In the 1970s’ Chile, neoliberal reform ideas entered economic policy under Augusto Pinochet, who took power in the then Marxist-dominated country in 1973. The roots of this change are traced back to the so-called Chicago Boys, who—although born in Chile—were educated at the University of Chicago as part of a cooperation with the Universidad Catolica and later spread the monetarist ideas in key positions in academia, business, and politics (Brender, 2010; Kogut & Macpherson, 2008). While there is a consensus that they can be described as technocrats who brought neoliberal thinking into Chilean government and society, some even see them as “revolutionary vanguard” (Clark, 2017). Importantly, however, both members and non-members of the group refer to them as the “Chicago Boys” and irrevocably associate them with a set of policy ideas. As such, they exemplify the impact that a shared biographical connection, based on shared education, for example, can have on long-term cooperation and change in policy.

Biographical ties and a resulting sense of belonging associated with policy ideas thus have the potential to influence long-term interactions between individuals. The Programmatic Action Framework (PAF) is a comparatively recent perspective that integrates this idea into policy process research (Bandelow et al., 2021; Hassenteufel & Genieys, 2021). By systematically incorporating elements from public administration, elite sociology, and social psychology (particularly the role of social identities), the PAF adds a complementary lens to explain policy change and stability. With regard to the former, the PAF integrates the assumption that the actors close to the bureaucratic state apparatus exert a direct influence on policy formulation. The sociology of elites sees the trajectories of policy actors as roots for their behavior, which is taken up in the PAF by looking at shared biographies as a source of cooperation between policy actors. Finally, the Social Identities in the Policy Process (SIPP) perspective (Hornung et al., 2019) provides an explanation for the long-term stability of social groups based on distinct social identities. These can include biographical/demographic identities, but—in the long run—also programmatic identities, which specifically refer to groups that form around a policy program. The observation that so-called programmatic groups push their policy programs until they achieve policy change is called programmatic action and gives the framework its name.

Research has noted that the shape of programmatic groups may vary depending on the political institutions in which they form (Hassenteufel et al., 2010), but has so far not addressed how these institutions influence the occurrence of programmatic action. However, to make the theoretical perspective a valuable addition to comparative public policy, it is indispensable to reflect on the institutional conditions necessary for programmatic action to be observed. This gap will be filled by this study by answering the question: Under which institutional conditions programmatic action does occur in the first place and how do political institutions contribute to the success or failure of programmatic groups? Given that the PAF has evolved from the programmatic approach, which was originally developed against the background of French policy-making and the specificities of the French political system, one might think that the PAF represents a unique account of French policy processes. However, PAF applications to other political systems with different institutional settings in terms of federalism, corporatism, and the structure of policy sectors show that the PAF actually demonstrates some traveling capacity across countries and policy fields (Davidian, 2021; Duque, 2021; Hornung & Bandelow, 2020). The present study complements PAF research with a systematic analysis of the institutional preconditions necessary for programmatic action to come about. It thus answers the question of the institutional conditions under which programmatic action takes place—an endeavor that has not yet been explicitly considered in previous research.

To this end, a systematic analysis of instances of programmatic action in two institutionally different states is conducted: France and Germany. This selection is based—in addition to the institutionally different conditions—on the premise that programmatic action has already been observed in these countries (Genieys & Hassenteufel, 2015; Hornung & Bandelow, 2020). The policy area on which the analysis focuses is that of health policy, partly because health policy is particularly suited to outlining institutional conditions of federalism and corporatism that are less visible in other policy areas.

1.1 Theoretical and Empirical Puzzle

Although formally established in 2018, the PAF’s ideational history dates back to the earliest studies of programmatic actors within the programmatic approach. Focusing on the evolution of French health policy since the 1980s, Genieys and Hassenteufel (2015) conducted a seminal study of what he called the emergence of a new welfare elite. He characterized this elite in terms of joint training at the École Nationale d’Administration (ENA) (National School of Administration) and mutual cooperation when these elite actors later occupied key positions in the French government. The welfare elite vigorously defended the health care sector against the austerity measures proposed by the “austerians”, which followed exactly the strategy exemplified by Ronald Reagan and Margaret Thatcher to dismantle the welfare state (Jensen et al., 2019). By protecting the social policy sector from the effects of these measures, the actors of the welfare elite were appropriately called the “new custodians of state” (Genieys, 2010). Protecting the state in this context means not only maintaining social spending to ensure financial strength, but also maintaining regulatory powers and authority in the social security sector (Genieys & Joana, 2017, p. 332). This relates to what has been explained in the discussion of the dismantling of the welfare state as the size of the public sector and administration, paradoxically a strong state as a result of neoliberal reforms (Gamble, 1994).

The neoliberal turn with which this introductory chapter opened was thus not only visible in Chilean economic policy under the influence of the Chicago Boys. At least since the 1970s or 1980s, welfare state policies in industrialized countries have entered an era of austerity (Finseraas & Vernby, 2011; Pierson, 2002). This is visible through the increasing use of cost-containment measures (Bonoli & Natali, 2012, p. 6), thus underpinning the argument of a neoliberal direction in which social policy has been moving since then (Ryner, 2009). In the Anglo-Saxon countries of the United States (US) and the United Kingdom (UK), retrenchment and austerity reforms were adopted under Thatcher and Reagan, respectively, with the goal of dismantling the welfare state, but they largely did not produce the desired results (Pierson, 1994; Starke, 2006, p. 105).

The conflict between custodians and austerians is very specific to the French case and an ideal-typical pattern that rooted primarily in the analysis of biographical trajectories. Depending on the positions the actors took in the social security or financing sector, they can be assigned to one of the two sides of the conflict. The analysis thus followed a sociological research design of elite research. Adapting this lens to policy research and countries other than France therefore requires a generalization of the conflict presented and a modification of the actor concept. As the programmatic approach was modified in light of the emerging PAF, the term elite was later dropped in favor of the framework-related terms programmatic actor (individual actor) or programmatic group (collective actor). This not only facilitates the delineation of actors by circumventing the need to define what comprises an “elite”, but also ensures transferability to other political and education systems, where elite thinking is less inherent.

To briefly clarify the roots of the “programmatic” term: The original perspective on “custodians” versus “austerians” was associated with the idea that the two sides were advancing different policy programs—long-term strategies and visions in shaping the health care sector—one pursuing the path of neoliberal austerity and the other that of social protection. The programmatic approach assumed competition between elite actors who formulate their views on the sector through policy programs and use their intellectual and power resources to implement them (Genieys & Hassenteufel, 2015, p. 281f). At least in health policy, reform paths seemed to follow a coherent policy program of a group of programmatic actors who used their resources to implement their ideas and compete against the so-called austerians in the struggle for authority.

Although the genesis of the PAF starts from a very specific institutional setting of French policy-making, with a powerful bureaucracy and an executive elevated by the ENA, programmatic action also occurred in German health policy, characterized by completely different institutional characteristics of both the political and the health care system. Looking at the policy processes in German health policy since the 1980s, there are interconnected policies that presumably follow a common coherent thread—a policy program—that cannot be explained by existing theoretical perspectives in policy process research (Hornung & Bandelow, 2020). Between the early 1990s and 2011, a programmatic group around the program “Competition in a Solidaristic Framework” (Knieps, 2017, p. 12) shaped German health policy (Hornung & Bandelow, 2020). The current version of the PAF has overcome the divide between “custodians” and “austerians” visible in France and adds to policy process research in general the explanation of programmatic action for policy change. The applicability of the PAF in policy process research has since been tested in several studies. However, although existing research demonstrates the applicability of the PAF in different institutional contexts, it is still under-theorized and insufficiently explored which institutional predispositions actually enable or hinder programmatic action. It is this theoretical and empirical puzzle this study ties in with.

As a consequence of the above puzzle, the guiding research question to be answered in this contribution arises systematically from the theoretical under-specification of the PAF in terms of the institutional opportunities and constraints of programmatic action and the empirical puzzle of why the PAF has developed the same explanatory power in terms of observed policy processes and policy outputs in the most diverse institutional settings of France and Germany. The overarching research question is therefore:

Under what institutional conditions does programmatic action take place? In other words: Under what institutional conditions do programmatic groups form on the basis of shared biographies and influence the policy process from policy formulation to policy adoption with their pursued policy program?

Specifically, and in concretization of this research question, it is also the central aim of this study to investigate which circumstances enable or hinder programmatic action, from programmatic group formation to programmatic success. To what extent do political institutions contribute to the success and failure of programmatic groups? The argument to be defended is that programmatic groups can form in any country, regardless of structural, institutional, and political embeddedness, but that certain institutions facilitate and hinder the formation and success of programmatic groups and policy programs. The empirical analysis will identify generalizable institutional conditions that are necessary for the emergence of programmatic action. The innovative contribution to policy process research consists above all in the fact that the institutions that are relevant for programmatic action are not the same as those usually considered in the fields of comparative politics and policy process research.

Answering the research question posed provides insights into processes of policy-making in contemporary France and Germany. It must be emphasized, however, that while observing parallel programmatic action in French and German health policy over a period of time may support the usefulness of the PAF, it does not equate to universal applicability of the PAF in every context. It merely supports the claim that PAF offers explanatory power in cases where other policy process frameworks tend to overlook important explanatory aspects, namely biographical trajectories and social identities connected to policy programs. This does not preclude, but in fact supports, the claim that PAF may well fail in other cases, where other theoretical perspectives fit much better.

1.2 Methodology

To gain insights into the processes of policy-making and the institutions that drive programmatic action, it is necessary to conduct an in-depth analysis of these processes. Since the research questions do not aim at examining average effects of variables on outcomes, but take a process perspective on the institutions of programmatic action and thus require fundamental insights into the policy process, a qualitative research design suggests itself. There is a standardized procedure for a study of programmatic action that includes six steps divided into three main tasks (see Fig. 1.1). These consist of a positional and sociological analysis of programmatic actors, an analysis of the links between programmatic actors and a policy program, and an analysis of the power of the programmatic group in terms of its resources and impact in the overall policy process (Hassenteufel & Genieys, 2021).

Fig. 1.1
A block diagram for the research protocol of programmatic action. It has 3 tasks, each with 2 analytical steps. The diagram has research material and method for each step.

Research protocol of programmatic action in policy process research. Source: Own visualization, slightly modified on the basis of Hassenteufel and Genieys (2021, p. 34)

Full size image

Institutions can be relevant at any level of this research protocol. They indicate the key positions that actors must occupy and influence the trajectories that policy actors reveal, they can be an expression of policy programs and a frame for programmatic actors to promote their policy program within. They can be critical in the allocation of resources to programmatic actors and influence the implementation of the policy program. Consequently, each of the three steps is carried out for the cases of programmatic action studied in order to be able to uncover the institutional effects on the whole process of programmatic action. To this end, the empirical analysis draws on several data sources. Official legislative texts, documents of the legislative process, and public reports and newspaper articles are used to identify programs. In addition, biographical files and information from government or analytical documents are used to analyze actors. Finally, interviews are used to obtain information about the entire process of programmatic action; these also include insider information and regular (informal) exchanges with actors in the health system. In particular, the expert interviews with key actors make it possible to gain a conclusive impression of the precise institutional opportunities and constraints they faced during their time as programmatic actors. It is therefore possible to augment valid information with interviews with actors who were directly involved in the policy-making process to illustrate the roots and trajectory of reforms under different institutional conditions. The multiple use of data from documents, publications, and expert interviews ensures validity and reliability by cross-checking information obtained through one method with information extracted from the other.

However, before proceeding with the analysis, it is necessary to present a reasoned case selection that allows the subsequent findings to be placed in the adequate context of generalization. To this end, the study is comparative in nature and follows a case selection based on the method of agreement dating back to Mill (1848). This method suggests that to explain the same phenomenon (programmatic action), a selection of cases with largely varying conditions (in this case institutional) allows for identifying the conditions that are necessary circumstances for the phenomenon.

In order to reveal the institutions relevant to programmatic action and to emphasize that these institutions differ from those prominently known and thematized in public policy research, it is necessary to select for empirical study cases that are institutionally most different in a given policy sector. At the same time, however, they must be cases in which programmatic action has been observed in the same policy sector over a similar time period. In the fairly recent history of the PAF, it has been noted that an application of the framework appears most promising in sectors close to the state with a considerable degree of state control, and in sectors that require sectoral policies to address structural, organizational challenges of fundamental but conflicting policy goals. Programmatic groups are rather found in sectors close to the state because a programmatic group must involve bureaucrats and, where appropriate, members of indirect public administration (self-governance) to bring together diverse resources and positions that enable coherent and consistent implementation of the policy program. Sectors close to the state are those in which the state acts not only as a regulator but also as a provider of services. Striking examples are health policy, education policy, and transportation policy (Mayntz & Scharpf, 1995). In these fields, a policy program developed by actors within the state administration and other sectoral actors is more likely to prevail than in policy areas where market-based governance is more present. PAF analyses focusing on health policy have been conducted in the different national settings of Spain, the UK, the US, France, and Germany (Hassenteufel et al., 2010; Hornung & Bandelow, 2020).

Taking these examples as a starting point and sticking to the policy sector of health policy, this analysis will take a closer look at two cases: The French and German political systems are characterized by a great divergence with regard to their institutions. On both the executive-parties dimension and the federal-unitary dimension of Lijphart’s classification of types of democracy, France and Germany each occupy the other half of the scale, placing them in the opposite quadrants of the coordinate system for both dimensions (Lijphart, 2012, pp. 244-245). France and Germany thus differ in many respects: France is characterized by a pluralistic interest group system, centralized decision-making processes, a government with a strong executive in a semi-presidential system, and a majoritarian electoral system. Germany is a parliamentary democracy with a special role for subnational states (“Länder”) through the second legislative body of the Bundesrat, a tradition of strong corporatist settings, and a proportional representation electoral system. In particular, the opportunity structures for policy actors inherent in federalism, corporatism, and the adjunct number of veto players are most different and therefore provide policy actors with different opportunities to pursue strategies in policy-making. The selection of France and Germany as case studies therefore follows the call of public policy scholars to select cases that say something about the link between systems and phenomena and to use the cases to systematically reveal these connections (van der Heijden, 2014, p. 36).

As regards the period of analysis, there are clear analyses of the French programmatic elite in health policy that emerged in the 1980s and secured its influence on policy-making at least until the mid-2000s (Genieys & Hassenteufel, 2015). The health care reform under Nicolas Sarkozy, entitled the Hospital, Patients, Health, and Territories Act (Loi Hôpital, Patients, Santé et Territoires; HPST), was also part of this ongoing success of a welfare elite toward a strengthened state in times of austerity (Simonet, 2018, p. 3). It is this welfare elite that will be the focus of later empirical analysis in order to understand what institutional conditions enabled its continued success. Continued because even in the current situation of the Corona crisis, the programmatic group appears to continue to have influence (Hassenteufel, 2020). The period of analysis in France spans from 1990 to 2020, a total of 30 years.

In Germany, the duration of the programmatic group was somewhat shorter, although it is debatable whether reforms in the 2010s can still be counted under the policy program label. The Pharmaceutical Market Restructuring Act (Arzneimittelmarktneuordnungsgesetz, AMNOG) appears to be the most recent health care reform that clearly falls under the program of introducing competition into the health care system with the goal of increasing cost efficiency and solidarity (Dingermann, 2013; Herr, 2013). Although the vision of competition seemed unbroken in 2014 (Manzei et al., 2014), the legislative period under health minister Hermann Gröhe 2013–2017 did not continue this line (Bandelow et al., 2019, p. 463). For this reason, the period between 1990 and 2011 is analyzed for the German case. However, in order to determine, in the true sense, the institutional factors that favor programmatic action, it is also necessary to examine a case in which programmatic action did not occur. In this case, too, the institutional circumstances should lead to the expected result. Therefore, the period after programmatic action in Germany, from 2011 onward, is also included in the analysis. The fact that the period of analysis from 2010 onward in the aftermath of the financial crisis shows a persistence of the programmatic group in France and an end to the programmatic group in Germany makes it all the more possible to clarify the institutional conditions of programmatic action.

In order to analyze the factors of success, it is necessary to define what a successful programmatic group and policy program are. Success is multidimensional and can refer to the realm of process, program, and politics (McConnell, 2010). Specifically, program and political success are relevant to PAF, but in a different understanding than in existing research. Since programmatic actors are not elected and thus cannot achieve political in terms of electoral success, their success is defined by having more authority and resources in the sector. Thus, a programmatic group is successful if its members move up the career ladder, hold key positions in the sector for a long time, and have resources (both financial and regulatory) to influence their policy sector. Indicators of financial and regulatory resources would be increased budgetary importance to their policy sector (e.g., increased spending) and an increased policy authority for health care decision-making.

The success of a policy program in terms of the PAF is unrelated to a normative outcome, that is, an assessment of whether the adoption and implementation of a policy has led to the desired result, at least insofar as it does not affect the overall goals of the policy program. To give an example, if the vision of the policy program is greater interlocking of care across sectors, and one reform step is the establishment of medical care centers overseen by multiple institutions to coordinate care across sectors, it is not of interest to the programmatic group promoting the policy program whether these medical care centers actually improve care in a normative sense by, for example, ensuring access to specialists or improving quality of medical interventions. As long as the overall goal of a closer sector integration, which is the vision of the policy program, is achieved, policy outcomes and evaluations such as good or bad policy are of secondary, not primary, interest. Policy program success is defined here as the adoption of a policy program and its persistence, which is completely independent of normative effects but depends solely on the interests of the programmatic actors who profit from its implementation (Wenzelburger & Zohlnhöfer, 2015, p. 19).

Revealing the factors of success of programmatic groups and policy programs that are embedded in the institutions that enable or hinder programmatic action is the central research interest of this study. However, while the success of policy programs and programmatic actors is easier to determine with an outside perspective, understanding the role of institutions is only possible by gaining a detailed picture of the policy process through inside information. Therefore, expert interviews with programmatic actors identified in the previous two analytical steps were chosen as the research material and method for this final analytical task. The interviews also subsequently allow validation of the previously generated results on the policy program and programmatic actors.

1.3 Outline

Taken these considerations together, it is worth recapitulating what this book does and does not aim to do. Starting from the observation that programmatic action occurred in the health care systems of France and Germany despite their most different institutional designs, the main argument to be advanced in the following chapters is that there are institutional predispositions in each nation that allow for the formation of programmatic groups and the resulting promotion and implementation of their policy program over several years. While there is already evidence of programmatic action in France and Germany during the period of observation, it has not yet been systematically analyzed what institutional factors enabled programmatic action. This is particularly relevant given that the political institutions of federalism, corporatism, and party systems differ considerably in the two states. To what extent can the institutional preconditions be generalized on a broader level to make PAF applicable beyond the French, German, and European borders? To achieve this goal, the main contribution of this book is to conduct a detailed systematic and comparative analysis of the presence and absence of programmatic action in French and German health policy over the past 30 years, and to identify the commonalities that have enabled this programmatic action in France and Germany.

It is worth noting that the analysis of programmatic action is possible by starting from one of the two sides of the pole: programmatic actors or policy programs. Given that this book is situated in the field of policy process research, the variable of interest remains policies. Consequently, the empirical analysis begins with a systematic overview on relevant sectoral policies. It devotes its attention, firstly, to the analysis of policy programs that encompass a consistent and coherent vision for the health sector, realized in separate but few comprehensive acts, accompanied only by minor adaptive measures. Secondly, tracing this vision back to the actors and the ideational roots of its content, the analysis reveals personal connections between actors and major, comprehensive health reforms, rather than those that only entail regulations for a very specific aspect of the health care sector (e.g., hospitals) without affecting another. Network graphs visualize the biographical links between programmatic actors. The underlying data stem from systematic analyses of official documents and biographical files, as well as interviews. Thirdly and finally, these interviews with key informants and members of the programmatic groups are used to address the central research interest, to uncover the policy processes during each period of programmatic action, with a specific focus on the institutional opportunities and constraints that led to the formation of these programmatic groups and that they faced in realizing their policy program. Thirty-one qualitative in-depth interviews reveal which institutions proved conducive to programmatic action and thus relevant to the ultimate formation of programmatic groups, their policy programs, and the success of both.

Although the health policy sector under investigation is related to the welfare state, the study conducted does not directly aim to explain welfare state retrenchment or expansion. As such, it is not designed to provide explanations for why a specific content of reforms was chosen, apart from question of the extent to which the content can be considered as an explanatory factor for the success of the policy program. Rather, the analysis sheds light on the opportunities and constraints that programmatic groups faced when promoting their policy program, both regarding the institutionally pertinent factors for programmatic group formation and programmatic groups’ as well as programs’ success, and regarding the group-specific and program-specific characteristics that facilitated that success.

Overall, this book is divided into four parts, of which this introduction is the first. The second part is devoted to theory. This first covers the theoretical foundations of the PAF laid in Chap. 2 (“Programmatic Action and Policy Processes”), including the debates to which it contributes and a delineation of the PAF from existing perspectives. Following this preliminary work, Chap. 3 (“Political Institutions and Public Policy”) links the PAF to existing institutional perspectives and derives hypotheses about the potential influence of institutions on programmatic action. Hence, this chapter is particularly designed to highlight the institutional circumstances that are typically assumed to influence policy-making and potentially explain the occurrence of programmatic action in the two countries. Based on the interim conclusion on the influence of institutions, the empirical part (part III) contains a detailed analysis of programmatic action in France and Germany (Chaps. 5 and 6), after having provided an overview of the institutional settings of health policy in the two countries (Chap. 4). The last part consists of two concluding chapters: The first summarizes the results of the empirical analysis with respect to programmatic action in France and Germany with a view on the institutions necessary for programmatic action (Chap. 7). The second (Chap. 8) answers the overarching research question which commonalities in most different political systems are drivers of programmatic action and thus facilitate or hinder the emergence and success of programmatic groups and policy programs. It also addresses the challenge of institutional change to programmatic action before emphasizing the limitations and directions for future PAF research that result from the study.