance policy against social risks, and the systems can be described by the manner in which they distribute public cover-

T. Houssoy (*) Clersé (UMR CNRS 8019), Université de Lille, Lille, France e-mail: [email protected] M. Del Sol Intitut de l’Ouest : Droit et Europe, Université de Rennes 1, Rennes, France e-mail: [email protected] P. Martin Centre for Comparative Labour Law and Social Security, Université de Bordeaux, CNRS, Pessac, France e-mail: [email protected] © The Author(s) 2021 C. Benoît et al. (eds.), Private Health Insurance and the European Union, https://doi.org/10.1007/978-3-030-54355-6_10

281

282 

T. HOUSSOY ET AL.

mix, or even  more favorably that of  welfare mix. This approach insists on the combination of different spheres in the provision of social services, namely the State, the market, the family and the sphere of “voluntary associations”. In this chapter, we argue that it    can be linked with the multi-pillar conception that prevails  in pension  reform  and studies. Thinking in terms of “pillars” typically means differentiating between the respective places of public-based schemes (pillar 1), professional schemes operating on a capitalisation basis (pillar 2) and individual voluntary retirement savings plans acting as life annuities (pillar 3). With regard to the growing role of professional pensions (pillar 2), studies on occupational welfare, which emphasize the role of  benefits and services received by employees on the basis of the employment relationship, constitutes a sound and frequent tool for the analysis of multi-pillarisation processes (Pavolini and Seeleib-Kaiser 2018). This line of studies allows to add issues of interactions and balance between pillars to the analysis, notably regarding the segmentation of financing in the public-private mix in  pensions coverage. It is time, we contend in the following lines, to think about potential points of intersection between this approach and the study of the public-private mix in healthcare. As  repeatidly shown by the contributions to this book, the public-­ private mix is indeed  a configuration that one can also find  in  national health insurance systems. In numerous European countries, access to care is achieved by means of (compulsory) public health insurance combined with work-related and individually contracted private insurance  policies,  depending on a set of arrangements that vary from a country  to another. As in pensions, formulas based on occupational welfare are thus present in the field of healthcare and are indeed relatively well developed within the European Union (EU). However, the proportion of the population covered by private insurance considerably varies across the EU.2 This also applies to the distribution between group and individual coverages by private health insurance (Sagan and Thomson 2016). Group age (more or less broad and universal), but also by the place granted to private insurance and a market to cover certain risks. 2  It is particularly high in those countries that leave it to the insurance market to cover the co-payment factor, that is, that part of the care not reimbursed by compulsory public health insurance. That is the case in France, where over 95% of the population is now covered by complementary private insurance, and in Slovenia, where the coverage rate is 84%. The Netherlands, Austria, Belgium and Ireland also have a high coverage rate (Sagan and Thomson 2016).

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

283

policies are overwhelmingly dominant in Eastern Europe and Scandinavia (between 80% and 100% of the private insurance market). They concerns the majority of the privately insured population in the UK, Belgium and the Netherlands (between 65% and 75%) but are significantly less developed in other countries like Italy, Austria, Germany and Ireland (between 13% and 42%). Although individual policies still constitued the majority on the private insurance market (52% of contributions paid in 2018) in France, the volume of group policies is rising sharply and could reverse the dominant position in their favour (DREES-IRDES 2019). However, it was mainly in the area of pensions that we saw major reforms in the early 2000s Europe. They systematically mobilised private insurance policies and, by extension, market mechanisms to complement a reduction in public pensions. These reforms particularly professional coverages, were activated through a set of tax incentives to cover a large proportion of the population immediately in conditions that are favourable for employees due to the employer contribution. It should also be noted that “pillarisation” is not organised in the same way for pensions as for health coverage. In the latter case, the distribution between public and private insurance depends on more complex factors: in a single country, the public coverage can be very strong for some pathologies and more residual for other types of care, opening different options for private insurance.3 Using  occupational welfare literature to analyse health insurance is therefore far from straightforward. Furthermore, it is an analytical framework rarely used in this field. However, the comparative analysis of health insurance systems, as this chapter intends to show, would benefit from using these studies. They make it possible to examine the systemic transformations that the mobilisation of  professional coverage can bring or cause, especially with regard to access to care and in terms of solidarity. More specifically, we argue that emphasizing on the “professional pillar” in the field of health coverage opens up a perspective that considerably enhances the analysis. This is not (or not only) a matter of helping to provide guidance for the privatisation or commodification processes; it is about considering the bridges or alternatives that the mobilisation, or even instrumentalisation, of professional coverage can build between 3  In France, for example, the care required by long-term disorders is covered by the Social Security, although responsibility for other pathologies is shared by Social Security and private insurance.

284 

T. HOUSSOY ET AL.

national solidarity (public system) and that prevailing on the health insurance market. In the next section, we  discuss  the notion of  pillar and question the place of occupational welfare (2). Then, we introduce a case of instrumentalisation of the “professional pillar” by public policies, namely the French one,  which appears quite singular to the point of constituting a quasi-­ natural experiment (3). Indeed, a law of June 14, 2013 requires private sector employers to provide their employees with complementary health coverage and to contribute at least half of its funding. This shift immediately  gave  an unprecedented place to collective  health insurance  in the country, between public health insurance and individual private insurance, as legislators  have institutionalised the role of professional  (or occupational) coverage. We then discuss the many implications of this largely unprecedented reform, and this both in France and in Europe. In particular, we look at the distribution of resources aimed at satisfying social needs (in this case, access to care) and the effects of the segmentation in terms of inequalities and the overall solidarity of the system (4). Finally, we reflect and discuss on the potential for collective bargaining to build solidarity in a trading environment (5). Indeed, contemporary versions of occupational welfare never fail to remind us that it is important not to “underestimate the role of the trade unions and the collective aspect of professional relationships in the negotiation of social protection programmes” (Natali et al. 2018).

2   The Growing Importance of Occupational Welfare in Europe as a “Pillar” of Social Protection One can argue that the French health insurance system tends to take the form of a “pillar-based” architecture: its main configuration increasingly articulates a nationwide public scheme with occupational complementary health insurance plans set up at company or at sector level. This description may become even more relevant as supercomplementary health insurance plans have played a bigger role since occupational health insurance (OHI) schemes became mandatory: 2.4 million people are estimated to have taken out such a health insurance contract by 2016 to top up their complementary coverage. For most of the rest of the population, complementary health insurance remains based on the voluntary signing of individual contracts.

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

285

This description, however relevant at first sight, cannot be understood as a direct translation of the “three-pillars doctrine” commonly used to discuss or promote pension reforms (2.1). Moreover, its limitations do not simply apply to an analysis of health insurance in France, but can be addressed to any uses of the concept of pillars for describing any particular institutional social protection configuration covering any kind of welfare requirement. We therefore offer a more generalist description centred around the idea of a pluralist welfare system (i.e. social risk needs in a given community are answered through State and non-State schemes, together forming a welfare system) and institutional complementarity (i.e. the overall outcome (or lack of) of a given welfare system must be seen in the light of the mutually beneficial interactions of its components—in other words, the whole is not the sum of its parts). Drawing on the conclusion that pillars are relevant, although limited to the description of the different components of a welfare system, a closer look at studies of occupational welfare schemes is of great help in refining an analysis of the new welfare system that emerged (2.2). However marginal occupational welfare was in social policy analysis, the few studies on the subject emphasise the importance of crossing a “provision” perspective with the questions of the general allocation of resources for welfare needs (finance perspective) with the way industrial relations actors set up occupational schemes and define the population that can access them and the solidarity they can create (regime perspective). 2.1   The Problem with the Pillar Perspective Taken at Face Value The three-pillars’ perspective has become a cliché of comparative social policy study and social policy reform, most notably when applied to pension policies (e.g. Immergut et al. 2006). This depiction articulates three pillars (the public, the occupational and the individual pillars), each on top of the other to achieve the coverage of a given social risk. Popularised by the landmark 1994 World Bank report “Averting the Old Age Crisis”, this apparently descriptive tool cannot simply be untied from a normative agenda of structural reform favouring the privatisation of pension management and financing. This derives from the manner in which the public-­ private mix is defined as the functional distribution of roles. The “three-pillar doctrine” has taken many forms, but is most commonly seen as a generalisation of a national case, that of Switzerland, by an epistemic

286 

T. HOUSSOY ET AL.

community of life insurers and pension consultants during the 1970s and early 1980s (Leimgruber 2012). Yet, the seminal World Bank report (1994) does not exclusively attribute an occupational character to the mandatory privately managed second pillar, nor a personal character to the voluntary top-up third pillar, as usually associated with the Swiss/classical three-pillar model. However, this report contains the most explicit exposition of the underlying common normative project any version of this model carries: an optimal pension scheme must exclusively distribute its redistributive function to a public scheme and its savings function to private schemes (ibid 1994, p. 15).4 This is a conception of the distribution of roles between public and private that is in phase with the then-prevailing New-keynesian influence and the (American) intellectual background of the report’s authors (ApRoberts 1996). Overall, it can be summed up by two tenets. Firstly, it supposed that the market remains the optimal determination mechanism of resource allocation and employment, and the State’s intervention must be limited to redistribution and the correction of market failures (Bönker 2005, pp. 83–86). This functionalist perspective results in the definition of the first pillar, a public scheme mostly dedicated to providing some form of minimum benefits that does not rely on market outcomes,5 and the definition of the second pillar as a mandatory and State-regulated funded scheme in order to obtain a large coverage and prevent under-savings.6 Secondly, it supposed that savings, through the insurance technique, are the “natural” form of social protection (Ewald 1986, quoted in Ramaux 2007, pp. 22–25). This tenet is the driving force of the “three-pillar doctrine” as a privatisation tool and derives from the statement that actuarial neutrality, cornerstone of the private insurance technique and intertemporal optimal savings management, “should avoid some of the economic and political distortions to which the public pillar is prone” 4  One made mandatory and tightly regulated by the State, and the other made voluntary and mostly under the full sway of classical insurance techniques and financial markets. 5  In relation to this, the 1994 report notices that “[this first] pillar has the unique ability […] to coinsure against long spells of low investment returns, recession, inflation, and private market failures” (ibid 1994, p. 16). It points at the way the State intervenes against the overall major failure of the market when its role is to honour long-term commitments: by socializing the cost of its fundamental inability to provide long-term and secure yields. 6  The necessity of such State’s intervention is framed as a counterbalance to either a high individual temporal discounting from the subscribers or as a limited access to the insurance market due to information asymmetries.

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

287

(World Bank 1994, p. 16). Such a narrow definition of social protection commands the scope of the privately managed pillars and makes it the dominant variable of the “three-pillar” architecture. We can derive that the rejection of any a priori functional distribution of role between “pillars” is a prerequisite for establishing the notion of pillar as a relevant descriptive tool for identifying and studying any form of welfare system as a pluralist institutional arrangement (e.g. Johnson 1999; Powell 2007a). We argue, in a similar manner to our study of the coherency of the “three-pillar doctrine”, that understanding a given system of schemes empirically requires not simply describing each element, but also identifying the state of institutional complementarity (Amable 2004; Aoki 2001; Deeg 2007) between each of them. Institutional complementarity can be defined as the state where “the co-existence (within a given system) of two or more institutions [here, health insurance schemes] mutually enhances the performance contribution of each individual institution” (Deeg 2007, p.  611). Drawing on Powell’s extensive literature review (Powell 2007b, p. 4) of past welfare pluralism7 studies, there would be a descriptive and prescriptive use of such a perspective. Although we do not adhere to such terminology, our use of the concept of institutional complementarity to describe a welfare system would fall in the category of “prescriptive” analysis. It draws from sharing the fundamental assumption (and observations in our case studies, as we will argue) that a welfare mix is not simply about the additive cumulation of all its components, that variations in the welfare mix over time cannot be considered a “mere rearranging of furniture in the drawing-room” (Mishra 1990, p. 112, quoted in Powell 2007a, p. 4). Using this framework on the French case in health insurance, the development of occupational health insurance must be understood as part of an overall reconfiguration of the relations between the national health insurance (NHI) scheme and the private health insurance market. Additionally, we can analyse how it is linked with changes in policy paradigm (Amable and Palombarini 2009). Indeed, as we will argue, the French case illustrates how a shift in the role of private insurance and the development of company- and sector-based complementary schemes is not so much a direct application of the “three-pillars doctrine” as a pragmatic and 7  The term “mixed economy of welfare” is also used widely, including by Martin Powell himself. We follow Johnson (1999, p. 22) in considering both terms “identical in meaning and […] therefore can be used interchangeably” (Powell 2007b, p. 2).

288 

T. HOUSSOY ET AL.

unplanned response to a shift in paradigm, here in health policy. Similarities with the case of pensions must, therefore, be understood as deriving from a similar reconfiguration favouring private management within the peculiarities of health insurance institutions and debates. 2.2  The Complexification of European Welfare Systems in the Early Twenty-First Century and the Development of Occupational Welfare Although wide variations exist in the distribution of importance of the State intervention relative to non-State interventions across social risks and countries in Europe, many studies of social policy focus on the role of the State in finance, provision or both. Such focus cannot be considered wrong-minded as many European welfare systems are unequivocially welfare state in the sense that private actors, for-profit or not, and alternative forms of social order to the State (Streeck and Schmitter 1985) are usually either dominated by State intervention or clearly marginalised when appraised on a quantitative basis (most notably in financial terms). However, as we argued, the non-additive nature of welfare systems, once we account for all its different components, gives grounds for assuming the complexities emerging from the interaction between components are an integral part of any analysis. Many recent proposals of pluralist analysis of social policies and social protection (Ebbinghaus 2011a; Goodin and Rein 2001; Powell 2007b) offers hints on how to analyse such systems. Overall, welfare systems are institutional arrangements that do not rest on consistent categories encompassing dimensions of governance, finance, benefits logics, and so on. Henceforth, there is little to no relevancy in uni-dimensional descriptions and typologies. For the sake of clarity, we limit ourselves to a two-dimensional reading of the situation, allowing for discussions on the interaction of multiple non-coincidental dimensions. Drawing on Ebbinghaus (2011a), a welfare mix for a given social policy can be described by answering five questions for each of its components: (i) who is covered, (ii) what kind of benefits are served, (iii) who pays, (iv) who provides and (v) who governs, decides and manages. Adopting the typology of pillars, and therefore a provision perspective (i.e. focusing on question 4), allows for a practical description of the main configuration a welfare system can take. However, it hardly helps to established clear links with the other four questions considered alone. Using the French health insurance system as an example, the NHI scheme can easily be identified

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

289

as the public pillar, but its evolution over time shows great shifts in the content of all of the other axes. An oversimplified presentation of this evolution would depict it as a social insurance that, between 1945 and today, moved from covering most private wage-earners to being close to a universal scheme, from serving mostly as “sick pay” to almost exclusively insuring health consumption expenditure, from being financed almost exclusively by social contributions to being mostly financed by taxes, and from self-administration by trade unions and employers associations to de facto State administration. A closer look at the occupational pillar might help select proper second dimensions for analysis. Occupational welfare has a long history despite being rarely put forward in social policy analysis during the pre-2000s period (Rein 1982; Rein and Wadensjö 1998; Shalev 1996). Coined by Richard Titmuss in his seminal lecture, The Social Division of Welfare (1958), it referred first to a vast and vaguely defined array of “benefits received by an employee through or as a result of his employment over and beyond the public benefits such as national insurance” (Sinfield 1978, p. 130). However, such a definition is ill-fitted to apprehend its contribution to the welfare mix as it hardly differentiates between simple company perks (such as company cars) and social benefits (Goodin and Rein 2001, pp. 773–774). Overall, its most fruitful contribution, both descriptively and analytically, lies in the original formulation of Richard Titmuss as a tool for understanding the social division of welfare. As a descriptive tool, looking at public welfare alone (i.e. “the” welfare state as commonly understood) is not enough; occupational and fiscal welfare must be added to the analysis8 (e.g. Hacker 2002, 2004; Howard 1997) in order to identify all socially determined tools for financing or providing services and goods fulfilling welfare needs.9 As an analytical tool, it enables the examination of the full structure of resource allocation for welfare needs, most notably to highlight the anti-redistributive outcomes of the welfare system that an exclusive focus on the public welfare too easily hides. Although it had an Anglo-Saxon focus, probably due to the substantial increase in occupational health insurance in the United States and 8  It refers to all forms of deductions, rebates or fiscal measures the State used to promote and orient private initiatives. 9  The most common example related to our case study is that occupational pensions or health insurance schemes contributions or benefits are often deducted from the calculation of taxes.

290 

T. HOUSSOY ET AL.

occupational pensions in the UK (e.g. Cutler and Waine 2001), occupational welfare has always been a significant but often neglected part of the European welfare systems (Farnsworth 2004; Rein 1996), including France. Such a shortcoming in the study of (mostly West and Northern) European welfare states is progressively addressed by an expanding literature since the mid-2000s (Ebbinghaus 2011b; Greve 2007; Natali et al. 2018). This literature mostly stems from the rising importance of occupational pensions in all types of European welfare state (e.g. De Deken 2018; Naczyk 2013; Trampusch 2007), but expanded to other kinds of social risk (e.g. Greve 2018; Järvi and Kuivalainen 2013; Seeleib-Kaiser and Fleckenstein 2009; Yerkes and Tijdens 2010). While the first trend focused on defining occupational welfare and assessing its scope and distributional outcomes, this second trend tends to focus on the institutional set-ups and actors involved, particularly those in industrial relations. Overall, two strands exist in this segment of the literature, representing opposite theses on the outcomes of the implications of actors in industrial relations, particularly trade unions: a solidaristic-collectivisation thesis (Johnston et al. 2011; Trampusch 2009) and an egoistic-dualisation thesis (Häusermann 2010; Palier and Thelen 2010). The solidaristic or collectivisation thesis argues that trade unions might push for negotiated occupational welfare in order to cope with a lack or retreat of the public welfare state (Yerkes and Tijdens 2010) but also to maintain or gain power and position and extend their constituency (Naczyk and Seeleib-Kaiser 2015). According to this thesis, this results in the development of solidaristic occupational welfare. Following prior studies emphasising important intra-labour divergent preferences in social policy and possibilities of cross-­ class alliances (Hall and Soskice 2001; Mares 2003; Swenson 1991; Thelen 1993), the egoistic or dualisation thesis posits that trade unions in core sectors, facing post-industrialism and decentralisation of collective bargaining, shifted from promoting encompassing social improvements nationwide to local agreements with management to protect their position on the labour market. According to this thesis, this results in the development of a welfare dualism where “insiders” benefit from generous occupational welfare schemes, leaving the “outsiders” facing little to no access to such a form of social protection. This dual interpretation of the concept of occupational welfare helps us to highlight two fruitful forms of interactions with the provision perspective: a finance perspective can clarify how the configuration of welfare resources allocation is organised among the different forms of health

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

291

insurance provision (Sect. 4) and a regime perspective focuses on the reconfiguration of the forms of solidarity prompted by the emergence of OHI schemes in France (Sect. 5).

3   The Activation of an Aspect of Welfare for a New Purpose: The French Case of Occupational Health Insurance Occupational welfare is a form of social protection that is not only ancient but never truly disappeared or became marginalised when the modern French welfare state, centred around public welfare, was instituted in 1945. However, the case of OHI shows that it neither grew from the immediate post-war period to become an important part of the French health insurance system, nor was it a component that received much attention from the public authorities until a later period starting in the 1980s. On the contrary, a historical perspective on OHI in France shows that it emerged recently when the French Government chose to “activate” this aspect of welfare as a tool for a larger renewal of public health policy. Even within this shorter period, the important role OHI plays nowadays is not so much the results of a planned process, but rather the results of accidents and opportunities. It follows that the history of OHI in France can be roughly understood as a succession of two periods. The first and longest one spans from the early industrialisation to the 1980s (3.1). During this period, OHI is firmly a part of the social policy of the firm (Rein 1982). Everything from setting it up to the forms it took resulted from the work of industrial relations actors. Public authorities, on the other hand, were mostly acting through the expansion of NHI. From the end of the 1980s onwards, changes in the political economy of the French health insurance system led to an increasing role played by private health insurance, under the guidance of the French Government (3.2). During this period, it was the overall goal of increasing access to complementary health insurance through ever-stricter regulation that was the main driver of institutional change. The 2004 Douste-Blazy Law is the tipping point when OHI is explicitly seized by public authorities as a tool for public health policy at large.

292 

T. HOUSSOY ET AL.

3.1  A Constitutive Part of the Social Protection of French Private Wage-Earners Since the Inception of a National Health Insurance Like many European welfare states (Farnsworth 2012), the roots of the modern French welfare state are to be found in the development of occupational welfare in the new industrial urban centres that emerged during the second half of the nineteenth century and early twentieth century (Castel 2002; Hatzfeld 1971). Up to the interwar period, these health insurance schemes remained concentrated on an ever-growing share of civil servants and industrial workers in the population. By the late 1920s, this constellation of occupational schemes for State employees, train drivers, coal miners, and so on covered 56% of the total population (Dreyfus 2009). The first attempt at creating a mandatory health insurance coverage took place under the provisions set up by the law of 1928,10 modified on July 1, 1930. Although it cannot be described as making OHI schemes mandatory, it was clearly based on the experiences these schemes provided. Moreover, while setting up a national framework, the laws of 1928 and 1930 allowed pre-existing occupational schemes to continue to exist and to act as substitutes for the new institutional framework. The ability of old occupational schemes to survive the progression of public welfare is one of the many compromises made for setting up this mandatory health insurance scheme that determined the evolution of French social protection to the present day. With the creation of the Sécurité sociale (literally, “Social security”) in 1945,11 the social policy of a firm was now an element of complementary rather than basic social protection. The period 1945–1946 corresponds to a phase of increasing employees’ power, directly or through their representatives. Work councils were instituted12 and given an important role in setting up and managing the firm’s social policy, contributing to an initial movement away from traditional paternalism. By 1946, the new social protection architecture enacted the existence of a second complementary  Law of April 5th, 1928 on social insurance.  Ordinance of October 4, 1945, on the organisation of Social Security and Ordinance of October 19, 1945, on the social insurances for insurees in non-agriculutral occupations. 12  Ordinance of Feburary 22, 1945, instituting companies’ councils (in French, comités d’entreprise, which were roughly what is commonly designated as workers’ councils). The comités d’entreprise were replaced by the comités sociaux et économiques (social and economic councils) by January 1, 2018. 10 11

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

293

tier where firms play a role and established a new kind of insurance institution for collective social risks coverage, set up through collective bargaining by trade unions and employers and administered by them.13 The decision to set up any occupational welfare scheme for the benefit of the employees and the determination of the employers’ participation have, since then, been left either to the discretionary decision of the employer or to collective bargaining, at either company or sector level.14 In these early days of the modern French welfare state, firms’ social policies and initiatives are neither significantly regulated nor instrumentalised by the national government as a prolongation of public social policies. One exception is the case of complementary occupational pensions which emerged under highly peculiar circumstances and forms (Friot 2012, pp. 93–122), leading to the establishment of a nationwide quasi-public scheme on top of the basic public scheme by the 1970s. Collective bargaining at sector level was crucial for the development of the coverage of many social risks. However, OHI was and remained a matter of collective bargaining at company level (Kerleau 2009, p. 17). This was the case since the early days of the modern French welfare state, as some of the pre-existing occupational schemes survived by transiting to company-based mutual funds,15 sometimes under the supervision of the workers’ council (Kerleau et al. 2008, p. 19). In any case, the initiative is in the hands of private actors. Within the firm, social protection is part of a large array of elements forming the remuneration of the employees, from perks to social advantages. No employer has any mandate to set up an OHI scheme but it can be used as a tool in the larger trade-offs constituting a firm’s wage policy. Most of the second half of the twentieth century is a period of development of social advantages earned through employment in a firm, but it is excessive to consider these dynamics as a process of institutionalisation of occupational welfare in France (contra Kerleau et al. 2008, p. iii). On one hand, setting up such welfare schemes remained highly discretionary and practices were highly differentiated across companies. On the other hand, as an element of trade-off and bargaining on wages, the forms and extent these occupational health schemes took were 13  Ordinance of June 8, 1946, on the common provisions for complementary welfare for employees (in French, régimes complémentaires des salariés). 14  A third procedure consists of setting up occupational schemes by directly consulting the employees by referendum. However, this procedure is rarely used. 15  Following the Ordinance of October 19, 1945, on the statutes of mutual benefit societies (in French, mutualité).

294 

T. HOUSSOY ET AL.

highly dependent on the local compromises reached between the industrial actors involved. Public authorities first ventured into making OHI schemes a tool of public policies in the 1970s and 1980s, through two means: the creation of tax and social contributions exemption on one hand, and the strengthening of regulation on the other hand. The French Government first introduced tax exemptions in 1975: employees could deduct the premium they paid for health insurance from their income tax, within a limited amount, providing the occupational scheme was collective (i.e. it was made accessible to a category or all of the employees of a given company) and mandatory (i.e. all employees of a category or company had to subscribe to the scheme) (Kerleau 2009, p.  18). During the 1980s, these advantages were expanded to the benefit of the employers by deducting their financial contributions16 to any occupational schemes from both the base value for the calculations of social contributions to Social Security and other legal and national social protection schemes, and from corporate tax. The second instrument the French Government used to advance public policy goals was the new regulation introduced by the pivotal 1989 Evin Law.17 The 1980s was a crucial turning point for the evolution of the private health insurance market in France (see also Chap. 7). Although OHI schemes were not the focus of this reform, the 1989 Evin Law prepared the way for the increasing role private health insurance was about to play in financing health consumption and fulfilling public health goals. Many central provisions to this day, such as the prohibition of medical selection and the guarantees of coverage maintenance in unemployment, were introduced by the 1989 Evin Law. Overall, this first wave of public interventions was relatively minor for OHI schemes. Public authorities incentivised their expansion and established some protections for employees, but never intended to fully instrumentalise these schemes or orientate their contents, finance and mode of governance.

16  French law makes a distinction between taxes (i.e. fiscal in French) paid to the State, and social contributions (i.e. social in French) paid to Social Security and other legal and national social protection schemes. They are organized as two separate sets of mandatory contributions and are governed by different sets of rules. 17  Law of December 31, 1989, reinforcing the social guarantees provided to persons insured against certain risks.

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

295

3.2  The Late Development of Occupational Health Insurance in France The last pivotal moment leading to the generalisation of OHI in France started with the generalisation of the basic NHI through the creation of the Couverture Médicale Universelle (CMU).18 During the debate leading to the vote on the 1999 CMU Law, the high dependency of any beneficiary to private insurance financing forced the public authorities to include a free complementary stage to this new scheme (CMU-c19). This resulted in two major modifications. The first major disjunction with the institutional path in place since 1945 was the paradigm shift in public policy from an objective of universal (basic) health insurance coverage to an objective of greater access to private complementary health insurance (Kerleau 2012). This shift in the evolution of the French health insurance system was the result of almost two decades of transformation of its political economy. However, nothing at that moment required OHI schemes to be detrimental to this new health policy goal. The second major disjunction with the previous institutional path is more accidental and helped push OHI schemes to play a larger role in the new health insurance system. Through an amendment of Article 21 of the 1999 CMU Law sponsored by Communist members of Parliament, two modifications of the labour law were included that truly set the stage for increasing coverage by OHI and for promoting collective bargaining as its main form of governance (ibid 2012, p. 9). From 1999 onwards, the inclusion of provisions on OHI is required in any sectoral branch agreements for benefiting from the erga omnes clause delivered by the Ministry of Labour.20 Moreover, setting up an OHI scheme must be included on the agenda of the Négociation Annuelle Obligatoire21 in each company with at least one trade union representative if such a scheme is absent at sector or company level. Although this last provision was intended to stimulate the expansion of OHI through company-level agreements, it did not include any 18  Literally, Universal Health Coverage. This was replaced by the Protection Universelle Médicale (PUMa, literally Universal Health Protection) by January 1st, 2016. 19  “c” standing for complémentaire (complementary). 20  An erga omnes clause (clause d’extension in French) allows the binding provision of a branch agreement to be extended to all companies and employees within that branch, whether a company or employee is affiliated to the signing employer associations or trade unions. For such a clause to exist, the French Minister of Labour must sign a decree. 21  Literally, Mandatory Annual Bargaining, introduced by the Law of November, 13, 1982, on collective bargaining and collective labour conflict resolution.

296 

T. HOUSSOY ET AL.

­ andate for concluding such agreements, only a mandate to initiate talks. m Therefore, besides being accidental modifications, these provisions remained, in practice, within the domain of incentives rather than direct constraining regulation, and therefore depended on the will of industrial actors. The 1999 CMU law opened up a short period of uncertainty concerning the tools public authorities would use to increase access to private complementary health insurance. Two laws, passed in 2003 and 2004, settled the question in favour of a combination of higher regulation on the content of health insurance contracts, special schemes directed to low-­ income beneficiaries and increased subsidisation of OHI schemes through fiscal welfare. The 2003 Fillon Law22 drastically reformed the French pension system and was an opportunity to promote funded occupational pensions by reforming the logics and use of fiscal welfare (Kerleau et al. 2008, p. 33). It did so by making conditions for accessing tax and social contributions exemptions stricter.23 This resulted in introducing a new logic in the use of fiscal welfare by the French Government, since its aim was not simply to promote occupational welfare, but also to dictate its main parameters. What was at stake was “organizing and rationalizing social protection within the firm” (Kerleau 2012, p. 186, our translation). By reforming the entire apparatus of tax and social contributions advantages, fiscal welfare became a “privileged tool of public regulation of [private] social protection” (Zemmour 2013, p.  53 our translation). Hence, the 2003 pension reform affected every form of occupational welfare and had the greatest impact in the case of OHI. If the role of the pension reform seems accidental, the 2004 Douste-Blazy reform24 was the moment when public authorities explicitly included OHI as a tool of public health policy. Building on the 2003 Fillion Law, it tightened the conditions for benefiting from tax and social contributions exemptions by requiring the benefits basket to abide by a legally defined standard. This standard, known as contrat solidaire et responsable,25 was not limited to setting minimum levels of coverage but also aimed to promote a financial regulation built on a model of responsible, individual and rational beneficiaries within the entire  Law of August 21, 2003, on pension reform.  Decree of May 9, 2005. 24  Law of August 13, 2004, on health insurance reform. 25  Literally, solidarity-based and responsible contract. Its first parameters were set up by the Decree of September 29, 2005. 22 23

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

297

French health insurance system. This goal required private health insurance not to offset the array of incentives and disincentives introduced in the NHI and therefore made insurance providers actors of the implementation of the public health policy (Batifoulier et al. 2007). Overall, OHI schemes were instrumentalised as means to facilitate access to private insurance by preventing most individuals from being alone and powerless on the health insurance market through a form of regulated collectivisation (i.e. strengthening the demand-side through group purchase). To that extent, the French case of OHI is similar to other recent experiences of occupational welfare (e.g. Johnston et  al. 2011). The new regulation that emerged in 2004 sponsored a model of occupational schemes close to a private version of social insurance. It only subsidised schemes with a mandatory subscription of employees (i.e. based on mutualisation) and equal co-financing by the employers (i.e. the same cost-sharing within the whole of the employees or an objectively defined group of them). This represents the model around which OHI schemes are organised up to the present day. However, even though the 2003 and 2004 laws increased regulation on these schemes, public intervention remained based on incentives rather than a direct mandate up to 2013. The intermediate period, while appearing calm, showed signs that one last paradigm shift was building up. Evidence of one such sign can be found in a commentary by Jean-Marie Spaeth (2008, p. 36), former chairman of the NHI Board of Administrator, who emphasised that while industrial relations actors lost most of their grasp on the administration of the NHI by 2004, OHI was an opportunity for the renewal of their action in social protection. Moreover, the Commission Nationale de la Négociation Collective26 delivered on February 6, 2007, to the Minister of Labour Gérard Larcher a report inviting employers’ associations and trade unions to reach a nationwide agreement on making OHI mandatory throughout the private sector (Kerleau 2009, p. 19). Such an agreement was reached on January 11, 2013,27 and prompted a significant process of paradigm shift that lasted until 2016 and the full implementation of the mandate. First, it definitively shifted OHI schemes within the domain of public policy by abandoning the logics of incentives (Del Sol 2014). This shift, however, cannot be reduced to the mandate put on private companies to  Literally, National Commission on Collective Bargaining.  Translated by the Law of June 14, 2013, on employment security (in French, sécurisation de l’emploi). 26 27

298 

T. HOUSSOY ET AL.

provide health insurance coverage, since public authorities also chose to drastically modify the regulation put in place in 2004. Beyond introducing a minimum employers’ share of co-financing (50%), it introduced a new standard benefits basket which set minimum and maximum benefit levels.28

4   A Political Option that Redefines the Distribution of Welfare Resources The reconfiguration of the socialised and private financing of healthcare in France represents a political decision that has transformed the role of national and complementary health insurance. Understanding this means, in the first instance, identifying the theoretical and ideological principles of the movements towards the privatisation of healthcare systems (André et al. 2016), in particular the various strategies for competition that can be identified in the recent transformations undergone by the Bismarckian health insurance schemes (Hassenteufel and Palier 2007), and an institutional complementarity between the two major types of health insurance (4.1). The first term enables several successive governments of different political parties to contribute to this reconfiguration by sharing a single representation of the problems facing the French health insurance system and therefore the solutions to be employed to resolve them. The second term makes it possible to facilitate the privatisation process. On the one hand, this has involved taking into account the risk of under-consumption of healthcare and the ethical concerns that the exposure of vulnerable or seriously ill populations to private health insurance can cause. And on the other hand, the segmentation of the financing in line with the type of healthcare enables complementary health insurance organisations to focus on the financing of ambulatory or primary care, insurance for which is financially more sustainable. However, this choice is not without impact on the solidarity in the health insurance system or on the inequalities of healthcare (4.2). Accordingly, the key aspect of solidarity in the French health insurance system is the result of the presence of Social Security, and especially of its financing in proportion to income. Furthermore, the historic flaws in the regulation of private medicine in France generate worrying levels of inequality of access and failures to take up care, mainly for financial reasons. This is in addition to inequalities in the quality of reimbursements  Decree of November 18, 2014, and Social Security Budget Law for 2014.

28

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

299

and in the financial cost of individual and group policies. Thus, the privatisation of the financing of ambulatory or primary care is a policy that encourages the proliferation of inequalities on the labour market and favours some socio-professional groups via group policies. 4.1  Privatisation and Segmentation of NHI Coverage as a Root of the Development of Occupational Health Insurance The development of group coverage for complementary health insurance is inseparable from the reconfiguration process of the roles between Social Security and the insurance market that has occurred in France since the early 1990s (see also Chap. 7). This process, presented as a “silent metamorphosis of health insurance” (Tabuteau 2010), can be summarised as a change in the distribution of private insurance financing which, while officially remaining complementary (i.e. provided to finance a treatment, as a supplement to Social Security financing), in general terms takes a more complementary role (i.e. becomes the principal financial sponsor). In France, this takes the form of a slow but effective withdrawal of the Social Security financing of so-called ambulatory or primary care, that is, everything ranging from a visit to the GP, to dental or eye care. In addition to the shortcomings in the regulation of private medicine costs, the main driver of this reconfiguration is the adoption of a cost containment policy for public healthcare expenditure. The change of policy generally originates from the new imperatives of European integration, whether in regard to the question of containing public expenditure as a result of the Maastricht Treaty and the introduction of the single currency (Palier 2017, pp.  55–56), or the question of competitiveness and the containment of labour costs (Barbier and Théret 2009, pp. 32–34). In practice, several estimates come together to show that chronic and hospital care are financed by Social Security up to approximately 90%, while the financing of ambulatory or primary care for the core population has been reduced and could fall to just over 50% (DG Trésor 2015; HCAAM 2011, p. 77, 2013, p. 203; Legal et al. 2009). Two aspects of this process interest us directly in the understanding of the “pillarisation” of the architecture of French health insurance as a system. On the one hand, the coherent existence of this system is based on the adoption of a theoretical and ideological underpinning by successive governments that has enabled them, without common party affiliation or overall unique reform, but by successive and incremental modifications, to justify and introduce the

300 

T. HOUSSOY ET AL.

privatisation of healthcare financing. On the other hand, the segmentation of the population and of the financing of healthcare generated by this theoretical and ideological underpinning is able to produce a form of institutional complementarity favouring the functioning of the health insurance market by removing the least solvent treatments and sectors of the public from its scope. Such a description closely matches what can be described as a “social liberal” or “New Keynesian” configuration, opposed to the social state logic that was prevalent in France for a long time after 1945 (Palier 2010; Ramaux 2012, pp.  173–178). This configuration can be described as a shifting of the role of government intervention towards a functional distribution of the roles fundamentally similar to that identified in the three-­ pillar pension model: public spending or regulation effectively play a role that is functionally equivalent to a correction of “market failures”. If we take this parallelism seriously, it should be emphasised that the types of failure generally highlighted by a large part of the health insurance economics are different from those in the case of pensions. First of all, the market failure usually put forward is not so much the result of an under-­ allocation (for retirement pensions, under-savings) but rather of an excess caused by the presence of risks of moral hazard on both sides of the exchange of medical goods or services: over-consumption of care goods and services (Pauly 1968) or physician-induced demand (Evans 1974). The second point is that both full insurance (i.e. no direct personal contributions in healthcare consumption) and the diversity of French-style insurance models (Dormont et  al. 2014; Franc 2017) are seen as sub-­ optimal. The gradual fragmentation of the health insurance system, to the point of now describing access to complementary cover as a “constellation” of measures (Del Sol 2014, p.  171), supports the idea that this institutional change cannot be reduced to the unilateral application of a programme dictated by an economics-based vision. It should also be borne in mind that the concept of moral hazard gradually took on the character of a theoretical and empirical failure, to the point where the financial contribution of the patient could be regarded as harmful to society, especially when it completely prevents access to treatment for the most serious illnesses (Nyman 1999a) or does not permit access to the necessary treatment for the worsened state of health of the poor (Nyman 1999b). While the analysis took a moralising turn in the face of this rather unique form of exchange, the analyses resulting from moral hazard survived in the form of “conventional wisdom” underlying the neo-liberal

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

301

policies attacking the European public health insurance systems. In this context, the desire to reduce public spending on health means increasing the contribution of patients, for which the economic theory emerging from M.  Pauly’s seminal work supplies a scientific credibility. However, the unequal and harmful consequences for the health of the population represented by such a strategy necessarily collide with ethical challenges, which are also supported by health economics. The privatisation of healthcare financing therefore involves drawing up a list of collective priorities, which took the form of an increase in the co-payment, sparing vulnerable groups and the most expensive care (Batifoulier 2012). It should therefore be noted that the constitution of legal, regulatory, tax and Social Security measures leading to the expansion of group health insurance can be interpreted as “opportunistic” measures, in the sense that they will only emerge during a broader discussion on the privatisation of healthcare financing without even really being the main subject. As discussed in the previous section, provisions favouring expansion by negotiation were made during the debates on the 1999 CMU law on universal healthcare coverage, principally intended to provide basic and complementary coverage to the poorest members of society. Likewise, the principal objective of the Douste-Blazy law of 2004 was to create a coherent system creating the figure of a patient who is a market player and the manager of his/her own health capital (Batifoulier et al. 2008). As well as supplementing the policy of access to complementary coverage for low-­ income households by supporting households that do not qualify for CMU-c, the social and tax benefits scheme it consolidates is above all a mechanism for expanding this type of scheme via state-approved complementary policies. The obligation to take out insurance via the company, although not incompatible with the implementation of this reform, is not necessary for the construction of the “autonomous” patient as the heart of this institutional change lies in the control and incentive mechanisms permitted by these policies. The systemic effect of the segmentation process produced by the concentration of the financing of national health insurance on major treatments and hospitalisation creates a form of institutional complementarity favouring private health insurance for ambulatory or primary care, focused on three main dimensions. According to an element of composition, by removing expenditure related to the most serious risks, this segmentation reduces the amount of the premiums (Legal et al. 2009, p. 68) and thus reduces the risk of not using private insurance. As well as increasing access

302 

T. HOUSSOY ET AL.

to healthcare, such a configuration makes it possible to increase the number of solvent beneficiaries on the most profitable part of the risk covered. According to an element of selection (i.e. the traditional element of adverse selection by the insurers), starting from an initial situation where a single form of coverage is provided by each insurer, making “major risk” private insurance coverage financially viable requires the possibility of capturing a large number of “healthy” subscribers to fund the insurance of the sick. Otherwise, the selection made possible by competition would work against such an approach. This selection can involve the insured “voting with their feet”29 or the diversification of policies by insurers in order to capture the “healthy” population. This mechanism would tend to reduce the coverage to the needs of this sub-population, thus aligning the premiums, and would influence the solvency of the insurance of chronic care by reducing the possibility of financial transfers between these two populations. Finally, according to an element of externalities, the guarantee offered by a high level of social management, of chronic care in particular, makes it possible to reduce the level of refusal of care of pathologies themselves able to cause an indirect rise in the use of “light” care. This leads to a reduction in negative externalities in the management of diseases most often reimbursed by private insurers, including when these insurers only cover ambulatory or primary care (Dourgnon et al. 2013). 4.2  A Policy Tool Reinforcing the Social Division of Welfare of the French Health Insurance System When R. Titmuss (1958) proposed one way of considering social protection beyond these public measures, his main concern was to show the many ways in which the social construction of the allocation of the resources intended to meet social needs greatly favours the middle and upper classes. Two central mechanisms must be added to obtain a more relevant picture. Firstly, the private initiative and the market form the dominant mechanism of the allocation of resources beyond the democratic debate and State allocation in Europe and in France. Next, by identifying fiscal and occupational welfare, R.  Titmuss also highlights their frequent association, amplifying the phenomenon of social division. His 29  The “healthy” population would tend to focus on those insurers that, specialising in covering them, would be able to offer lower premiums at the cost of excluding the more seriously ill.

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

303

original idea could also bring to light the link between this construction and the way in which the actions and interests of social groups attempt to acquire the resources intended to meet social needs, and thus to understand how their conflicts and distribution of power between them explain the form this system takes (Sinfield 1978). Without being able to achieve this final level of analysis, we argue that the new segmentation of healthcare financing, the increased number of mechanisms to access complementary coverage and the development of group coverage have reinforced the social division of the French health insurance system. Firstly, the privatisation of healthcare financing threatens the solidarity of the system. Adapting the Gini index to show the degree of difference between contributions and consumption per income decile, Jusot et al. (2017) show that most of the systemic solidarity is the product of national health insurance, due to its financing based on income and therefore unrelated to health needs. Conversely, complementary health insurance plays an insignificant role due to its predominantly flat-rate funding: only 10% of individual policies and 16% of group policies between 2011 and 2013 have a rate that is based on income (Montaut 2018). Indirect forms of solidarity through funding are mobilised, especially in favour of seniors (63% of individual policies) and families due to the free complementary coverage for some (89% of individual policies) or all children (59% of group policies), although these have very little impact on the systemic solidarity generated by complementary insurance. France is characterised by worrying levels of inequality in healthcare. On the face of it, we can expect that, in the event of illness, access to healthcare will be divided evenly according to socio-economic groups if it is accessible without barriers. However, an individual’s declared state of health tends to improve in line with income: in 2014, while 30.6% of the population declared themselves to be in a poor state of health,30 that represents 38.9% of the population of the first income quintile (between €0 and 952) and this share decreases with each quintile to represent 22.0% of the population of the last quintile. The result is that if each beneficiary meets their healthcare needs, then the distribution of access to healthcare must be more concentrated on disadvantaged people. However, in Europe, we are seeing inequalities in access to specialist care in favour of the 30  This term refers to all of the responses to the 2014 European Health Interview SurveyHealth, Health Care and Insurance Survey (EHIS-ESPS) survey formed from a declaration on the state of health corresponding to “quite poor”, “poor” and “very poor”.

304 

T. HOUSSOY ET AL.

wealthiest groups, and these inequalities are particularly marked in France: the richest 50% have twice as much access to specialists as the poorest 25% (Jusot 2013). This is as much the case for the annual probability of reliance on care as for the number of visits, and places France at the head of most of the ratings among the European countries, although it ranks lower for access to general practitioners. The performance of the French health system can be linked to problems with the regulation of private medicine prices: in 2017, extra fees represented 2.7% of the income of GPs (26.6% for Sector 231) but 17.7% for specialists (32.6% for Sector 2). In general terms, the probability of reliance on care and the number of visits to the doctor are significantly different, all things being equal (including monitoring the state of health), between socio-economic groups, both for general refusals and by type of care (Dourgnon et al. 2012). This is easy to see with the help of descriptive statistics (Table  10.1): in 2014, refusal of care systematically decreases when an individual has complementary coverage, and refusal for financial reasons is the leading cause. Thus, the process of the privatisation of the financing of routine healthcare is a factor in the potential development of healthcare inequalities, especially in Table 10.1  Percentage of the population who have reported to have foregone care in 2014, in accordance with their insurance situation Refusals of care reported General population

Private complementary

Complementary CMU

Without complementary

By type of care for financial reasons

16.8

15.3

21.7

40.1

10.1

9.0

16.4

25.1

5.2 25.0

4.4 23.1

6.8 33.2

19.6 53.0

15.9 2.9

15.7 2.6

19.2 7.1

16.5 3.6

By decisive factor

Dental care Optical care Doctors Financial reasons Delays Distance

Source: by the author from data of the 2014 EHIS-ESPS survey (Célant et al. 2017, pp. 132–145) Note: the data concerning refusals for financial reasons are reconstructed data.

 Price regulation in private medicine is divided into two sectors in France. “Sector 1” concerns private practice doctors who have agreed to apply the state health service contract. “Sector 2” concerns private practice doctors who have not agreed to apply this contract and can therefore apply a free pricing system. “Extra fees” are that part of the price of care over and above the agreement price. See Chap. 7 for further details. 31

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

305

France. The introduction of CMU-c can be interpreted in this way and was to some degree capable of curtailing this effect for the most disadvantaged populations, as is shown by the intermediate level of refusals of care observed among these beneficiaries. Likewise, it causes additional effects in terms of reducing the remaining amount payable by households with the Long-Term Conditions scheme (Dourgnon et al. 2013), assuming a large share of the expenditure on chronic care. The inequality in the issue of the financial accessibility of care resulting from the existence of the dependence of French households on complementary coverage is coupled with the quality of the reimbursements offered by this coverage. It stems from the segmentation created by the coexistence of individual and group coverage. The group coverage offered by the companies or branches is of systematically better quality than that available in the individual segment (Fig. 10.1). Where 30–40% of beneficiaries only of group policies after 2010 receive bad or average coverage (classes C to E), this rises to 90% in the individual segment. This initially stems from the favourable position group requests have over suppliers, even if they are only very basically organised when compared to individual 100% 90% 80%

2% 5%

3%

3%

4%

3%

8%

6%

7%

6%

37%

40%

22%

70%

39% 43%

44%

60% 50%

31%

30% 20% 10% 0%

29% 2006

54%

56%

53%

15%

9%

9%

13%

20%

28%

24%

26%

9% 8% 3% 1% 2010 2011

9% 2% 2012

3% 6% 2013

10%

43%

40%

53%

30%

19% 26%

26%

26% 21%

21%

19%

22%

2010

2011

2012

2013 E

D

6% 2006 C

B

A

Fig. 10.1  Changes in distribution by level of coverage of beneficiaries between 2006 and 2013. (Source: DREES (2016, p. 45), from data on the most popular policies, 2006 and 2013 editions.) Note: The policies are rated from the best coverage (A) to the worst (E). For the methodology, see Garnero and Le Palud (2014). The left-hand part corresponds to individual contracts, the right-hand part to group contracts.

306 

T. HOUSSOY ET AL.

70.00% 60.00% 48.70%

50.00%

51.00%

40.40%

40.00%

33.70%

30.00% 20.60%

20.00% 10.00% 0.00%

8.40%

≤ 650 €

651€ / 1000€ Collectif

1001€ / 1400€ Individuel

CMU-c

1401€ / 2000€

2001€/3000€

> 3000€

Sans complémentaire

Fig. 10.2  Rate of coverage by complementary health insurance, based on income per consumption unit and the type of coverage in 2012. (Source: DREES (2016, p. 53), from the 2012 ESPS survey, The Institute for Research and Information in Health Economics (IRDES.))

requests. This can also be seen in the distribution of policy expenses: in 2017, while policies had an average return on contributions in the form of benefits of around 79%, this hides the fact that this ratio rises to 85% for group policies and falls to 73% for individual policies (Adjerad 2019, p. 3). From the perspective of inequalities between socio-economic groups, this disparity in reimbursements favours the most advantaged income groups (Fig. 10.2); thus, only 8.4% of households with an income below €650 have group coverage and this share increases in line with income to reach 51% for households with an income over €3000. Thus, the segmentation of the complementary market between group and individual policies according to differences in the quality of reimbursements and access by the private salaried employee is broadly favourable to the wealthiest, especially managers. The withdrawal of part of the individual segment reduces the mutualisation margins for those people accessing it, which penalises salaried public employees (83% of individual coverage), the self-employed (78%) and, in particular, retirees (94%). To a large extent, the social division in access to healthcare stems from a private health insurance market that replicates the inequalities of the labour market. This can also be seen in private sector employment: in 2009, 66.3% of salaried employees with permanent contracts accessed complementary coverage through their company, as against 24.8% for

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

307

those on fixed-term contracts (Perronnin et al. 2012, pp. 74–75). Finally, to the de facto subsidising by the employer’s co-payment must be added the externalisation of the cost of the coverage provided by the preferential tax/Social Security system. Thus, at the macroeconomic level based on values for 2014, the Inspectorate General of Social Affairs (IGAS) has estimated that the public authorities spent, via aid and reduced income, 6.4–8.3 billion euros on purchasing private health insurance (Bruant-­ Bisson and Daude 2016, p. 81). In this respect, aid for group policies for companies represents 3.5 and 5.4 billion euros, just over half the total amount. At the microeconomic level, simulations make it possible to estimate that this tax/Social Security system could represent subsidisation of up to 25% of the total amount, unevenly split between the employer and the employee. While this ultimately represents reduced spending on salaries for the employer, the employee benefits from it at the cost of a lesser entitlement to national social welfare schemes owing to a reduction in the basis for calculating contributions, caused by substituting salary with a contribution to complementary insurance (Zemmour 2015).

5   The Implications of the Development of Occupational Health Insurance for Regulation by Collective Agreements: Reflections on the French Case The French Act of June 14, 2013, makes it mandatory for companies to provide complementary health insurance, thus institutionalising the status and role of collective insurance coverage. The fact that the law makes this mandatory makes the French situation unprecedented, to say the least. Furthermore, over and above its particularities, analysing the French case through the lens of occupational welfare is a way of examining the role of collective bargaining in the design of occupational insurance. Doing so also offers insights into how occupational welfare is deployed in the EU legal environment which, by various regulatory means, enshrines the right to collective bargaining. The fact that, in France, occupational welfare entails employers taking out an insurance policy also raises the question as to whether social partners are capable of having an impact in the health insurance market and regulating it (5.1).32 Ultimately, the core question  The issues are not the same when there is no requirement for insurance intermediation.

32

308 

T. HOUSSOY ET AL.

under discussion here is the collective, solidarity-based dimension of occupational insurance (5.2). 5.1  Collective Demand for Insurance: How Much Room Is There for Collective Bargaining? Positive EU law leaves plenty of room for collective bargaining in the realm of social protection (5.1.1). However, the fact that collective bargaining is acknowledged as being able to play a role in this field at EU level does not compel Member States to grant any particular amount of room for collective demand for insurance. The French case offers interesting evidence in this respect (5.1.2). 5.1.1

 ollective Bargaining in Social Protection: The View C from the Top In the well-known Dutch case of Albany (the setting up of sector-based pension funds via a collective agreement),33 the European Court of Justice (ECJ) took the view that the collective agreement setting up a complementary pension scheme addressed a social policy goal, since it was designed to ensure a set level of pensions for all workers within the occupational sector under consideration, thus directly contributing to the improvement of one of employees’ working conditions, that is, their compensation (albeit deferred compensation). In a subsequent ruling, a health care insurance scheme was analysed as contributing to the improvement of working conditions, by guaranteeing that the workers had “the necessary means to meet medical expenses but also by reducing the costs which, in the absence of a collective agreement, would have to be borne by the employees”.34 More recently, the same reasoning was applied in the case of a collective agreement concluded in France in the artisanal bakery sector setting up complementary health insurance, with the reimbursement of healthcare costs “contributing to improved working conditions”.35 From this, it follows that just like public authorities, social partners can achieve a degree of regulation in the field of social protection by negotiating collective agreements (Del Sol 2015). To do so, they have full  ECJ case C-67/96, Albany, September 21, 1999 (point 59), ECR 1999, p. I-5751.  ECJ case C-222/98, Van der Woude, September 21, 2000 (point 25), ECR 2000, p. I-7111. 35  ECJ case C-437/09, AG2R, March 3, 2011, ECR 2011, p. I-973. 33 34

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

309

autonomy to conclude an agreement and determine contents, since nothing in EU economic law requires them to adopt a particular definition of welfare benefits or determine the characteristics of the social protection scheme they intend to set up in any particular way.36 They may undertake minimal bargaining, or be more ambitious as to the type of benefits, the terms on which they are allocated, how much is allocated, and how they are funded. This jurisprudence is all the more applicable in that article 28 of the Charter of Fundamental Rights, adopted in 2000, which states that “Workers and employers … have … the right to negotiate and conclude collective agreements at the appropriate levels…” This relevance has been significantly heightened since the right to collective bargaining achieved the status of fundamental freedom in EU law. Indeed, when the Treaty of Lisbon was signed in 2009, the Charter of Fundamental Rights was granted the same legal value as the treaties themselves (art. 6§1 Treaty on the Functioning of the European Union (TFEU)). 5.1.2

 ollective Bargaining in Social Protection: The View C from the Bottom Despite a favourable EU legal environment, it cannot be asserted that collective bargaining is a “natural” vector for setting up social protection schemes in the occupational sphere, for instance to express a collective demand for health insurance. In reality, the role of collective bargaining in the field of social protection depends on parameters that are defined at the national level. The French case constitutes a topical example of this issue. In 2013, a law was passed that enshrined the right to occupation-based complementary health insurance. It thus made this right dependent on a particular employment status and relationship, that is, being an employee of a private sector company. However, the occupational origins of this right do not automatically entail the coverage being collective in nature. French regulations are ambiguous in this respect. Some aspects of the applicable legal framework could be understood as minimising the collective aspect, or the significance thereof. The following analysis is focused on the situation in France; nevertheless, it also raises broader issues whose implications could also emerge in other countries: firstly, the issue of potential competition 36  However, the principles of non-discrimination and equality of treatment are also binding on social partners, as well as on the authorities.

310 

T. HOUSSOY ET AL.

between the legal vectors used to deploy occupational insurance; and secondly, the key issue of the degree to which regulation via collective bargaining can or may intervene. Collective agreements as a vector of occupational social protection—exclusively, or in competition? For the purposes of deploying mandatory complementary health insurance, French law has prioritised implementation by means of collective bargaining. As a result, social partners in the relevant branch (sector of business) are the first to be required to deal with this new item in bargaining; if no agreement is reached at branch level, the law invokes the setting up of complementary health insurance by means of bargaining at company level. However, branch or in-company negotiators are not bound by any obligation of results (i.e. concluding an agreement); they are bound only to a best effort undertaking. If no agreement is reached, the obligation to have insurance will be met by means of a unilateral decision on the part of the employer. Of course, any such decision does not rule out there having been prior consultation with employee representatives, but there is no legal requirement for any such consultation; neither is it consubstantial with this form of commitment. Determining the contents of health insurance is not put up for discussion with employee representatives, let alone the issue of having any solidarity parameters. In this case, the collective dimension is quantitative (the number of people insured) and not qualitative. Moreover, this legal procedure is not agnostic in terms of who is actually covered. It must be combined with older legislation: France’s Evin Law of December 31, 1989. Article 11 of this Law specifies that no employee present in the company at the time when an employer takes a unilateral decision to implement collective coverage “may be compelled against their will to pay the relevant contributions…”. All such employees are thus granted the right to refuse insurance if they are charged for part of its funding. A survey conducted to evaluate the effects of the 2013 Act for the first time highlights the finding that in fact, almost half of all newly insured companies resorted to a unilateral decision by the employer (DREES-IRDES 2019). In these firms, the insured group may shrink due to a number of employees having opted not to sign up for the occupational insurance coverage. Contractual regulation—by sector or by company? In terms of social protection, it could be thought that negotiating a collective agreement is, other things being equal, the best way to bring the qualitative parameters

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

311

of coverage into the equation. However, resorting (whether exclusively or otherwise) to collective bargaining is not sufficient to take the collective dimension into account. Indeed, it must be emphasised that the level at which the agreement is concluded and its scope (sector of business or firm) may have an impact. Historically, in France, the law has entrusted occupational branches (or sectors of business) with a key role in regulating complementary social protection, and in framing the standards that emerge at company level. While companies were left room to implement some degree of regulation, in most cases, this regulation has been secondary to that of the branch, or dependent on how the branch was regulated. More recent legislation (including an order dated September 22, 2017) has however led to this state of affairs being partially called into question, with regulation negotiated at company level becoming more autonomous in legal terms, including in the realm of top-up social protection (Vincent 2019a, b; Rehfeldt and Vincent 2018). Moreover, it should be noted that this phenomenon of decentralisation of collective bargaining is a trend that may be observed in many EU countries, all the more so since the 2000s (European Commission 2012; Koukiadaki et  al. 2016; Müller et  al. 2019). The resulting retreat of sector-based regulation is not without its effects, irrespective of the subject under negotiation; indeed, it leads to a form of de-collectivisation (Laulom 2017a, b). This effect is all the more acute when it comes to social protection, since decentralisation carries with it the seeds of a threat to both the scope and the degree of solidarity (see below)—and these are consubstantial components in any social protection scheme. Some closing thoughts. At the time of the 2013 reform, France did not equip itself with any means to “counter” insurance market forces by giving any real preference to the demand for insurance from broad groups. This may come as a surprise inasmuch as the authorities have historically offered significant incentives to encourage the development of complementary social protection in the occupational sphere (see above). For instance, in 2003, the law opted to reserve such incentives for mandatory affiliation schemes for employees that were collective, that is, benefiting all employees, or objectively determined categories of employee. Indeed, mandatory collective insurance cannot be an emanation of insurance operators left to themselves. It can only be an emanation of the

312 

T. HOUSSOY ET AL.

State37 or social partners, via collective bargaining. It may therefore appear paradoxical that French law has not granted exclusive accreditation to social partners to deploy mandatory complementary health coverage benefiting employees. Indeed, this type of accreditation would be the only legal avenue via which a collective demand for insurance could be established. However, in application of the 2013 Act, in the absence of any collective branch or company agreement, implementation is by means of a unilateral decision by the employer;38 this does not express any collective demand. In other words, occupational health insurance does not necessarily result in a coverage that has been established collectively. The unilateral decision means that any policy taken out will cover only a specific set of employees. In one sense, this echoes the definition of “group insurance policy” given by the French Insurance Code (Code des assurances): a policy taken out by a legal person or head of establishment for the purposes of affiliating a number of individuals (subscribers) that must have a link of the same nature with the policyholder. The confines of the group are defined by the identity of the link with the policyholder and not the links and/or interests that may unite the members of the group in question. By contrast, the vocation of a collective agreement is to cover a group of employees whose interests are taken into account by the employee negotiators with a view, in principle, to seeing the general interest of the group emerge in the negotiating process.39 5.2  Solidarity-based Demand for Insurance: How Much Autonomy Do Collective Agreements Provide? Since the Law of June 14, 2013, the role of collective occupational insurance in France has become institutionalised (see above). However, it should be borne in mind that occupation-based health insurance has long formed part of the market, more specifically the health insurance market. This raises the question of the extent to which collective agreements can affect competition between insurance companies on grounds of solidarity.  For instance, when they entrust a Social Security scheme to private-law bodies.  The first data from after the 2013 Act emphasises the fact that the establishment of topup health cover is not negotiated as a priority. It is frequently implemented by a decision on the part of the employer, in particular in very small establishments. See DREESIRDES (2019). 39  On the difficulty of representing interests that can be divergent on a larger scale than that of the company or branch, see Ebbinghaus (2006). 37 38

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

313

The issues. In a competitive system, demand (in theory) seeks to benefit from there being more than one supplier to obtain the best price (the concept of economic efficiency). Solidarity-based goals are therefore not defining features of this system; however, the latter does not necessarily ignore them altogether, either. The same therefore applies to EU competition law, which allows for adjustments of free competition in the presence of services of general economic interest (hereafter, SGEIs), for instance in the form of selective benefits or exclusive rights for certain suppliers. Thus, for example, in the field of social protection, the ECJ acknowledges the existence of SGEIs, in particular if solidarity-based elements are imposed on the managing operators of the schemes in place (see Chap. 4). Indeed, it is generally argued that if the scheme is characterised by a degree of solidarity that imposes significant obligations on the managing insurer, the service the latter provides cannot be made available at the same price as that of other operators that are not subject to the same constraints. In other words, this insurer is less competitive on the insurance market because solidarity has a cost.40 This raises the issue of exclusive management rights being granted to a given insurance company. Indeed, these rights are the main legal vehicle by means of which solidarity can be introduced in the field of insurance. This is permitted by positive EU law. The latter allows for priority to be granted to a form of social protection that is partly protected from market forces (Laigre 1996; Driguez 2006; Del Sol 2016). When this paradigm shift occurred in 2013, France seems to have missed an opportunity to establish private collective insurance incorporating a high degree of solidarity. The main reason for this is to do with the way the right to collective bargaining has been made somewhat secondary, since collective agreements can no longer assign a management monopoly to an insurer. How collective bargaining rights have become secondary in the French reform. Collective agreements give social partners the ability to establish a health insurance regime that is structurally solidarity-based and fulfils the criteria to qualify as an SGEI.  The degree of differentiation from the 40  In the Albany case (previous ruling), the sector-based complementary pension scheme managed exclusively by a pension fund was characterised, in the Court’s own words, by a high degree of solidarity. This was due, in particular, to the independence of contributions from the risk, the obligation to accept all workers without any prior medical examination, pensions continuing to be built up even without further contributions being paid in the event of incapacity for work, and the fund taking into account outstanding contribution payments due from the employer in the event of the latter’s bankruptcy.

314 

T. HOUSSOY ET AL.

operation of “traditional” private insurance present on the market is what makes it possible to assess the implementation of the solidarity principle and thus provide justification for adjusting these rules of competition. The collective agreement should therefore not simply consist of a set of guarantees, but rather establish a fully fledged scheme. The rules configuring the degree of proportionality that is normally established in insurance between contributions and risks, as well as between contributions and benefits, therefore become crucial; everything thus depends on the degree to which this configuration alters the traditional workings of risk-based insurance. This is, in substance, what emerges from the AG2R ruling (op. cit.) in which the ECJ refers to a number of factors: funding by means of fixed-rate contributions, such that the rate is not proportional to the insured risk, with the result that age, state of health and particular risks inherent in the position occupied are not taken into consideration;41 the type of benefits and extent of coverage not being proportional to the amount of contributions paid; the existence of “free” entitlements not directly related to a contribution (e.g. the temporary maintenance of coverage for heirs in the event of the death of the employee). If the collective agreement is concluded at the level of a branch and management of the social protection scheme has been outsourced to an insurance body, the general-interest missions are usually accomplished by granting exclusive management rights to the insurer, failing which the economic conditions would not be acceptable for it and in some cases could threaten its financial equilibrium. To facilitate this, positive EU law allows for the collective agreement itself to grant a management monopoly to an insurance company. This monopoly may thus arise from the inclusion in the collective agreement of a clause designating the insurance company that will manage the negotiated scheme; this clause then requires companies in the branch to take out a collective insurance policy with the designated insurer. However, in 2013, the French Constitutional Council ruled the contractual clauses on designation invalid. In doing so, it closed the door to the establishment of collective insurance that was solidarity-based in structural terms, on the grounds of the freedom of contract for employers, as well as the freedom of enterprise (Kessler 2015). Indeed, henceforth, social partners in a branch can no longer impose the insurance body on 41  In other words, pricing was not determined on the basis of each employee’s loss experience.

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

315

companies in the sector of business in question. At best, the collective agreement may include a recommendation clause, but this nevertheless allows firms full freedom to choose the insurance company—the one recommended by the branch, or another insurer. As a result, it is no longer possible to create genuine occupational social protection schemes, since no insurance company would agree to ensure a strongly solidarity-based scheme without being able to count on every firm within the branch. The right to collective bargaining is therefore limited, and in a certain sense relegated, with the employer’s freedom of contract being granted superior status. This leads to a paradoxical state of affairs: while the specifically social function of the designation clauses makes them fully compliant with EU competition law, they have disappeared from French law because they run counter to the French constitutional principle of freedom of contract. Moreover, if there is no branch agreement, or the agreement does not have a recommendation clause, the concept of pooling between firms within the occupational sector in question also runs into difficulties. Firms are thus subject to the law of the market and nothing else and have to take out a policy directly from an insurance company. Given this state of affairs, when establishing pricing for the policy, the insurance company will perform an actuarial analysis of the risks, taking into account factors such as the size of the company, its loss experience (e.g. the rate of occupational accidents and diseases), absenteeism and certain demographic factors (such as sex, age and state of health). Firms may be faced with a range of rates for equivalent levels of coverage. As a result, firms with low bargaining power (often the case for smaller companies) are penalised, since no pooling with larger firms at branch level is organised. Ultimately, this penalisation may impact employees, since their share of the funding may be higher, or the levels of coverage lower, in order for the cost to be affordable. As C. Vincent has quite rightly pointed out, “the practical conditions in which complementary health insurance has been implemented have opened up the way for the health risk market to be reconfigured. Local, autonomous regulation is taking the place of historic national governance” (Vincent 2019a, b). A potential challenge to the relegation of collective bargaining rights? In the realm of top-up health insurance, the disappearance of designation clauses would appear to sound the death knell of the idea of an entire occupational branch approaching private social insurance companies. The ruling by the French Constitutional Council also “mistreats” the right to

316 

T. HOUSSOY ET AL.

collective bargaining by restricting the freedoms of social partners. However, the European legal environment has enshrined the right to collective bargaining via a variety of legal vectors (Dorssemont et al. 2019; Bruun et al. 2016; Dorssemont 2016; Jacobs 2013). Firstly, article 28 of the Charter of Fundamental Rights of the EU recognises the right of collective bargaining and action (Dorssemont and Rocca 2019); secondly, article 6 of the European Social Charter asserts “the right of workers and employers to collective action in cases of conflicts of interest”;42 in principle, this compels the large number of Member States who have signed or ratified the Charter to adopt the necessary measures to ensure that this right can actually be exercised.43 Indeed, it is on the grounds of a breach of this article 6 that a French trade union federation, Force Ouvrière, decided to bring a case before the European Committee of Social Rights (hereafter, ECSR) as part of a collective complaint. The trade union argued that the fact that social partners cannot make use of designation clauses of which the purpose is to pool risks at branch level constitutes unjustified interference in the right to collective bargaining and is detrimental to the right of social partners to negotiate freely. The ECSR upheld the trade union’s argument, taking the view that the prohibition of designation clauses was in breach of article 6§2 of the Charter.44 In particular, it specified that there was “no fundamental reason to uphold freedom of contract on the right to collective bargaining” (§68). As one author put it, “the ECSR ruling establishes equivalency between economic freedoms and the right to collective bargaining”; “by positioning itself in such a way as to focus on an assessment of the seriousness of the legislative measure in terms of its impact on the right to collective bargaining, the ECSR has not addressed the question in mere technical terms, but rather in terms of values” (Ginon 2019). This does not grant social partners carte blanche though, because it does not establish any precedence of the right to collective bargaining over economic freedoms either. Everything is a matter of proportionality. The ECSR’s assessment of proportionality has led to it emphasising that “the designation of an  It is worth noting that article 6 is in the part of the Charter entitled “Solidarity”.  This Charter has been adopted, not by the European Union, but by the Council of Europe. It is controlled by the European Committee of Social Rights by means of two mechanisms (national reports and collective complaints). However, the rulings handed down by the Committee are not enforceable in domestic legal orders. 44  ECSR, July 3, 2018, Confédération Générale du Travail Force Ouvrière (FO) versus France, complaint no. 118/2015. 42 43

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

317

insurer by the social partners was a mechanism based on the principle of solidarity” (§72), contrary to the mechanism of the recommendation, which “leads to a complementary ‘two-track’ social protection scheme” (§73).45 Similarly, to the ECJ jurisprudence on SGEIs in terms of competition law, the ECSR has staked out a possible route towards balancing market principles and social interests. The issue here is to avoid a form of Balkanisation of occupational complementary social protection schemes46 liable to penalise the most vulnerable employees and the smallest firms. In France, it is now up to the authorities to alter the trajectory set by the Constitutional Council’s ruling. It would appear that an alternative path can be traced. However, this would entail taking the ECSR’s ruling, which is not binding, into account, and in doing so, adding grist to the social partners’ mill. Such an outcome appears unlikely. Indeed, the 2013 Act has in fact established a right to complementary health insurance for employees in the private sector. This right is a means of redistributing the public and private sector funding of health insurance for a large part of the population. Seen thus, the ambiguous role assigned to collective bargaining is doubtless not as paradoxical as it might appear at first sight; promoting an aspect of collective bargaining was not in fact its intended aim.

6   Conclusion Despite the fact that the vast majority of individuals (more than 95%) have complementary health coverage in France, the Government has made complementary insurance compulsory for a  segment of the population (salaried employees in the private sector) although it has traditionally been voluntary. This development reveals the heights reached by the marketisation of health insurance as the requirement to take out private health coverage is now institutionalised. As a result, this extension of complementary coverage could be seen as contributing to the privatisation or commodification process of health insurance. This type of analysis, commonly used, takes the public-private mix as its point of reference and provides 45  The Committee criticised the very broad nature of the prohibition against designation clauses. It took the view that doing so failed to take into account “certain branches with specific needs, presenting ‘bad risks’ or activities where employees frequently change employers” (§74). 46  One author talks in terms of “a myriad of independent micro-perimeters” (Petit 2019).

318 

T. HOUSSOY ET AL.

information on the role of the State, the existence of non-State players (such as firms, social partners and insurers) and the place of the market. Such a point of reference could be used for all social risks. It also facilitates a comparative approach between national systems as it ultimately enables a sort of architectural modelling of the systems. In view of the very high level of complementary coverage that existed prior to the 2013 law, however, this marketing does not significantly alter the balance between public and private insurance. On the other hand, by choosing to favour group insurance and requiring employers to provide at least half of the funding for the health coverage of their staff, one enters the domain of occupational welfare which is consequently positioned on the market. This marketing dimension created by the 2013 law has a certain uniqueness. This chapter therefore took the option of employing those studies dedicated to occupational welfare. We aimed at providing an analytical framework that makes it possible to grasp the complexity and the institutional complementarities between the different components of a system. It required using a framework seen as more “receptive” to complex and original institutional configurations. By placing occupational welfare schemes at the centre of the analytical framework (here, in the area of health coverage), it is possible to examine several aspects of the system: the role of the State (if necessary via tax incentive policies promoting occupational insurance coverage) and of the market; the distribution of roles in meeting the social requirement to pay for health care; the systemic effects of institutional arrangements in terms of solidarity. In our view, this justifies the preference for the expression “welfare mix” rather than “public-­ private mix”. Thus, while the existence of non-State players can be an indicator, or even a criterion, of the privatisation of financing, it can also reveal a lot more about the system when it is combined with other aspects. For example, some non-State players can, through their involvement or by “allocation”, represent alternatives between the State and the market. However, this analytical framework is not well suited to the modelling of national systems. On the other hand, with its multi-dimensional interpretation of a system, it can make its realities more intelligible and consequently provide general keys to understand of a particular case. Thus, the evolution of the French case cannot be reduced to new balances within the health insurance market between individual and group coverage. On the contrary, the analysis performed makes it possible to grasp the issues and systemic effects of the reconfiguration that has taken place. It underlines the way in which it affects the welfare mix and its capacity to satisfy the

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

319

need for the provision of care. The occupational welfare prism is of relative interest in understanding the financing privatisation process. It is, however, particularly relevant when it comes down to analysing the very substance of a social protection system, especially in relation to health coverage. What is true for the analysis of the French case is also true for other national systems, despite the notable differences between them. Indeed, the health problem is universal and can be viewed independently of the employment status. It is connected to public health issues, to the interactions with healthcare service providers and to questions of both horizontal (between the healthy and the sick) and vertical (between income groups) redistribution. The mobilisation, or even instrumentalisation, of professional (and therefore category-based) insurance coverage is therefore not a neutral policy choice, regardless of whether the choice is strategic or more accidental. If we look beyond the analytical aspect alone, the French case is a strong indicator of the solidarity issues that can be highlighted by occupational welfare schemes. They force us to (re-)consider the question of solidarities, which is important or even central, when protection against health risks is at stake. We must first take a look at occupational welfare, as a component of the system; a look that will in a way be introspective. Unlike public Social Security schemes, occupational welfare schemes are not, by definition, structured by a principle of solidarity. On the other hand, by construction, they can be infused with a solidarity whose form, extent and degree are all variables. The question of solidarity does not stop at the borders of the professional “pillar”. In the field of health coverage, occupational welfare schemes certainly have the potential to be established between the State and the market and even to be an alternative to the market: they can instil solidarity, unlike individual private health insurance which would be barely regulated  in that direction, without being able to be as ambitious as a compulsory legal Social Security scheme based on a principle of national solidarity. As a result, it is the solidarity of the system as a whole that is impacted by a growth of  occupational welfare  schemes. The view taken must be quite panoramic in order to appreciate the links between the various components of the system and the distribution of solidarity produced by the institutional arrangements. Yet, in terms of solidarity, these arrangements will produce effects that are necessarily dependant on choices related to occupational welfare. This could result in welfare mix configurations that are extremely variable from one system to the next with the

320 

T. HOUSSOY ET AL.

(apparent) paradox that a high level of solidarity within the occupational “pillar” could destabilise the systemic solidarity.

References Adjerad, R. (2019). Complémentaire santé: la hausse modérée de la part des contrats collectifs se poursuit en 2017. Études & Résultats, 1112. Paris: Direction de la Recherche, des Études, de l’Évaluation et des Statistiques (DREES). Amable, B. (2004). The Diversity of Modern Capitalism (326 p). Oxford: Oxford University Press. Amable, B., & Palombarini, S. (2009). A Neorealist Approach to Institutional Change and the Diversity of Capitalism. Socio-Economic Review, 7(1), 123–143. André, C., Batifoulier, P., & Jansen-Ferreira, M. (2016). Health Care Privatization Processes in Europe: Theoretical Justifications and Empirical Classification. International Social Security Review, 69(1), 3–23. Aoki, M. (2001). Toward a Comparative Institutional Analysis (480 p). Cambridge, MA: MIT Press. ApRoberts, L. (1996). Les acteurs sociaux selon la doctrine de la Banque Mondiale. La Lettre de l’Observatoire des Retraites, 8(9). Barbier, J.-C., & Théret, B. (2009). Le système français de protection sociale (128 p). Paris: La Découverte (Repères). Batifoulier, P. (2012). De la responsabilité individuelle aux priorités collectives. Une analyse institutionnaliste des politiques de la demande de santé. Éthique et économique/Ethics and Economics, 9(2), 25–44. Batifoulier, P., Domin, J.-P., & Gadreau, M. (2007). La gouvernance de l’assurance maladie au risque d’un État social marchand. Économie Appliquée, 60(1), 101–126. Batifoulier, P., Domin, J.-P., & Gadreau, M. (2008). Mutation du patient et construction d’un marché de la santé. L’expérience française. Revue Française de Socio-Économie, 1, 27–46. Bönker, F. (2005). Changing Ideas on Pensions: Accountig for Differences in the Spread of the Multipillar Paradigm in Five EU Social Insurance Countries. In P.  Taylor-Gooby (Ed.), Ideas and Welfare State Reform in Western Europe. Hampshire: Palgrave Macmillan. Bruant-Bisson, A., & Daude, M. (2016). Contribution au rapport au Parlement sur les aides fiscales et sociales à l’acquisition d’une complémentaire santé, 2015-143R. Paris: Inspection Générale des Affaires Sociale (IGAS). Bruun, N., Lörcher, K., Schömann, I., & Clauwaert, S. (Eds.). (2016). The European Social Charter and the Employment Relation (536 p). Oxford: Hart Publishing. Castel, R. (2002). From Manual Workers to Wage Laborers: Transformation of the Social Question (496 p). Piscataway: Transaction Publishers.

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

321

Célant, N., Guillaume, S., & Rochereau, T. (2017). Enquête Santé Européenne— Enquête Santé et Protection Sociale 2014 (EHIS-ESPS), 566. Paris: Institut de Recherche et de Documentation en Économie de la Santé (IRDES). Cutler, T., & Waine, B. (2001). Social Insecurity and the Retreat from Social Democracy: Occupational Welfare in the long boom and financialization. Review of International Political Economy, 8(1), 96–118. De Deken, J. (2018). The Corrosion of Occupational Pensions Solidarity in the Netherlands. Transfer: European Review of Labour and Research, 24(1), 43–56. https://doi.org/10.1177/1024258917748274. Deeg, R. (2007). Complementarity and Institutional Change in Capitalist Systems. Journal of European Public Policy, 14(4), 611–630. Del Sol, M. (2014). Généralisation de la couverture complémentaire santé des salariés : éléments de controverse. Droit social, 165–173. Del Sol, M. (2015). Droit européen et protection sociale négociée: mesure de l’autonomie des partenaires sociaux. In P. Turquet (Ed.), La crise de la protection sociale en Europe. Adaptation ou refondation (pp. 95–105). Rennes: Presses Universitaires de Rennes. Del Sol, M. (2016). Protection sociale: le droit de l’Union européenne entre monopole et concurrence. In M.  Borgetto, A.-S.  Ginon, & F.  Guiomard (Eds.), Quelles(s) protection(s) sociale(s) demain ? (pp. 107–125). Paris: Dalloz (Thèmes et commentaires). Dormont, B., Geoffard, P.-Y., & Tirole, J. (2014). Rebuilding the Health Insurance System. Publication web sérielle. Notes du CAE, 42. Paris: Conseil d’Analyse Economique (CAE). Dorssemont, F. (2016). The Right to Bargain Collectively: A Matrix for Industrial Relations. In N. Bruun, K. Lörcher, I. Schömann, & S. Clauwaert (Eds.), The European Social Charter and the Employment Relation. Oxford: Hart Publishing. Dorssemont, F., Lörcher, K., Clauwaert, S., & Schmitt, M. (Eds.). (2019). The Charter of Fundamental Rights of the European Union and the Employment Relation (712 p). Oxford: Hart Publishing. Dorssemont, F., & Rocca, M. (2019). Article 28—Right of Collective Bargaining and Action. In F. Dorssemont, K. Lörcher, S. Clauwaert, & M. Schmitt (Eds.), The Charter of Fundamental Rights of the European Union and the Employment Relation. Oxford: Hart Publishing. Dourgnon, P., Jusot, F., & Fantin, R. (2012). Payer nuit gravement à la santé : une étude de l’impact du renoncement financier aux soins sur l’état de santé. Économie publique/Public economics, 28–29, 123–147. Dourgnon, P., Or, Z., & Sorasith, C. (2013). The Impact of the Long-term Illness Scheme (LTI) on Inequalities in the Utilisation of Ambulatory Care Between 1998 and 2008. Publication web sérielle, Question d’économie de la santé, 183.

322 

T. HOUSSOY ET AL.

Paris: Institut de Recherche et de Documentation en Économie de la Santé (IRDES). DREES. (2016). Les complémentaires santé. Acteurs, bénéficiaires, garanties. Édition 2016, Panoramas de la DREES, S 16-070. Paris: Direction de la Recherche, des Études, de l’Évaluation et des Statistiques (DREES). DREES-IRDES. (2019). Enquête PSCE—Protection sociale complémentaire d’entreprise. Paris: Direction de la Recherche, des Études, de l’Évaluation et des Statistiques (DREES). Dreyfus, M. (2009). L’émergence tardive des assurances sociales en France. In M.  Dreyfus (Ed.), Les assurances sociales en Europe (pp.  79–129). Rennes: Presses Universitaires de Rennes (Pour une histoire du travail). Driguez, L. (2006). Droit social et droit de la concurrence. Bruxelles: Bruylant. Ebbinghaus, B. (2006). Trade Union Movements in Post-industrial Welfare States: Opening up to New Social Interests? In K. Armingeon & G. Bonoli (Eds.), The Politics of Post-industrial Welfare States: Adapting Post-war Social Policies to New Social Risks (pp. 123–142). London: Routledge. Ebbinghaus, B. (2011a). Introduction: Studying Pension Privatization in Europe. In B.  Ebbinghaus (Ed.), The Varieties of Pension Governance: Pension Privatization in Europe (pp. 3–22). Oxford: Oxford University Press. Ebbinghaus, B. (Ed.). (2011b). The Varieties of Pension Governance: Pension Privatization in Europe (440 p). Oxford: Oxford University Press. European Commission. (2012). Labour Market Developments in Europe 2012. European Economy, 5. Brussels. Evans, R. G. (1974). Supplier-Induced Demand: Some Empirical Evidence and Implications. In M. Perlman (Ed.), The Economics of Health and Medical Care: Proceedings of a Conference held by the International Economic Association at Tokyo (pp.  162–173). London: Palgrave Macmillan UK (International Economic Association Series). Farnsworth, K. (2004). Welfare through Work: An Audit of Occupational Social Provision at the Turn of the New Century. Social Policy & Administration, 38(5), 437–455. Farnsworth, K. (2012). Occupational Welfare. In B. Greve (Ed.), The Routledge Handbook of the Welfare State (pp.  30–39). Oxon: Routledge (Routledge Handbooks). Franc, C. (2017). Le partage de la couverture maladie entre assurances obligatoire et complémentaires—Les défauts d’un système mixte. Médecine/Sciences, 33(12), 1097–1104. Friot, B. (2012). Puissances du salariat (438 p). Paris: La Dispute (Travail et salariat). Garnero, M., & Le Palud, V. (2014). Les contrats les plus souscrits auprès des complémentaires santé en 2010. Document de travail. Série Statistiques, 191.

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

323

Paris: Direction de la Recherche, des Études, de l’Évaluation et des Statistiques (DREES). Ginon, A.-S. (2019). Réflexions sur les clauses de désignation: du Conseil constitutionnel au Comité européen des droits sociaux. Revue de droit sanitaire et social, 331–343. Goodin, R. E., & Rein, M. (2001). Regimes on Pillars: Alternative Welfare State Logics and Dynamics. Public Administration, 79(4), 769–801. Greve, B. (2007). Occupational Welfare. Winners and Losers (208 p). Cheltenham: Edward Elgar Publishing. Greve, B. (2018). At the Heart of the Nordic Occupational Welfare Model: Occupational Welfare Trajectories in Sweden and Denmark. Social Policy & Administration, 52(2), 508–518. Hacker, J. S. (2002). The Divided Welfare State. The Battle over Public and Private Social Benefits in the United States (466 p). NewYork: Cambridge University Press. Hacker, J. S. (2004). Privatizing Risk without Privatizing the Welfare State: The Hidden Politics of Social Policy Retrenchment in the United States. The American Political Science Review, 98(2), 243–260. Hall, P. A., & Soskice, D. (2001). Introduction. In P. A. Hall & D. Soskice (Eds.), Varieties of Capitalism: The Institutional Foundations of Comparative Advantage (pp. 1–68). Oxford: Oxford University Press. Hassenteufel, P., & Palier, B. (2007). Towards Neo-Bismarckian Health Care States? Comparing Health Insurance Reforms in Bismarckian Welfare Systems. Social Policy & Administration, 41(6), 574–596. Hatzfeld, H. (1971). Du paupérisme à la Sécurité sociale, 1850–1940. Nancy: Presses Universitaires de Nancy. Häusermann, S. (2010). The Politics of Welfare State Reform in Continental Europe. Modernization in Hard Times. Cambridge: Cambridge University Press (Cambridge Studies in Comparative Politics). HCAAM. (2011). L’assurance maladie face à la crise. Mieux évaluer la dépense publique d’assurance: l’ONDAM et la mesure de l’accessibilité financière des soins. In HCAAM (Ed.), Rapport d’activité. Paris: Haut Conseil pour l’Avenir de l’Assurance Maladie (HCAAM). HCAAM. (2013). Rapport d’activité 2013. Paris: Haut Conseil pour l’Avenir de l’Assurance Maladie (HCAAM). Howard, C. (1997). The Hidden Welfare State: Tax Expenditure and Social Policy in the United States (266 p). Princeton: Princeton University Press (Princeton Studies in American Politics: Historical, International, and Comparative Perspectives). Immergut, E. M., Anderson, K. M., & Schulze, I. (Eds.). (2006). The Handbook of Western European Pension Politics (965 p). Oxford: Oxford University Press.

324 

T. HOUSSOY ET AL.

Jacobs, A. (2013). The Right to Bargain Collectively under Article 11 ECHR. In F. Dorssemont, K. Lörcher, & I. Schömann (Eds.), The European Convention on Human Rights and the Employment Relation. Oxford: Hart Publishing. Järvi, L., & Kuivalainen, S. (2013). Equally to All? The Significance of Collectively Negotiated Sickness Benefits in Cross-Nordic Social Policy Analysis. European Journal of Social Security, 15(4), 341–357. Johnson, N. (1999). Mixed Economics of Welfare: A Comparative Perspective (309 p). London: Prentice Hall Europe. Johnston, A., Kornelakis, A., & d’Acri, C.  R. (2011). Social Partners and the Welfare State: Recalibration, Privatization or Collectivization of Social Risks? European Journal of Industrial Relations, 17(4), 349–364. Jusot, F. (2013). Les inégalités de recours aux soins: bilan et évolution. Revue d’Épidémiologie et de Santé Publique, 61, S163–S169. Jusot, F., Legal, R., Louvel, A., Pollak, C., & Shmueli, A. (2017). Public and Private Health Insurances: How do they Contribute to Social Solidarity. Publication web sérielle, Question d’économie de la santé, 225. Paris: Institut de Recherche et de Documentation en Économie de la Santé (IRDES). Kerleau, M. (2009). Le nouveau cadre institutionnel de la protection sociale complémentaire d’entreprise : quels enjeux pour les mutuelles santé ? Revue internationale de l’économie sociale (RECMA), 312, 15–34. Kerleau, M. (2012). De la couverture maladie universelle aux politiques d’accès à l’assurance-maladie complémentaire : diversité des modèles et des protections. Revue Française de Socio-Économie, 9, 171–189. Kerleau, M., Durand, F., Fretel, A., & Hirtzlin, I. (2008). Pratiques et enjeux autour de la protection sociale complémentaire d’entreprise. Tome 1. Rapport de recherche, Mire “Gouvernance de la protection sociale”. Paris: Centre d’Économie de la Sorbonne. Kessler, F. (2015). L’émergence (trop tardive ?) de la notion de régime de protection sociale complémentaire. Revue de droit sanitaire et social, 352–363. Koukiadaki, A., Tavora, I., & Martinez Lucio, M. (Eds.). (2016). Joint Regulation and Labour Market Policy in Europe During the Crisis. Brussels: European Trade Unions Institute (ETUI). Laigre, P. (1996). L’intrusion du droit communautaire de la concurrence dans le champ de la protection sociale. Droit social, 82–90. Laulom, S. (2017a). From Decentralisation of Collective Bargaining to De-collectivisation of Industrial Relations Systems? In S.  Laulom (Ed.), Collective Bargaining Developments in Times of Crisis (pp. 15–28). Alphen aan den Rijn: Kluwer Law International, Bulletin of Comparative Labour Relations. Laulom, S. (Ed.). (2017b). Collective Bargaining Developments in Times of Crisis (368 p). Alphen aan den Rijn: Kluwer Law International, Bulletin of Comparative Labour Relations.

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

325

Legal, R., Raynaud, D., & Vidal, G. (2009). La prise en charge des dépenses maladie des assurés sociaux en fonction du risque constaté : quelle contribution de l’assurance maladie et des organismes complémentaires ? In Comptes nationaux de la Santé 2009 (pp. 57–69). Paris: Direction de la Recherche, des Études, de l’Évaluation et des Statistiques (DREES). Leimgruber, M. (2012). The Historical Roots of a Diffusion Process: The Three-­ pillar Doctrine and European Pension Debates (1972–1994). Global Social Policy, 12(1), 24–44. Mares, I. (2003). The Politics of Social Risks. Business and Welfare State Development (344 p). Cambridge: Cambridge University Press. Mishra, R. (1990). The Welfare State in Capitalist Society: Policies of Retrenchment and Maintenance in Europe, North America, and Australia (152 p). New York: Harvester Wheatsheaf. Montaut, A. (2018). Tarifs et prestations: que révèle le nom des organismes de complémentaire santé?. Publication web sérielle, Études & Résultats, 1047. Paris: Direction de la Recherche, des Études, de l’Évaluation et des Statistiques (DREES). Müller, T., Vandaele, K., & Waddington, J. (Eds.). (2019). Collective Bargaining in Europe: Towards and Endgame (238 p). Brussels: European Trade Unions Institute (ETUI). Naczyk, M. (2013). Agents of Privatization? Business Groups and the Rise of Pension Funds in Continental Europe. Socio-Economic Review, 11(3), 441–469. Naczyk, M., & Seeleib-Kaiser, M. (2015). Solidarity Against All Odds: Trade Unions and the Privatization of Pensions in the Age of Dualization. Politics & Society, 43(3), 361–384. Natali, D., Pavolini, E., & Vanhercke, B. (Eds.). (2018). Occupational Welfare in Europe: Risks, Opportunities and Social Partner Involvement (2018 p). Brussels: European Trade Unions Institute (ETUI) & Observatoire Social Européen (OSE). Nyman, J. A. (1999a). The Value of Health Insurance: The Access Motive. Journal of Health Economics, 18(2), 141–152. Nyman, J.  A. (1999b). The Economics of Moral Hazard Revisited. Journal of Health Economics, 18(6), 811–824. Palier, B. (2010). The Long Conservative Corporatist Road to Welfare Reforms. In B. Palier (Ed.), A Long Goodbye to Bismarck? The Politics of Welfare Reforms in Continental Europe (pp.  333–387). Amsterdam: Amsterdam University Press (Changing Welfare States). Palier, B. (2017). La réforme des systèmes de santé (128 p). Paris: Presses Universitaires de France (PUF) (Que sais-je ?). Palier, B., & Thelen, K. (2010). Institutionalizing Dualism: Complementarities and Change in France and Germany. Politics & Society, 38(1), 119–148.

326 

T. HOUSSOY ET AL.

Pauly, M. V. (1968). The Economics of Moral Hazard: Comment. The American Economic Review, 58(3), 531–537. Pavolini, E., & Seeleib-Kaiser, M. (2018). Comparing occupational welfare in Europe: The case of occupational pensions. Social Policy & Administration, 52(2), 477–490. Perronnin, M., Pierre, A., & Rochereau, T. (2012). Enquête sur la protection sociale complémentaire d’entreprise en France (PSCE 2009) (Vol. 554). Paris: Institut de Recherche et de Documentation en Économie de la Santé (IRDES). Petit, B. (2019). Clauses de désignation: le Conseil constitutionnel désavoué par le Comité européen des droits sociaux. Revue des droits et libertés fondamentaux, chron. no 05. Retrieved from http://www.revuedlf.com/droit-social/ clauses-de-designation-le-conseil-constitutionnel-desavoue-par-le-comiteeuropeen-des-droits-sociaux-ceds-3-juillet-2018-cgt-fo-c-france-n118-2015/ Powell, M. (Ed.). (2007a). Understanding the Mixed Economy of Welfare (272 p). Basingstoke: Bristol University Press (Understanding Welfare: Social Issues, Policy and Practice). Powell, M. (2007b). The Mixed Economy of Welfare and the Social Division of Welfare. In M.  Powell (Ed.), Understanding the mixed economy of welfare (pp.  1–21). Basingstoke: Bristol University Press (Understanding Welfare: Social Issues, Policy and Practice). Ramaux, C. (2007). Quelle théorie pour l’État social  ? Apports et limites de la référence assurantielle. Revue française des affaires sociales, 1, 13–34. Ramaux, C. (2012). L’État social. Pour sortir du chaos néolibéral (470 p). Paris: Mille et une nuits (Essai). Rehfeldt, U., & Vincent, C. (2018). The Decentralisation of Collective Bargaining in France: An Escalating Process. In S. Leonardi & R. Pedersini (Eds.), Multi-­ employer Bargaining under Pressure: Decentralisation Trends in Five European Countries (pp. 151–184). Brussels: European Trade Unions Institute (ETUI). Rein, M. (1982). The Social Policy of the Firm. Policy Sciences, 14(2), 117–135. Rein, M. (1996). Is America Exceptional? The Role of Occupational Welfare in the United States and the European Community. In M.  Shalev (Ed.), The Privatization of Social Policy? Occupational Welfare and the Welfare State in America, Scandinavia and Japan (pp.  27–43). Basingstoke: Palgrave Macmillan. Rein, M., & Wadensjö, E. (1998). Enterprise and the Welfare State (416 p). Cheltenham: Edward Elgar Publishing. Sagan, A., & Thomson, S. (2016). Voluntary Health Insurance in Europe. Role and regulation, World Health Organization, European Observatory on Health Systems and Policies, n°43, 50 Seeleib-Kaiser, M., & Fleckenstein, T. (2009). The Political Economy of Occupational Family Policies: Comparing Workplaces in Britain and Germany. British Journal of Industrial Relations, 47(4), 741–764.

10  IN BETWEEN THE MARKET AND PUBLIC HEALTH INSURANCE… 

327

Shalev, M. (Ed.). (1996). The Privatization of Social Policy? Occupational Welfare and the Welfare State in America, Scandinavia and Japan (383 p). Basingstoke: Palgrave Macmillan. Sinfield, A. (1978). Analyses in the Social Division of Welfare. Journal of Social Policy, 7(2), 129–156. Spaeth, J.-M. (2008). Les syndicats et l’assurance maladie. Les Tribunes de la santé, 18, 29–37. Streeck, W., & Schmitter, P.  C. (1985). Community, Market, State-and Associations? The Prospective Contribution of Interest Governance to Social Order. European Sociological Review, 1(2), 119–138. Swenson, P. (1991). Bringing Capital Back in Social Democracy Reconsidered: Employer Power, Cross-Class Alliances, and Centralization of Industrial Relations in Denmark and Sweden. World Politics, 43(4), 513–544. Tabuteau, D. (2010). La métamorphose silencieuse des assurances maladie. Droit Social, 85–92. Thelen, K. (1993). West European Labor in Transition: Sweden and Germany Compared. World Politics, 46(1), 23–49. Titmuss, R. M. (1958). The Social Division of Welfare: Some Reflections on the Search for Equity. In R.  M. Titmuss (Ed.), Essays on the “Welfare State” (pp. 34–54). Londres: Allen & Unwin. Trampusch, C. (2007). Industrial Relations as a Source of Social Policy: A Typology of the Institutional Conditions for Industrial Agreements on Social Benefits. Social Policy & Administration, 41(3), 251–270. Trampusch, C. (2009). Collective Agreements on Pensions as a Source of Solidarity. Journal of Comparative Social Welfare, 25(2), 99–107. Trésor, D.G. (2015). Quel avenir pour le dispositif de prise en charge des affections de longue durée (ALD)  ?. Lettre Trésor-Eco, 145. Paris: Ministère de l’Economie et des Finances. Vincent, C. (2019a). La reconfiguration du marché du risque santé par la négociation: de l’interprofessionnel à l’entreprise et du contrat individuel au collectif. Revue de droit sanitaire et social, 810–818. Vincent, C. (2019b). Collective Bargaining in France: From Hesitant Decentralized Bargaining to a Rush Towards Prioritizing Enterprise Level. In T.  Müller, K. Vandaele, & J. Waddington (Eds.), Collective Bargaining in Europe: Towards and Endgame (pp.  217–238). Brussels: European Trade Unions Institute (ETUI). World Bank. (1994). Averting the Old Age Crisis: Policies to Protect the Old and Promote Growth. Policy Research Report. Oxford: Oxford University Press & World Bank. Yerkes, M., & Tijdens, K. (2010). Social Risk Protection in Collective Agreements: Evidence from the Netherlands. European Journal of Industrial Relations, 16(4), 369–383.

328 

T. HOUSSOY ET AL.

Zemmour, M. (2013). Les dépenses socio-fiscales ayant trait à la protection sociale: état des lieux, Publication web sérielle. LIEPP Policy Paper, 2. Paris: Laboratoire Interdisciplinaire d’Evaluation des Politiques Publiques (LIEPP). Zemmour, M. (2015). Exonérations, exemptions et dépenses fiscales : quels coûts pour la protection sociale ? La Revue de l’IRES, 87, 3–34.