Private Health Insurance in France: Between Europeanization and Collectivization

Abstract

Policy Points.

• Private health insurance (PHI) in France has been facing critical changes over the past 30 years. A complementary and voluntary scheme, it has been historically dominated by nonprofit entities. However, the share of for‐profit insurance companies in the sector has significantly increased. Nonprofit firms also changed their strategies and mimicked some for‐profit behaviors.

• The present paper argues that this process is a result of the conflict‐provoking coevolution of the insurance and health care sectors. Trying to improve the regulation of the insurance industry as whole, two European directives first jeopardized the business model of nonprofit entities. Then, two national reforms designed to improve health coverage significantly increased competition among insurers, notably in the area of corporate‐level contracts.

• Decoupling the insurance and health care sectors has become a major source of policy feedback and unexpected outcomes of reforms affecting the very organization of PHI.

Context

In France, private health insurance (PHI) has an exceptionally high level of coverage and accounts for 13.7% of health expenditures. A complementary and voluntary scheme, it has been historically dominated by nonprofit, mutual benefit societies. Over the past 20 years, however, the market share of for‐profit insurance companies has increased by 47%. Financialization of the field developed, and competition based on new risk management strategies also increased. The broad aim of this paper is to characterize and to elucidate the causes of this trend. More specifically, we are interested in how and to what extent a series of supranational and national policies contributed to this situation.

Method

Our data come from three sources. We first reviewed documents published by health insurers, government reports, and newspaper articles. We then conducted two semistructured interview campaigns between September 2017 and May 2018. The first mostly covered private and public actors and their involvement in European Union (EU) policymaking (n = 21). The second series of interviews was conducted with another group of actors directly involved at the French level (n = 16).

Findings

Our findings support preliminary observations. PHI in France, we argue, is indeed facing a development of competition and marketlike instruments. Four major policies (two EU directives and two national reforms) played a significant role in this outcome. Surprisingly, however, it has never been the purpose of legislators and policymakers: while EU directives created a regulatory framework for insurance activities within the Single Market, policies adopted at the national level initially aimed at improving health coverage. We show that it is the interactions and the noncoordination among all of these policies that explain their unexpected outcome.

Conclusions

The trend described in this paper is twofold. The first is Europeanization, as PHI in France is increasingly affected by EU legislation. Since this framework tends to favor larger firms and for‐profit companies, a reduction in statutory coverage can no longer be considered a quasi‐neutral transfer from (publicly owned) social security to nonprofit providers. At the same time, PHI is shifting toward collectivization: as competition increases, complementary health coverage is becoming gradually standardized and based at the corporate level. Together, these changes are likely to reduce freedom of choice and individual welfare, an assumption supported by studies published on the most recent period.

The french health care system features a mixture of public and private financing, including in the delivery of health services themselves. Coverage is provided to all legal residents through a National Health Insurance (NHI) scheme, mostly funded by social security contributions paid by employers and employees through an automatic deduction on salaries—and this even if fiscal resources have played a growing role in health care financing over the past 25 years. Essentially made up of employer and employee representatives, NHI funds (caisses) are quasi‐public entities largely overseen by the health minister. Both parties negotiate on a yearly basis the overall funding for health care in France. Apart from this task, the bulk of NHI activities is to process reimbursement checks for health care providers and patients through a network of regional and local funds.1 The range of health services covered is quite wide, whilst most beneficiaries are reimbursed at a rate of 75% to 80%, with some exceptions for long‐duration diseases that are fully funded by the public purse. In this context—and since the founding of the NHI system—a share of health coverage has been left to private health insurance (PHI), subscribed by individuals or corporations on a voluntary basis. PHI now has an exceptionally high level of coverage and plays a significant role in the reimbursement of copayments by the public purse for treatments and services in the statutory health care system.2

With few exceptions, PHI in France has rarely been studied, per se, within international comparisons. Indeed, the same statement applies to small‐n studies of all French health care. There are obvious reasons for this. In spite of gradual increases during the 1990s as a function of the reduction in statutory coverage for certain health services,3 the share of PHI has stabilized around only 13% of total health expenditure for more than a decade; for instance, it was 12.8% before the 2007‐2008 financial crisis.4 Thus, there seem to be no obvious changes here. Moreover, case studies of the French system insist, particularly by contrasting this situation with the United States, on the large market share captured by nonprofit firms when it comes to PHI.5 When the ratio of social security contribution to health expenses decreases and that of private insurers rises, this is certainly a transfer from quasi‐public to private providers; however, this is not exactly the same kind of “privatization”6 as it would have been if most of the latter group were made up of for‐profit companies. Sometimes portrayed as a “champion” of universal health coverage,7 the French system is thus usually seen as one of pacific and efficient coexistence between statutory and complementary schemes, which moreover are very similar in nature.

Without frontally contesting this broad‐brush picture, this paper begins with a different set of assumptions. What most comparisons neglect, we argue, are the significant changes that have occurred within PHI. Over the past 20 years, the market share of for‐profit companies has indeed increased by 47%. At the same time, most nonprofit insurers have changed their positioning and have adopted some of the same strategies as these companies, namely by engaging in mergers or takeovers and by adopting a risk management approach—points to which we will return. Last but not least, the most recent figures (2019) suggest a five‐point decrease in coverage by today's PHI contracts (from 95% to 90%).

The prime motivation of this paper is, therefore, to describe these changes and to explain why they have occurred. Since PHI is heavily regulated and highly dependent on reforms affecting the NHI sphere, our claim is that this process is the result of policies likely to change the way it is organized. This said, these reforms also appear to have originated from the two different sectors upon which PHI is anchored. As in other countries, private health insurers in France are compound entities. They can be said to be both part of the insurance industry as a whole and part of the health care system per se. A public policy or regulation adopted in one of these two fields may have strong implications for the other. When rules change for the insurance industry, one might observe direct and indirect effects for the health care system, and vice versa.

On the basis of this claim, this paper offers a descriptive theory that enables analysis to connect the change within PHI in France to a range of reforms originating from the health care and insurance industry sectors. We characterize this overall process as a case of coevolution, a term used to designate the reciprocal influence between policies adopted in the health and insurance sectors. The term coevolution has previously been applied to the different subsectors of health care, which depict it as a “diversified policy field.”8 Formally defined as the “mutual influence and adaptation of the institutional connection between related policy areas” over time, coevolution has been the subject of growing attention from health policy and administration scholars over recent years.9 When considering institutional change, it invites research to analyze the interrelated effects of past policy choices made in separate, though interdependent sectors. When an entity such as PHI is altered by coevolution, there can be widely differing cross‐sectoral outcomes, ranging from noncoupling (no convergence) or decoupling (divergence) to loose or tight coupling between policy sectors.10 Actors’ (eg, firms or interest groups) policies and institutions (instruments, rules, and norms) can come to be more or less distinct as the two sectors coevolve. When they become increasingly interdependent, one sector can play a leadership role in terms of ideas, interests, or policy orientation. Coevolution may also result in integrated programs or common objectives, when policymaking explicitly tries to integrate the two sectors. Conversely, coevolution can mitigate policy coordination and result in policy “disintegration,” a major issue in the health sector where intersectoral policy action may be needed to attain certain policy goals.11 As we shall demonstrate, the current situation of PHI is best described by the latter type of outcome.

In this context, the notion of coevolution appears to be an appropriate vehicle with which to grasp changes that have affected PHI in France over the recent period. While a first series of reforms reshaped the insurance industry as a whole, others both originated in and were directed toward the French health care system. More precisely, there have been a series of directives and regulations adopted at the level of the European Union (EU) that have deeply restructured how insurance firms now operate. At the same time, a range of policies adopted at the national level have sought to reinforce the relationships between the NHI and PHI, initially to increase health coverage. EU‐led reforms of the insurance sector did not consider health dimensions or potential effects on PHI; at the same time, national policies affecting PHI were made without any consideration for the effects of past EU directives and regulations on insurers. Thus, coevolution resulted here in divergence of the two sectors. As a consequence, both series of policies have had unexpected outcomes due to their noncoordinated interactions. More precisely, we show that PHI is currently pressured by two mutually reinforcing trends: Europeanization on one hand, with an increasing presence of EU legislation in the daily activity of PHI operators; and collectivization on the other, resulting from reforms adopted at the national level pushing toward a standardization and corporate‐level demand for health insurance products and services. A twofold process, this change has had a range of implications for the provision of health care and individual welfare as a whole.

Our findings are not restricted to the case under study. When there is no institutional coordination between insurance and health sectoral regulations, they tend to diverge. Such a form of decoupling generates policy “feedback”12 that policymakers are unable to anticipate. Some wider lessons might thus be drawn from this case, especially for countries in which PHI has some sort of linkage to a public system, but when regulations and policies fail to articulate the two sides of its dual nature.

The rest of the paper is organized as follows: In the first section, we present our data and briefly outline the main features of PHI and introduce our hypothesis. We also identify which policies may have induced its evolution over the recent period. The next two sections explore the relationships between these policies and the changes of mutual benefit societies, together with their effects on convergence between different players in the field. The fourth section concludes.

French Private Health Insurance: A Review of Recent Reforms

Sources and Approach

The data presented in this article come from three sources. We first reviewed documents and reports published by private health insurers (mutual benefit societies, for‐profit companies, and provident institutions; see below) and by their main regulators, namely, the Autorité de Contrôle Prudentiel et de Résolution (ACPR, the French pensions and insurances supervisor). Then, we extracted data from reports published by the Fonds CMU (CMU Fund, in charge of providing health coverage to people excluded from the compulsory regime in relation to private insurers) and the French Ministry of Health, mainly the Directorate for Research, Studies, Assessment and Statistics (DREES). These documents are listed in the references.2, 4, 25, 38, 43, 47, 48, 54, 61, 63, 65, 66 We then conducted two semistructured interview campaigns between September 2017 and May 2018. The first campaign was conducted by the authors. It mostly involved private and public actors and their involvement in European Union policymaking (n = 21). A second series of interviews was supervised by the authors and conducted by a group of students from Ecole des Hautes Etudes en Santé Publique with another group of actors directly involved in this process at the French level (n = 16). Interview sources are listed in Table 1. Additional sources include reports and information drawn from a body of 1,632 articles from major French national newspapers as well as sectoral press of the insurance industry (Actualités sociales hebdomadaires, L'Argus de l'assurance, La Tribune de l'assurance, Protection sociale informations) collected between October 2018 and February 2019.

Table 1.

Semistructured Interviews

Reforms	Civil Servants, Administrative Services, Ministerial/European Commission (EC) Agents	Private Health Insurers and Representatives	Managed Care Organizations Members, Professionals, and Representatives
Insurance directives and Solvency II (n = 21)	General Secretary (CEO), UNOCAM 10/2017 Former Ministry of Health cabinet member, former Social Security Director, General Inspector of Social Affairs 11/2017 Former Public Insurance Director 11/2017 Former member, Insurance and Pensions division, Ministry of Economic and Financial Affairs 12/2017 Former Seconded National Expert, EC/DG Growth 03/2018 Former Seconded National Expert, EC/DG Financial Institutions 04/2018	Social Policy & Prospective Advisor, Harmonie (MBS) 10/2017 Former President, MGEN (MBS) 11/2017 Former Vice President, Humanis (IP) 11/2017 Former President, French Mutuality Federation 11/2017 Former Head of European Affairs, French Mutuality Federation 12/2017 President, Union des mutuelles d’Île‐de‐France (MBS) 01/2018	—
	Policy Officer 1, EC/DG FISMA/Insurance and pensions division 05/2018 Policy Officer 2, EC/DG FISMA/Insurance and pensions division 05/2018 Policy Officer 3, EC/DG FISMA/Insurance and pensions division 05/2018	Former President Advisor, MGEN (MBS) 02/2018 Head of Institutional Relations, MGEN (MBS) 02/2018 Head of European and International Affairs, Humanis (IP) 02/2018 Former CEO, French Mutuality Federation 03/2018 Head of European Affairs, Solidaris (MBS) 04/2018 Performance and Insurance Director, Harmonie (MBS) 05/2018	—
Managed Care and Employment Security Act (n = 16)	Medical Doctor, General Inspection of Social Affairs 05/2018 General Secretary (CEO), UNOCAM 05/2018	CEO, French Mutuality Federation 05/2018 Head of Marketing and Innovation, MBA Mutuelle (MBS) 05/2018 Business Development Director, Humanis (IP) 05/2018 Commercial Development Director, GMF (MBS) 05/2018 Member of the Executive Committee, Allianz (IC) 05/2018	Health Economist, Health Institute (think tank) 05/2018 Economist, Asteres (consulting firm) 05/2018 Representative of the Consumer Association UFC 05/2018 CEO, Santéclair (managed care organization) 05/2018 Commercial Development Director, Kalivia (managed care organization) 05/2018
			CEO, Génération (brokerage firm)
			05/2018
			Optician affiliated with a managed care organization
			05/2018
			Audioprosthesist affiliated with a managed care organization
			05/2018
			Nurse affiliated with a managed care organization
			05/2018

Abbreviations: IC, insurance company; IP, provident institution; MBS, mutual benefit society; CEO, Chief Executive Officer (Directeur général); UNOCAM, Union Nationale des Organismes d'Assurance Maladie Complémentaires.

French Private Health Insurance: Between National Health Insurance and the Market

With NHI (Assurance maladie) covering most health expenditures, private health insurance in France has developed as a voluntary insurance scheme. It operates according to a complementary logic by covering services that are not fully reimbursed by the NHI. It is thus different from supplementary health insurance that may be found in other European countries such as Spain, Italy, or the United Kingdom. By contrast, supplementary health insurance generally guarantees “a wider choice of providers, faster access to treatment or […] superior accommodation in hospital.” Supplementary health insurance also tends to cover a rather small share of health expenses.3

Until recently, PHI as an industry in France was mostly made up of democratic, nonprofit organizations (mutuelles de santé, or mutual benefit societies). A mutual benefit society is a nonprofit firm led by an executive board elected by its members; each person affiliated with a mutual may vote or be elected to head the organization. Created before the NHI system, mutual benefit societies were maintained after its foundation by the 1947 Loi Morice, which guaranteed them a role in copayments for treatments and services within the statutory health care system.13 Since then, their development has been encouraged by the state by a range of tax credits and incentives.

A second group of private insurers was made up of provident institutions (hereinafter “IPs,” for institutions de prévoyance), created in 1945. They mostly operate in the area of occupational benefits. Involving employer and employee representatives, IPs offer regimes that may encompass health care but mostly deal with work disability and invalidity. Like mutual benefit societies, IPs are nonprofit organizations. Until recently, they enjoyed, together with mutual benefit societies, a de facto monopoly.

Since the end of the 1980s, a third category of private insurers has been formed by commercial insurance companies (hereinafter “insurance companies”).14 By contrast with mutual benefit societies and IPs, they take the form of for‐profit joint stock firms. Their market share was initially negligible as compared to that of mutual benefit societies. Over the past 20 years, however, it has increased by 47%. This movement has been paralleled by a significant decrease of the market share captured by mutual benefit societies. As the three actors operating on the segment of PHI in France, mutual benefit societies, IPs, and insurance companies diverge in terms of governance, commercial targets, and market share (Table 2).

Table 2.

Similarities and Differences Between Private Health Insurers in France

Type	Governance	Economic Model	Commercial Targetsa
Mutual benefit societies	Democratic	Nonprofit	Older people, employees, civil servants, midlevel executives
Provident institutions	Paritarian	Nonprofit	Unskilled workers, senior executives
Insurance companies	Shareholding	For‐profit	Farmers and self‐employed professionals

^aMossialos, Thomson.³

In spite of their differences, a unified term has increasingly been utilized in French administrative reports or legislative texts to designate all entities that provide “complementary health insurance” (assurance maladie complémentaire). The use of a single term is explained by the growing presence of IPs and insurance companies in the market. But it is also a by‐product of recent sectoral changes. Over the past 20 years, successive governments have tried to better integrate and associate private health insurers within the governance of the NHI. In 1999, these three actors were involved in the implementation of the CMU (couverture maladie universelle), a scheme introduced by the Parti Socialiste (left‐wing) government of the day. Until then, access to NHI was granted only to its contributors—namely, employers and employees and their families. This left aside many social groups, such as precarious workers or migrants. In this context, CMU (literally “universal health coverage”) is a device allowing people previously excluded from NHI to receive its benefits; at the same time, CMU‐C (CMU‐complémentaire, or complementary) was also created in collaboration with private insurers, to manage the complementary share of universal health coverage.15 CMU‐C involved the different private health insurers at the same level and in a collaborative fashion. In the same vein, a 2004 law created a single representative group for all categories of private insurers: Union des organismes complémentaires d'assurance maladie, or the National Union of Complementary Health Insurance Organizations (UNOCAM).16 This law also granted the UNOCAM board the right to participate in the determination of reimbursed benefits and “the basket” of care in the statutory health system—and this because the share of health costs that is not covered by public insurance is mechanically paid by private insurers.

Scholars of French health care agree that this process has resulted in both a more explicit recognition of the role of PHI in the overall system and a homogenization of for‐profit (insurance companies) and nonprofit actors (mainly mutual associations), at least from the perspective of their relationship to the statutory regime administration. However, such specialists tend to diverge when it comes to analyzing the causes of these trends. For some, they have been generated by a tacit, yet powerful logic of privatization of French health care, defined here as a transfer of resources and ownership from public to private insurance. Indeed, and since the 1980s, one can note a stagnation of the share of health spending paid for by NHI. This shift coincides with the first mentions in reports and legal texts of the field of “complementary health insurance,” a field that “should be organized and regulated.”17 The creation of CMU‐C and UNOCAM is seen as institutionalizing this “scrambling” trend by reinforcing the integration between private and public insurance—and offering the basis for further disengagement from the public sphere. Without necessarily rejecting this interpretation, other specialists insist that PHI homogenization and recognition have been paralleled by a process of “statization” or “technocratization” of NHI governance.18 From this perspective, successive reforms have reinforced the role of the state at the expense of unions and employer representatives, shaping the development of two interdependent spheres—statutory insurance led by NHI representatives and complementary insurance under the umbrella of UNOCAM—both overseen by a new “regulatory health care state.”

Though useful in order to understand the overall governance of French health care, both conceptions miss a part of the story: the sectoral changes that have affected the organization of PHI as an industry and the way they have interacted with the most recent reforms of the French health care system. Stated differently, PHI, as in other countries, is always shaped by reforms that originate in two sectors that might be more or less interdependent. The first, as we have just seen, is health care governance per se. Yet, PHI is also an industry regulated by norms, rules, and policy instruments that can be more or less interdependent or act jointly with health care governance.19 The coevolution of these two sectors is thus likely to induce heavy transformations and may affect the outcomes of reforms implemented in each of them.

Such a broader perspective enables analysis to identify the adoption of several policies in insurance and health care sectors that may have induced deep institutional changes for PHI. Not all of these policies were intended to establish a new relationship between the two sectors; yet, they have both had an impact on the organization of PHI and, ultimately, affected the provision of several health benefits for which the share of the NHI has been historically weak. Over this period, mutual benefit societies, insurance companies, and IPs have indeed been affected by a series of major reforms. The first were adopted at the level of the EU; they mostly affected PHI as an industry. Then, subsequent health care reforms were adopted at the French level, initially to increase health coverage through and by PHI. However, the effect of past EU policy choices was not taken into consideration by policymakers at the French level. Coevolution thus resulted in institutional divergence between the industrial and health care dimensions of PHI—or more precisely, between the policy orientation, rules, and norms of each sector. As a result, this lack of coordination has generated a range of unexpected outcomes.

In 1992, private insurers first became governed by EU “Insurance” directives, irrespectively of their status.20 These texts initially aimed at creating a European space for insurance goods and services; they also were intended to structure financial operations on the market. Mutual benefit societies and IPs were not their main target. However, as private entities, both were constrained to comply with the new legislation's requirements. One of the directives’ most obvious consequences has been to align these operators on the legislative and regulatory frameworks of for‐profit companies, especially in terms of governance—all of which could have potentially jeopardized their democratic (mutual associations) and paritarian (IPs) structures. Moreover, this framework was markedly reinforced in 2009 via the Solvency II directive.21

After the financial crisis, the goal of European legislators was to deepen a more prudential regulation of insurance market. Formally, the Solvency II directive forced insurers to meet a standard of technical provisions and capital requirements. For insurance companies, it notably reduced the possibility of a firm engaging in risky behaviors or investment strategies. The consequence was exactly the opposite for mutual associations and IPs: as nonprofit entities, they traditionally had reinvested their surpluses into benefits or services to their members. Under Solvency II, they have been incited (in order to respect the capital and technical provision requirements) to set aside financial provisions.

At the national level, French regulatory agencies and legislators as of 2009 began to encourage the development of managed care organizations (MCOs), initially to reduce the cost of medical technologies for which the share of PHI has historically been high (notably for dental and optical care, see further). In 2014, a law (Loi Le Roux)22 granted mutual benefit societies the right to practice differentiated reimbursement rates, a common strategy for insurance companies but, until then, strictly forbidden for mutual benefit societies. This policy has fostered the development of risk management within mutual associations and, potentially, their control over the behaviors of their members. Finally, and without any obvious link to the previous reforms, a national industrial agreement was signed by employer and employee representatives23 and became part of the law in 2013 (Accord national interprofessionnel, within the Employment Security Act).24 To attain a better coverage of workers, it introduced an obligation for employers to provide their workers with a PHI scheme. By creating new outlets, this last reform potentially paved the way for increased competition among insurance companies, mutual benefit societies, and IPs—the latter appearing to benefit from this reform, due to their positioning in regards to corporation‐level‐based contracts.

Each of these reforms has seemed to involve different actors and addressed specific sectoral issues. Their basic common point is to have affected norms, rules, and conventions institutionalizing PHI in France from separate (sectoral) perspectives. The Insurance and Solvency II directives are likely to have contributed to the formation of an institutional framework for the insurance industry at the scale of the EU and perhaps trivialized the business model of nonprofit entities. They also created an unprecedented regulatory framework, based on strict prudential and governance rules. The reforms adopted in the French context (development of MCOs for mutual benefit societies and the Employment Security Act) were initially justified in terms of “better access to care.” But they seem to have also created, at least in legal terms, the condition for increasing competition and business combinations.

If there has been no obvious coordination among these sectoral evolutions, official figures tend to suggest that their consequences for PHI have been many. In its most recent (2017) report, the Directorate for Research, Studies, Assessment and Statistics (DREES; Ministries of Health, Labor, and Public Action) collated data from different regulatory sources.25 It revealed a significant rise of the market share of for‐profit companies, rising from 19% in 2001 to 30% in 2016. During the same period, that of mutual benefit societies decreased to hit a low of 52% in 2016. Some indications also suggest a financialization of nonprofit actors. Among other indicators, the concentration rate for these entities increased dramatically: there were 1,528 mutual benefit societies in 2001 and only 365 by 2016; 57 IPs in 2001 and only 25 in 2016. These figures are mainly explained by the rise of strategic alliances, mergers, and acquisitions within this category of health insurance, all of which led to the formation of large and private “social protection” groups. But can we identify a causal effect of one or several of the reforms mentioned here in regard to this change? Did they interact or even generate cumulative effects? If so, what does this tell us about the provision of health care in France?

More generally still, answering these questions amounts to identifying which mechanistic explanation accounts for this outcome by supplementing basic figures by within‐case analysis and by considering the temporal sequence of events. We need to identify the proper effect of the four reforms under study. Two were clearly directed toward PHI as an industry (EU Insurance directives and Solvency II), while two were sectoral evolutions of French health care (MCOs and the Employment Security Act). Control mechanisms also need developing. Consequently, we have integrated within our analysis some shifts that cannot be directly attributed to the policies in question—in particular, autonomous market dynamics or actors’ strategies separated from the logics of reforms. Accordingly, in the next two sections we will unpack this process to look for causal chains among the observed events. For each sequence identified, we will also examine the temporalities and feedback effects of prior reforms adopted at different points in time.

The European Dimension of PHI Changes

The Impact of the “Insurance” Directives

European Union treaties are categorical as to the limited competencies of the EU when it comes to welfare, especially in the area of health policy.26 For this reason, the influence of the EU, largely documented, has regularly taken circuitous routes. In this respect, health policies or services have typically been affected by legislation relative to the architecture of the Single Market, such as fiscal governance.27 The situation has nevertheless been different for the particular case of private welfare providers, such as pension funds or PHI. Considered enterprises in the European sense of the word, all insurance schemes that are not explicitly part of the statutory system are entities for which European rules and regulation may legitimately apply, especially in terms of competition policy. It is thus easily understandable that the first sequence of reform of PHI in France opened with the adoption of European directives. Its effects have been perceptible in the long run and led to a new regulatory framework at the end of the 2000s. These changes in turn deeply affected the governance and the organization of each PHI operator, especially mutual benefit societies.

When the Insurance directives were passed at the European scale in 1992,20 their main objective was to facilitate competition between insurance companies in the Single Market. Once an insurer had obtained an administrative agreement to carry out its activity within a member state, it could sell its products and services on the whole internal market. This opening up to competition prevents governments from applying product and price controls; only financial regulation is permitted. Discrimination among insurers through regulatory and fiscal means is also prohibited.28 If national health services or insurance schemes were explicitly excluded from these directives, they nevertheless included entire segments of the private sector operating within this field. In most EU countries, this change has not been a major issue. PHI constitutes less than 5% of total health expenditures in Europe as a whole. Two countries—Ireland and the Netherlands—have a much bigger private sector, since several groups are excluded from some aspects of the “means‐tested” statutory health insurance scheme.29 However, these countries negotiated with the European Commission a possibility for governments to heavily regulate this industry in the interest of the “general good” if “contracts covering health risks serve as a partial or complete alternative to health cover provided by the statutory social security system.”26 In spite of this provision, health care was not the main purpose of the Insurance directives. Pensions and financial activities of insurers, as well as corporate governance, were much more salient issues during the negotiation. Yet, the consequences of these texts for the health insurance industry have been far more important for PHI in France. The absorption of mutual benefit societies and IPs within these texts, particularly, opened a first sequence of institutional change. Indeed, most of their activities have henceforth been partially indexed to the judicial and regulatory evolutions that have taken place at the scale of the EU, for which they were largely unprepared.

At the national level, the directives implied guaranteeing that competition was fair and, by extension, that the devices that could be more favorable to a particular category of operators were to be removed. Scheduled for 1994, their implementation constituted a major issue for French mutual benefit societies, and this more than for insurance companies and IPs. For the former, they extended a competitive environment in which they already operated. For their part, IPs, managed between labor organizations and employee representatives, aligned their position on that defended by the European Trade Union Confederation: these actors considered that the Insurance directives could constitute an opportunity to expand at the European level their conception of private insurance, based on social dialogue and in close relationships with labor organizations. Moreover, IPs were firmly positioned as regards collective contracts, at the scale of large corporations or professional branches; consequently, they did not see this change as a threat to their activities.

The positioning of mutual benefit societies is, however, more ambiguous. A latent divide rapidly emerged between two opposing positions, a development that can be explained by the dominant member profiles of each entity and by their relative proximity to French labor organizations. Mutual associations that specialized in the complementary health coverage of civil servants and were closest to the General Confederation of Labor (CGT), a union that develops a statist and class‐struggle conception of industrial relations,30 expressed their strong opposition to the market logic behind European directives. They feared a trivialization of their democratic, nonprofit model within one big European market. By contrast, mutual associations operating on an interprofessional basis (ie, not dedicated to a specific category of worker) and closer to the Democratic Confederation of Labor (CFDT), a union that promotes Christian‐democracy, social‐democracy, and Fabianism,30 saw the directives as an opportunity for building a European mutualist movement, as well as creating synergies with other nonprofit organizations across the continent.31

Ultimately, a compromise was found within the National Federation of French Mutuality (FNMF), an organization supported by some civil servant mutual associations that were in favor of the line defended by interprofessional mutual associations. FNMF representatives publicly took position in favor of the integration of French mutual benefit societies within the framework of the Insurance directives, but in exchange for the recognition of a specific status that would dissociate mutual benefit societies from other insurance companies—namely, on the basis of their long‐standing vocation in health care and their democratic, nonprofit business model.32 On the basis of the values shared among mutual benefit societies, the goal of the FNMF was thus to reject the principle of equal treatments between insurer operators as established by the draft directives, a corollary of consumers’ “freedom of choice” as promoted by the European Commission. However, agents from the commission quickly rejected this demand; according to their view, the Insurance directives, as with other EU business legislation, never recognize the ‘organic criterion’ and only consider the nature of a firm's activities.33 Stated differently, mutual benefit societies and IPs could keep their democratic organization and their nonprofit business model, since it was not the purpose of EU law to modify these traits. But they could not claim a specific status on this basis, since they did not undertake the same kind of operations as for‐profit insurance companies. Consequently, the French mutual benefit societies were forced to accept the European directives without gaining any compensation in return.

At the national level, the legal Europeanization of mutual associations has had direct effects upon institutionalized relationships within the sector of PHI, especially for those in charge of the complementary coverage of civil servants. Although competition for whole corporation‐level contracts slightly increased between insurance companies and IPs, these mutual associations seemed relatively protected on this market segment. In addition to financial support, the state provided them with personnel and benefits in kind. But this situation changed when a mutual excluded from these aids successfully brought legal action in the Conseil d'Etat (France's highest administrative court) and prompted the European Commission to denounce, on the very basis of the Insurance directives, the principle of state aids to civil servant mutual associations. The Conseil d'Etat and the commission agreed with the view of the mutual that brought this case and pressed the French government to abrogate these “unfair subsidies,” considered to confer illegal selective advantages within a competitive market.34

These decisions quickly induced changes. Each call for tenders now had to be organized before any selection of a complementary health insurance by the state for its employees; more important, insurance companies, IPs, and mutual benefit societies now had the possibility to compete on the same level. The promotion of this new instrument was thus a direct consequence of European directives; if they did not modify the governance or the core principles of mutual benefit societies, they certainly increased competition among the different players in the field, since they contributed to equalizing their opportunities to develop in different segments, including the protected ones, such as the complementary health coverage of civil servants.

Paradoxically, this change has been paralleled with a symmetrical “statization” of the values of mutual benefit societies. Indeed, following their legal injunction, agents from the European Commission recommended French authorities offset this opening up to competition by maintaining some public aids or fiscal allowances within the sector.35 To them, the problem was not that several contracts were favored on the basis of a range of principles but that these contracts being offered by only one category of market actors led mechanically to selective advantages. In its opinion, the European Commission suggested generalizing these aids to any kind of private health insurer on the basis of its acceptance of a range of principles initially promoted by mutual benefit societies. In this respect, the standard for PHI contracts was reformed in 2006. When the subscription of a complementary health contract was not conditioned or based on the health status of a person, and if the insurer in question did not charge additional fees if this person's health status deteriorated, the contract became eligible to a range of tax allowances. If competition between insurers has increased, these typical mutual association principles have thus been generalized through state regulation.

Solvency II: An Increasing Homogenization of Insurers?

An underlying implication of envisaging European integration as a consequence of the Insurance directives (fast and without many difficulties for IPs and insurance companies, slower and more conflictual for mutual associations) leads de facto to a mere indexation of their activities to change in EU legislation. From this perspective, the supranational side of PHI change was naturally accentuated during the second half of the 2000s, when new directives were adopted. The latter are seen as simply adding an issue of security to the principle of equal treatment among operators.

In 2009, the Solvency II directive was adopted at the scale of the EU.20 Legitimized on the basis of security principles—namely, that of financial operations of insurance activities—it provided new quantitative measures of equity capitals for enterprises subject to Insurance directives. The Solvency II directive is itself a translation of Basel agreements, a prudential regulatory device ostensibly providing for a better assessment of banking risks, on the basis of new requirements in terms of equity capitals and financial strength.36 Solvency II was explicitly presented as a response by EU authorities to the financial crisis. The text's most explicit goal was to set up a “risk‐based” approach to regulation, in accordance with the very basic idea under which the riskier a firm's business, the more precautions it should take in conducting its operations. Formally, these texts set a threshold of minimum capital requirement (MCR), below which an entity is considered unsustainable and should face a withdrawal of its insurance authorization. In the same vein, a solvency capital requirement (SCR) was also defined. It designated the amount of capital required to meet quantifiable risks on an existing portfolio. More simply, it is the level of capital that an entity should possess to “absorb an exceptional shock,” such as the necessity to fulfill all of its engagements.37

The consequences of Solvency II vary according to the situation of the different operators. For insurance companies, this rule explicitly depreciated financial strategies based on the acquisition of higher‐yielding bonds, but that may result in heavy losses. Nevertheless, it appears that this change did not significantly alter the structure of insurance companies, since most of them were able to comply with these requirements. Indeed, these actors had already “marketized” their nonhealth activities and were thus familiar with this kind of prudential regulation.

The impact of Solvency II, however, is clearly deeper for mutual benefit societies and IPs. These two categories of actors previously disregarded the private accumulation of profits; their gains served directly to improve the benefits provided to their members or to increase their insurance coverage. To comply with Solvency II requirements, however, they now needed to make provisions and to find ways to increase the value of their funds—this in order to possess enough capital to satisfy both the MCR and SCR ratios. In addition, most of them are small and medium‐sized firms, whose funds might be considered far below these ratios. This situation was to have a direct impact on the configuration of the sector. Between the preliminary sessions prior to the adoption of Solvency II and its official date of enforcement (January 2016), mutual benefit societies and IPs engaged in a vast movement of concentration in order to reach a critical size to comply with the criteria laid down in the directive (Figure 1).

Figure 1.

The Decreasing Number of Mutual Benefit Societies and Provident Institutions^a

^aData adapted from ACPR, DREES, CMU Fund2, 4, 25, 54 [Color figure can be viewed at http://wileyonlinelibrary.com]

If the overall number of insurance companies thus remained stable, a significant decrease in the number of IPs and mutual benefit societies occurred. Our interview data reveal in the case of IPs that this trend cannot be attributed solely to the enforcement of Solvency II. To realize economies of scale, employer and employee representatives decided to engage in several mergers, years before and irrespective of the European legislative context. Solvency II only contributed to accelerating a trend that was already happening. However, the same data suggest that Solvency II was a major driver of mutual benefit societies’ mergers and takeovers. Within a short time period, these actors faced first the impact of Insurance directives in 2001 (which already had more stringent solvency rules than the previous scheme38) and then, deeper prudential requirements through Solvency II. More dispersed and fragmented, they were much more exposed. This situation encouraged an increase in the number of strategic alliances and mergers and acquisitions between entities and the formation of big mutualist firms. The latter were instrumental in this shift, such as in the case of Vyv care, the result of the merger of Harmonie and MGEN—France's biggest civil servant mutual association—with Harmonie and MGEN being themselves the result of previous mergers.39

But this situation was also the product of the longer‐term effect of Insurance directives. A clear combination of these policies can be seen. Facing an increasingly competitive environment due to the growing presence of insurance companies and the loss of their selective advantages, mutual benefit societies have considered themselves forced to adapt and to respond to the decrease in their market share. But Solvency II also induced changes at the organizational level. Since the regulatory framework became risk‐based and reinforced, nonprofit entities, previously formed of elected members (mutual associations), and labor organization representatives (IPs) needed to hire new skill profiles. Their democratic and paritarian governance is now partially shaped by specialists of the quantitative measure of risks, notably actuaries and data scientists. By creating new regulatory requirements, the frame institutionalized by Solvency II thus contributed to weakening the capacities of elected members and worker representatives to shape dialogue with the regulator or with other national authorities in charge of the activity of these entities.

As part of the insurance industry, PHI was thus deeply transformed by the Insurance and Solvency II directives. If efforts have been made at the national scale to protect and extend principles and values shared by nonprofit actors, these sectoral evolutions contributed to decoupling PHI from national policies and regulations that were more heavily connected to health care system governance. In this way, these texts have indeed introduced a range of marketlike instruments and principles within a sector partially governed by other logics at the French level, notably those of mutual benefit societies. In some respects, this process can be considered a trend that has contributed to homogenizing different categories of actors while simultaneously aligning them to the rules and norms that govern other (ie, nonhealth) insurance activities. If not all of the previously mentioned changes can be attributed to European legislation, given that the trend in concentration that affected IPs was partially driven by other strategies, they have resulted in a decoupling of the insurance and health care dimensions of French PHI.

When a European Legislative Framework Collides With a National Political Agenda

The New Frontiers of Risk Management

The process of change of PHI opened up by European legislation has been paralleled and paradoxically reinforced by a series of reforms and policies at the French level. However, there is no clear relationship between these two streams of sectoral changes, since the principles defended at the national level, mostly focused on the health care system, contrast with the rationale behind supranational interventions on the insurance industry. While European legislation has pushed in favor of a more integrated framework for the different operators of PHI on the basis of principles of security and liberty of insurance activities, changes at the French level have been mostly concerned with better access to care, without much consideration for this first series of reforms.

During the second half of the 2000s, most actors of PHI have sought to develop, initially without much explicit strategizing, new instruments for risk management—a term that within the insurance industry designates claims management, a prevention of their aggravation and the reduction of repair costs. To be more efficient in this area, the different categories of health insurers have tried to develop MCOs of variable sizes and forms. Their basic common trait is to rely on agreements passed between private health insurers on the one hand and between health care professionals or facilities on the other. These providers are committed to respecting fixed prices or rates, which allow health insurers to control their expenditures—knowing that insured persons are incited to consult these professionals and facilities by the modulation of rates and levels of reimbursement. Since they generate the possibility for insurers to collect data on insured persons, MCOs facilitate claims management and have allowed them to structure their supply (ie, range of products).

Beyond these basic features, French MCOs are somewhat different from most types of MCOs that may be found in other countries, typically in the United States.40 In France, the NHI system guarantees patient freedom of choice in terms of medical consultation and care consumption. Competition among private health insurers is limited to a small share of health expenditures. Moreover, the NHI and the state in France regulate the price of health care goods and services. It thus makes competition among care providers less effective.41 In this context, MCOs have developed in France for hearing aids, dental care, and optical care—benefits for which the coverage of PHI has historically been high. Patients and professionals are financially incentivized to participate in these networks; products and practices are heavily controlled by PHI, and the latter also try to organize behaviors of both through these entities.42 But the similarities between French MCOs and health maintenance organizations (HMOs) or preferred provider organizations (PPO) end there, due to the features of the French health care system. Patients still have the possibility to see health care professionals who are not members of the MCOs of their own health insurer. MCOs do not replace the NHI system, which continues to reimburse a certain rate of health expenditures, while the share of PHI remains the same. More fundamentally, the power of PHI and its capacity to influence price increases with MCOs remains far less important than in US PPOs and HMOs.

During the same period, management platforms have also developed. Playing the role of brokers between insurers and health care professionals or facilities, these entities are now in charge of the former for copayment and negotiating commercial advantages with professionals; for the latter, they channel patients and handle information systems. These quasi‐industrial platforms have in turn contributed to the emergence of large MCOs after 2010. Albeit significant, these initial developments were made without any formal regulatory framework, health insurers generally determining their own criteria in terms of quality and health services evaluation.43 Insurance companies were the main actors building these entities, an approach they had already developed in other segments (ie, nonhealth) of their activities. However, risk management through MCOs spread after 2009, notably following an opinion by the French Competition Authority, a government‐based competition regulator. It expressed support for these entities, praising their “pro‐competition impact,” their transparency, and their positive effect on prices.44 This opinion gave key support to an intensified development of MCOs and, by extension, industrial platforms. Thereafter, the constitution of large networks or professionals and facilities came to be seen as an opportunity for other categories of health insurers, especially nonprofit actors. If it did not originate this strategy, the Competition Authority's opinion nevertheless legitimized a posteriori the development of managed care without appealing for greater regulation. Using the argument of free competition, insurers subsequently, through MCOs, increasingly based their relationships with health care professionals on financial and managerial criteria. The feedback effects of European directives were observable during this short period of time: in a context of intensified competition with insurance companies, IPs in particular developed their own MCOs to increase the profitability of their collective, business‐level contracts.

The development of MCOs by private health insurers quickly interacted with partially independent transformations that have occurred within the French health care system, and this in relation to governmental attempts to increase access to care without extending the coverage of the NHI system. The growing implication of mutual benefit societies in MCOs has been instrumental in this shift. Yet, their historical values have always been in strong opposition to the logic conveyed by the managed care model. From the end of the 19th century until the end of the 1990s, the mutualist movement developed its own network of health and social facilities.45 Since its origins, this project had been conceived as a political alternative to the development of individualistic medicine, based on the autonomy of physicians (in a mutualist health facility, they are employees); the contemporary model of French medicine developing for its part largely in reaction to the collective approach fostered by mutual benefit societies.46 These facilities were subsidized by mutual benefit societies on the basis of patient needs, using their surpluses to increase the benefits offered to their members. They have never been subject to budgetary constraints. The mutualist conception of managed care spread on the basis of these principles throughout the 20th century, aiming at enhancing access to care.

In 2001, the coming into force of the European Insurance directives modified this approach. Indeed, these texts required insurance firms to specialize; to prevent a random use of their funds, they pushed all enterprises subject to European legislation to focus on insurance activities. For mutual benefit societies, this implied separating themselves from their own facilities, which could be considered as charities under European law and thus incompatible with an insurance activity. To retain these entities that were constitutive of their very model, most French mutual benefit societies accordingly decided to separate their activities into two autonomous firms, one dealing with insurance operations, the other with health and social facilities—an umbrella structure being generally created to articulate both and to ensure financial solidarity between the two.47 Within this framework, the insurance side was to be fully submitted to the Insurance directives while the other was not. Most of these transformations occurred after 2001, the date of the enforcement of the European directives.

Within the FNMF, a debate took place over the following years on the future of mutualist health and social facilities. A diagnosis quickly emerged: in the context of an organizational convergence between the different actors of PHI, they presented themselves as a major component of mutualist identity and as a vehicle for its nonprofit, solidarity‐based conception of health and access to care. However, if this position is firmly defended by the federation, this formula is seen less and less as a viable solution by individual mutual benefit societies.

This situation is a direct consequence of the growing engagement of insurance companies and IPs in MCOs. The size of these networks is significantly larger than mutualist ones; moreover, these actors have engaged in hard bargaining with health care professionals to obtain the lowest possible prices, an approach that mutual associations have traditionally refused to adopt in their own facilities.

Mostly during the second half of the 2000s, several big mutualist groups began to develop an activity based on MCOs, this change mainly taking the form of an integration of former mutualist facilities and structures into larger entities. In most cases, their administration was delegated to a platform gathering together IPs, insurance companies, and mutual benefit societies.48 This shift is also a consequence of the trend toward concentration initiated by Solvency II. Big mutualist groups, which initially developed to meet the new regulatory requirements, gained as a result greater market power in their relations with health care professionals, making it appropriate for them to impose their own prices and rates upon these actors. The position defended by the FNMF on managed care organizations had shifted in accordance with these changes by the end of the 2000s; the issue is no longer to find solutions to protect the model of mutualist health and social facilities but to determine how and to what extent MCOs could be compatible with mutualist values. The diagnosis had thus been subtly reformulated. Since they participate in the reduction of health expenses and access to care, MCOs are presented as fully compatible with mutualist principles as initially promoted through mutualist health and social facilities. Using the opinion expressed by the Competition Authority to legitimize this change, these actors have argued that their efficiency is also of great interest to their members.

Given this new diagnosis, mutual benefit societies have increasingly involved themselves in MCOs. They have also begun to develop a modulation of levels of reimbursement, with patients benefiting from higher coverage if they decide to consult a professional who is under contract with an MCO. However, the spreading of this approach to insurance among mutual benefit societies was suddenly halted in 2010 by the Court of Cassation, an appellate court of the highest instance, which considered the adoption of this practice by mutual benefit societies to be illicit—and this because, in the French context, mutual associations still operate within a legal framework that distinguishes them from insurance companies and prohibits several practices.49 Publicized by national newspapers, this case reintroduced a difference between mutual benefit societies and other types of private health insurers.50 Fearing that this decision would be remobilized in future cases, the FNMF's president pressed the government to restore equality of treatment among private health insurers—which might be considered an implicit recognition of their common identity, at least from a legal point of view. The mutualist movement thus became an advocate of free competition. Indeed, this initiative led to the adoption of Loi Le Roux in 2014, legalizing the modulation of levels of reimbursement by mutual benefit societies.51

From a governmental perspective, its implicit support for MCOs through this law has been in no way disinterested. Most funds allocated by private health insurers encompass benefits for which the coverage of NHI has been historically weak—as previously mentioned, dental care, optical care, and hearing aids (Figure 2).

Figure 2.

Share of Private Health Insurance Benefits by Category^a

^aData adapted from ACPR, DREES, CMU Fund2, 4, 25, 54 [Color figure can be viewed at http://wileyonlinelibrary.com]

Rather than extending public coverage, an approach preferred by successive governments has been to put pressure on health care professionals to lower their prices in order to limit out‐of‐pocket expenditures. However, this strategy has proven ineffective. In this context, the development of large MCOs is seen by several agents within the Ministry of Health as a way to contain the rise of these costs through an implicit delegation of rate and price negotiations to private health insurers. In November 2014, a decree relative to PHI contracts reinforced this trend.52 Its goal was to curb the rise of extra fees charged by several health care professionals, such as medical doctors but, more importantly, dentists and opticians. To this end, it forced private health insurers not to reimburse the complementary share of health benefits exceeding a certain amount of money. Thus, they cannot guarantee to insured persons, even for the most generous contracts, that their expenses will be fully covered in any case, especially if they have consulted a health professional who charges extra fees. Private health insurers cannot thus reimburse more and are constrained to pay less. If they want to maintain high levels of coverage, they are incited to act directly on prices as charged by these professionals—and this mainly through MCOs.

The tacit recognition of MCOs in the public sphere (through opinions expressed by Competition Authority, Loi Le Roux, and the 2014 decree on the regulation of PHI contracts) has thus constituted an additional step toward the changing nature of PHI in France. In this context, the transformations of this industry at the EU level of the insurance sector interacts with a national political agenda within the health care sector. Although this process has increased a risk‐management‐based competition among insurers, it has also reinforced homogenization among private health insurers, especially through the authorization granted to mutual benefit societies to practice modulation of reimbursement rates. One should also note that these policies have legitimized and reinforced a trend that had already developed among private insurers. However, most prior engagements from health insurers toward MCOs can be explained by the frame institutionalized by European legislation. Again, the case of mutual benefit societies is particularly revealing from this perspective: the development of MCOs has been a causal result of the fact that they now compete on equal footing with IPs and insurance companies. This trend has been reinforced by the Solvency II directive, since it has led to the formation of big mutualist groups more able to engage in hard bargaining with health care professionals. Without directly supporting MCOs, the government has seen this approach to risk management as a way of delegating the containment of health costs to the private sector.

Generalization of Complementary Health Insurance: Better Access to Care or Increasing Competition?

During the 2010s, another series of policies adopted in France has accentuated the change of PHI and combined with the previously mentioned ones. However, they have not resulted in a better integration between the logics behind insurance and health care sector reforms, as these two systems have endured coevolution without much explicit coordination. As a result, policy feedback continues to generate unexpected effects. The high point of these changes has been the ratification of the Employment Security Act in 2013 by employer and employee representatives, converted into law a few months later.53 This historical sequence corresponds to the compulsory generalization of complementary health coverage to all employees working in the private sector as of January 2016. In this context, the major changes were expected for small enterprises, only 33% of which gave complementary health coverage to their employees in 2009.54 In particular, this reform has developed a major aspect of the process of change in PHI since it sees labor organizations and the state as forcing private firms to subscribe an insurance contract to cover their employees—resulting not only in increasing competition among insurers but also in a new distribution of individual and firm‐level‐based contracts. However, the initial objective of this reform is unclear: this critical shift appears to be a compromise resulting from the linkage between two independent sequences, the first being related to health policy per se, and the second to the internal structure of the relationships between employer and employee representatives at the national level.

If the Employment Security Act and the accompanying 2013 law can be formally considered an extension of the market, from the state's perspective they fit with an older concern. As seen for the case of MCOs, a common objective shared by successive French governments since the 1980s has been to limit out‐of‐pocket payments without substantively extending NHI coverage.55 Privatization has not, however, been the hidden rationale behind this agenda. Over the past 30 years, the global, gross increase in health expenditures has not been paralleled with a significant variation in the respective share supported by public and private insurances, still around 13% for the latter in 2016. Instead, a more explicit agenda has repeatedly been to socialize the field of PHI, or at least to orient its development via a series of policies likely to foster governmental objectives. In this context, several governments of different political colors have considered that encouraging the generalization of PHI was a viable option. Here, the goal of public policy has been to make sure that each person affiliated with the NHI holds PHI for the reimbursement of treatments and services that are not, or are poorly, covered by the public one (see Figure 2).

A first move in this direction occurred at the end of the 1990s. As with the CMU, the then socialist government tried to get populations covered who had previously been excluded from PHI. This reform focused on the usual target groups of universal health coverage—in particular, retirees and employees at the bottom end of the wage scale.56 CMU‐C (“complementary”) was implemented in 2000, a device completed in 2004 by Assistance for Complementary Health Coverage (ACS, for Aide à la complémentaire santé).16 ACS took the form of a financial aid to purchase private insurance, allocated to individuals whose incomes are slightly above the threshold for benefiting from CMU‐C.57 They may then freely choose a complementary health insurance on this basis. CMU‐C and ACS can thus be considered as a market institutionalized by the state. These policies reinforced a partnership between the state and private insurers. The state removed barriers to access for a product from which several social groups were de facto excluded; private health insurers got new customers in exchange for a financial contribution through an additional “solidarity” tax on insurance contributions. In 2015, public regulation of these devices was reinforced with the aim of better targeting beneficiaries and maximizing their performance from the viewpoint of public health. Public actors in charge of CMU‐C and ACS have since established a set of eligible contracts on the basis of a call for tenders. Depending on their interest in public health, but also on a cost‐effectiveness balance, a range of contracts were to be selected and proposed to the beneficiaries of ACS. If individual choices were to remain free, public regulation was considerably reinforced, since it reduced the supply to a limited number of “good” products.

The development of CMU‐C and ACS might be considered attempts to de‐privatize the field of private health insurance. For this segment, public spending increased and regulation was reinforced. However minor in appearance, this process is nevertheless quite distinct from “marketization via compensation,”58 in which the share of public spending is expanded but priority is given to market solutions supported by private actors. Moreover, it is not, as in other countries such as the Netherlands, an increase in the share of private spending heavily regulated by the government59 nor a transfer of several risks from the public to the private sector, since both devices were designed for specific social groups. Here, a share of health insurance that was already private has been socialized, in the sense that public expenses and state capacities to shape private contracts have been extended. Nevertheless, it is true that marketlike instruments have developed on the basis of state intervention. Indeed, CMU‐C and ACS now cover people previously excluded from the market; rather than grant them directly with health coverage, public intervention has organized their free choice and created a new space for competition between private health insurers. Nevertheless, market development has corresponded here more to the “default option” than to the initial governmental intention:60 It appears more as a way of expanding health coverage in a context of budgetary austerity.

Indeed, years after their implementation, the initial objectives of CMU‐C and ACS—better health coverage and, ultimately, better access to care—have proven difficult to reach. Contributing factors are the design of these devices, which remain voluntary, and the fact that these financial aids are not automatically granted to eligible beneficiaries. Instead, beneficiaries first need to apply through a complex process.61 Worse still, data from the Ministry of Health reveal that both policy instruments remain largely unknown, especially ACS; for the newly elected socialist government in 2012, this market‐way quickly came to be considered an imperfect response. In parallel with the better regulation of CMU‐C and ACS in 2015, agents from the Ministry of Health therefore began to identify alternative routes to increasing health coverage. A solution that would consist of generalizing complementary health coverage through a range of fiscal mechanisms was considered. A proposal was elaborated on the basis of this diagnosis: since retirees, precarious workers, and young people are the main populations excluded from complementary health coverage, the government sought to remove fiscal aids for several PHI contracts (such as those set at the corporation level) to reallocate these aids to uncovered individuals, and thus to provide them directly with complementary health coverage. The proposal was presented in detail by President François Hollande during the 40th FNMF Congress.

Most mutual benefit societies indicated their strong support for this initiative.62 However, and despite explicit support from the FNMF, this attempt failed for unexpected reasons resulting from independent shifts within industrial relations. At the same time, employer and employee representatives were negotiating a national interprofessional agreement on “competitiveness and employment security.” Initially, complementary health insurance was a peripheral issue in these negotiations: they mainly focused on amendments to the Labor Code and on employment standards. Although the CGT as well as other unions strongly opposed the propositions of employer representatives, the latter asked for more “flexibility,” especially in terms of working time. In this context, members of the CFDT accepted part of these demands but imposed a compensation: all employees should have access to a corporate‐level complementary health insurance, to be funded at least 50% by employers. An agreement was finally reached in 2013 on the basis of this compromise—without, however, the signature of the CGT.

Albeit officially hostile to this decision that nullified the governmental proposal, the FNMF was unable to take up a position in this debate. Indeed, the federation could not follow up on its initial claims because many mutual benefit societies that were already well established in the field of corporate contracts saw this policy as a lucrative commercial opportunity. Others had even built strategic alliances with IPs favorable to it. Ultimately, the transcription of the agreement into law (the Employment Security Act)22, 23 constituted an important milestone in the generalization of complementary health insurance in France.

The government tried to present this change as having two significant advantages. On the one hand, it would reinforce the role of unions and employer representatives who had recently faced an important decrease in their prerogatives in terms of governance of the statutory regime. On the other, it would considerably extend the level of coverage of complementary health insurance. It would thus provide the same benefits that the initial governmental proposal had aimed for.63 However, it should be noted that a reduction of health inequalities, which was the main purpose of the first attempts to generalize complementary health insurance, was shelved during this reform. Since generalization is now to be enacted at the scale of corporations and on the basis of employment, it de facto excludes more marginalized populations, such as retirees or the unemployed. If CMU‐C and ACS have nevertheless been maintained, the issues related to access to these devices have not been dealt with. More fundamentally still, this reform has created new space for competition among private health insurers. Indeed, it established the conditions for a displacement of the focus of PHI, traditionally dominated by individual‐level contracts. Since 2013, one can thus note a progressive rise of collective, corporation‐level contracts (Figure 3).

Figure 3.

The Rise of Collective Complementary Health Insurance Contracts^a

^aData adapted from ACPR, DREES, CMU Fund2, 4, 25, 54 [Color figure can be viewed at http://wileyonlinelibrary.com]

IPs and insurance companies have been the main beneficiaries of this change. In many respects, the consequences of the Employment Security Act are similar to the development of European legislation and to the recognition of risk management through MCOs. This is because it develops and intensifies competition among private health insurers in a new area, that of corporations. It has also reinforced the homogenization of different actors. Until then, few mutual benefit societies developed activities at this scale. They thus have “much more to lose in terms of market share.”64 To meet this challenge, they have been incited to build strategic alliances with IPs that are far less well established on collective contracts—indeed, to a lesser extent, this is also the case for insurance companies. Correlatively, this situation appears likely to favor the further development of risk management through encouraging MCOs to offer the most “cost‐effective” contracts. However, these later developments are first and foremost an unexpected result of decoupling the insurance and health sectors. Different agendas are pursued in each area.

These above‐mentioned effects are thus unexpected outcomes of the interactions between these policies. Most of them—increased competition, development of risk management, financialization of nonprofit actors—were not expected or intended by the reforms adopted at the EU or national levels. They are largely the product of the fact that EU‐led reforms have almost never considered the health or social dimensions of insurance activities; and of the fact that health care reforms involving the private sector at the national level have not considered the effects of the previous ones on private health insurers, their organization, and their strategies.

Europeanization, Collectivization, and Welfare

According to this work, PHI in France can be said to be facing two mutually reinforcing trends that have induced deep institutional changes in the field. The first is Europeanization, as exemplified by the Insurance and Solvency II directives; the most salient legal and regulatory features of the sector have now been fully integrated within European rules and norms. This dimension will almost certainly have major political effects in the coming years, notably in relation to future legislation. Most of them are likely to be European transpositions of international agreements on insurance operations, especially in terms of financial regulation or insurance product distribution. One example is the recently adopted covered agreement between the EU and the United States, which will introduce a common prudential regulatory framework for insurance activities on both sides of the Atlantic. Such a trend might be both slowed down and, in other respects, reinforced by more specifically national evolutions.

The second dimension is best described as collectivization and is partly autonomous from EU legislative developments. The rise of MCOs, especially through huge industrial platforms involving IPs, mutual benefit associations, and insurance companies, is largely independent from the EU‐scale political agenda. The same applies to the consequences of France's Employment Security Act, an agreement that was passed without any linkage to European constraints. If they certainly occurred within a context of intensified competition, these two evolutions tended for their part to induce a collectivization of complementary health coverage. This dimension is observable through the development of collective, corporation‐level contracts, weakening the principles of free and voluntary subscription of PHI contracts that had previously been a major feature of the French system ever since the foundation of the NHI system. Even this change could incite private health insurers to differentiate among themselves in order to meet this new demand. One might also expect a standardization of their contracts in the long run in relation to the size and form of the different types of firms. Such a development would, again, contradict the traditional principles of PHI in France, which has historically specialized in individual contracts. Collectivization is also visible through the development of MCOs, since they constitute the basis for a symmetrical standardization of care and health products.

If freedom of choice is still guaranteed in the PHI market, it is limited de facto by an intensification of EU regulation and fiscal constraints on insurance contracts on one hand, and by managed care development and corporation‐level contracts on the other hand.

According to the most recent figures, this change from voluntary to mandatory complementary health insurance, and from individual to collective contracts, may also have deeper consequences. If we still lack empirical data on the cross‐health consequences of the interaction among all of these reforms, a recent study has shown that the generalization of complementary health insurance at the corporation level, particularly, would deteriorate welfare.65 On the basis of a sample of 6,122 individuals, the authors demonstrate that this policy could reduce welfare for half of the population, since employers are likely to pass on through salaries the cost of insurance contracts. From a collective perspective, the gains for beneficiaries would be counterbalanced by the loss incurred by persons covered by an individual contract or by being forced to insure after the reform. The only net beneficiaries (around 7% of the population) would be short‐term unemployed persons, who could keep their former complementary health coverage.65 Other recent (2018‐2019) data suggest a decrease in the level of coverage of PHI, which has dropped three to five points (from between 95% and 93% to 90%) over the past two years.66 The same study also shows that 17% of French citizens would be tempted by self‐insurance. Such disaffection could be explained by a disappointment related to the new corporation‐level contracts as expanded by the Employment Security Act.

Some conceptual and practical lessons might be drawn from this case study. What it has described is a process of institutional decoupling of the motives and the policy orientations within the two areas on which PHI is anchored: the insurance and health care sectors. Unexpected results of past policy choices and policy feedback are common findings of health policy and administration studies. This paper has shown that the coevolution and the nonintegration between these two sectors—which could have been subfields of the health care sector—are powerful sources for such outcomes. Irrespectively of the role it plays within the overall health care system from one country to another, PHI, particularly, is exposed to such cross‐sectoral evolutions that may have deep consequences on the provision of several health benefits or services. If such a disconnection between two sectors has clearly been accentuated by a segmentation between EU and national levels, research findings are likely to be the same wherever health care and insurance sectors are highly differentiated, regardless of the level of governance where these evolutions occur.