Political Institutions and Public Policy

Abstract

In order to shed light on this missing link between programmatic action and political institutions, this chapter reviews how institutions are currently discussed in policy process research in order to derive hypotheses that may explain under which institutional conditions programmatic action should or should not take place. The goal of this overview of the state of the art is twofold. On the one hand, it serves to assess the contribution of the PAF to existing theories of the policy process and the understanding of institutions in it. In doing so, it becomes clear why a new theoretical lens is needed and where and why the PAF is able to fill gaps left by others. On the other hand, the established approaches to explaining policy change and stability with respect to policy processes contain assumptions and hypotheses about the role of institutions that can be integrated into the PAF and help sharpen the analytical power of a look at the institutional conditions for programmatic action. They do this by formulating mechanisms between theoretical concepts that can also be adapted, or at least assumed to be relevant, to the formation of programmatic groups and the success of the group and its program. At the very least, they lay the groundwork for the question that asks about the influence of institutional settings familiar in comparative politics on policy change.

In order to shed light on this missing link between programmatic action and political institutions, this chapter reviews how institutions are currently discussed in policy process research in order to derive hypotheses that may explain under which institutional conditions programmatic action should or should not take place. The goal of this overview of the state of the art is twofold. On the one hand, it serves to assess the contribution of the PAF to existing theories of the policy process and the understanding of institutions in it. In doing so, it becomes clear why a new theoretical lens is needed and where and why the PAF is able to fill gaps left by others. On the other hand, the established approaches to explaining policy change and stability with respect to policy processes contain assumptions and hypotheses about the role of institutions that can be integrated into the PAF and help sharpen the analytical power of a look at the institutional conditions for programmatic action. They do this by formulating mechanisms between theoretical concepts that can also be adapted, or at least assumed to be relevant, to the formation of programmatic groups and the success of the group and its program. At the very least, they lay the groundwork for the question that asks about the influence of institutional settings familiar in comparative politics on policy change.

With these two goals in mind, the following chapters systematically review the state of the art based on five main lines of literature and with a particular focus on institutions. Overviews of theories of the policy process have been published in the form of textbooks in various languages (Maillard & Kübler, 2015; Schubert & Bandelow, 2014; Weible & Sabatier, 2017; Wenzelburger & Zohlnhöfer, 2015) and serve as an initial source to identify the sample considered here. What is innovative about the PAF and previously neglected in these textbooks is the focus on long-term periods of programmatic stability in policy change, traced back to programmatic groups and the social psychological explanation of individual behavior as the basis for collective action and resulting policy change. When one thinks of explanations for programmatic stability, institutionalist theories immediately come to mind. Thus, a first look at the research focuses on explanations for policy stability. This begins with institutional explanations of federalism and veto players and continues by examining the involvement of different interests in the policy-making process, for example, through corporatist structures or public protest. A particular emphasis is then placed on the role of bureaucratic recruitment and policy advice, as these institutions are assumed to be particularly relevant to programmatic action.

However, before beginning with the outlined theoretical groundwork, it is necessary to define institutions as they are to be understood from a PAF perspective. In political science, a general distinction is made between a broader and a narrower understanding of institutions, with the narrower understanding commonly associated with the “old institutionalism” (Peters, 1996). Within the neo-institutionalist tradition, institutions are defined by looking at the influence they exert on actors and thus include the informal rules that guide actors’ behavior (Wenzelburger & Zohlnhöfer, 2014, p. 314). The PAF uses a broader understanding of institutions as not only the rules governing behavior, but also the existence of agencies and organizations that shape the interaction between individuals and thus contribute to group formation and group action. This view likewise includes institutions that can stand in the way of programmatic groups as they seek to translate their program into policy decisions. In this way, the definition is similar to that of March and Olsen (2008, pp. 4-5) that institutions are determined by repetitive processes that need to be explored in terms of their effect on individuals and the resulting institutional stability (Capano, 2018), and that institutions (rules, but also agencies and social systems) are also carriers of identities that influence policy actors.

3.1 Veto Players and Federalism

Institutionalist analysis is closely related to the notion of which institutions impede or enable change. In comparative politics, institutional analysis is often linked to whether the political system is parliamentary or presidential, although the usefulness of this dichotomy for analyzing institutions is debatable (Cheibub et al., 2014). Essentially, the difference is whether or not the government is dependent on parliament, such that parliament can remove the government. It has been argued previously that the apparent balance between cabinet and parliament is symbolic rather than politically relevant because of the interdependence of offices and support on both sides (Price, 1943, p. 319). From an office-seeking perspective, the parliament will do everything possible to avoid dissolution and new elections. In presidential systems such as the US, the cabinet and parliament are separate and independent because of independent elections. The archetype of semi-presidentialism was developed by Maurice Duverger with reference to France, although it is controversial whether France can be considered this semi-presidential archetype (Elgie, 2009) and French political scientists reject this typology (Pasquino, 2005).

Among the various typologies in comparative politics is that of Arend Lijphart (2012), who differentiates patterns of democracy based on two dimensions. These dimensions consider, first, executive capacity, including party systems, cabinet functioning, legislative-executive relations, electoral systems, interest groups, and, on a second dimension, the degree of federal or unitary organization of politics. The extent to which these institutions contribute to or prevent policy change is a matter of debate. Pierson (1995, p. 472) outlines the various interaction effects that federalist structures have on social policy change.

In part because of this complexity, which makes it difficult to compare analyses across different political systems, veto player theory has emerged as an alternative lens on institutions (Tsebelis, 2002). Rather than analyzing the entire institutional setting of a given political system and its consequences for policy-making, this approach is interested only in those institutional actors that are necessary to change the status quo (Tsebelis, 2000, p. 442). Institutional veto players are metonymic with veto points (Immergut, 2006, p. 568). They describe the very institutions that are mandatorily involved in adopting a policy. When within an institutional veto player, multiple parties are required for a majority decision, they are referred to as partisan veto players (Tsebelis, 1995, p. 302f). Since the work of George Tsebelis, there has been an increasing number of studies on veto players and their power that test the five main hypotheses of veto player theory (C. Jensen & Wenzelburger, 2020; Wenzelburger, 2011; Zohlnhöfer, 2003). These postulate that the greater the number of veto players, the more cohesive they are, the greater the ideological distance between the veto players and between the current and previous governments, and the shorter the duration of government, the less likely a change in the status quo will occur (Schmidt, 2002, p. 152). In this regard, research provides evidence on the relationship between veto players government types (Angelova et al., 2018), institution-building (Hallerberg, 2011), globalization pressures (Ha, 2007), international trade policy and the change in tariffs (O’Reilly, 2005), the rule of law (Andrews & Montinola, 2004), budget composition (Tsebelis & Chang, 2004), and many others. In recent years, research also indicates that legal actors are conditional veto players depending on whether their preferences are absorbed by another institutional veto player (Brouard & Hönnige, 2017). In order to assess the institutional settings in the two most different contexts in this respect—France and Germany—the institutional and partisan veto players in these states are reviewed below.

For parliamentary systems, among which he also classes France, Tsebelis (1999, p. 593f) argues that, in addition to the governing parties, the second body of the legislative arena or the head of state could potentially be considered veto players. Since the French president, as head of state, has some veto power over government decrees, he could be seen as an institutional veto player only in times of divided government. However, Tsebelis (1995, p. 306) also emphasizes France as an exceptional case in which a strong president and a bicameral legislative system are de facto coded as a system with only one institutional veto player, namely the National Assembly (NA). Since the beginning of the Fifth Republic in 1958, French parliamentarism has been described as rationalized for several reasons. One is the government’s ability to have bills voted on as a package vote; a second is its ability to combine a vote on a bill with a vote of confidence (Hayward, 2004, p. 83). Other limitations include the government’s ability to rule by decree, to set the parliamentary agenda, and the degradation of the parliamentary committees (Keeler, 1993, p. 521f). Another is the lack of a comprehensive right to introduce amendments to bills. Navarro, Vaillant, and Wolff (2012, pp. 58/74) argue that the constitutional reform in 2008 gave greater weight to committee work, in which those who hold multiple offices, in particular, attend meetings regularly, contribute efficiently to tabling legislative amendments, and provide oversight of the government. The reform also strengthened the role of parliament by limiting the president’s ability to couple proposals with a vote of confidence and to require parliamentary approval for certain foreign and domestic policy decisions (Michel, 2008). Now, amendments to bills most often come from the government or parliamentary committees and are also more successful when introduced in committees than in parliament (Kerrouche, 2006, pp. 355-357). It can be concluded that the legislative arena of the National Assembly in France is of particular importance with regard to parliamentary committees, which have a special relationship to the respective ministers (Fasone, 2017, p. 16).

Given that the constitutional amendment in 2008 also allowed the opposition to initiate an abstract review, the Conseil Constitutionnel (constitutional council) was strengthened as a potentially second institutional veto player. Unlike comparable institutions in other countries, the Conseil Constitutionnel is able to exert influence during the legislative process and is therefore quite powerful; it can decide that legislative texts are not in conformity with the constitution and, on this basis, delete parts of legislative texts (François, 1997). With the 2008 reform, it is now also able to retroactively declare laws unconstitutional. In general, however, Brouard (2009) describes constitutional vetoes as electorally incentivized and politicized. They have been used primarily in electoral competition to signal positions to voters and to control the government (Brouard, 2009, p. 386/396). The Conseil Constitutionnel thus represents a conditional veto player, depending on its ideological composition (Brouard & Hönnige, 2017).

A key feature of French semi-presidentialism is the divided executive, where the prime minister sits in parliament and the president resides in the Elysée Palace without having the right to enter the NA. As a rule, the relationship between the president and the prime minister, whom he appoints as well as the ministers of the cabinet (after nomination by the previously appointed prime minister), is cooperative. Since there are only a few months between NA elections and presidential elections, the winning political party is in most cases the same—but only since the 2002 institutional reform (Grossman & Sauger, 2009). Scholars have argued that there is an institutional presidential bias in the sense that the ability to reshuffle governments gives the president dominance over the prime minister in intra-party relations and an opportunity for blame avoidance, but this does not seem to work well (Grossman, 2009, p. 277). Presidential dominance does not stop with the government, but usually extends to the majority party in parliament (Cole, 1993, p. 63). However, dominance over the majority in parliament can tip, especially when the parties securing the majority are in conflict (Lazardeux, 2009, p. 288) and when ideological distance within parliamentary coalitions is significant, weakening party unity in voting (Godbout & Foucault, 2013, p. 324). Indeed, while the prime minister and cabinet are in a weaker institutional position because they need the backing of parliament and the president, the centralization of power at the presidential level makes the president the main figure of accountability and provides fewer opportunities for coalition building to share blame (Bezes & Le Lidec, 2015, p. 502). In the face of strong opposition in parliament, both enacted and failed legislation lower the executive’s popularity (Becher & Brouard, 2020). Finally, regarding this institutional setting in France, the government’s relationship with the president, as well as with parliament, should not be underestimated when explaining reform success.

If the president and prime minister do not share party affiliation, this constitutes divided government, known as cohabitation. Here, the prime minister’s power again depends on the majority in parliament and is linked to whether the majority party expects electoral success in the upcoming elections, which would place the prime minister as a likely presidential candidate (Lazardeux, 2015, p. 88). In terms of legislative efficiency, which is supposed to be lower under divided governments because of the risk of gridlock, productivity levels actually do not differ significantly between divided and unified governments because actors are able to cooperate, but major structural reforms are more likely under a unified government (Baumgartner et al., 2014, p. 444).

Another type of divided government exists in the case of different majorities in the Senate and the NA. Nevertheless, the Senate is not considered a true veto player, since in case of disagreement on a bill, a mediation committee is convened. After its report, however, the NA still has the final decision-making power, so the Senate can delay a law but not prevent it. However, since the Senate in France is indirectly elected, it is mainly local representatives who sit in the Senate, which often reflects the conflict between the national government and local governments, especially the Assembly of Mayors (Assemblée des Maires de France).

German federalism entails the institution of the Federal Council (Bundesrat), which is involved in the legislative process. Although not all laws are subject to the explicit approval of the Bundesrat, it is an institutional veto player for those that are, and especially in policy areas that affect Länder interests (Zohlnhöfer, 2003, p. 130). In health policy, this concerns, among others things, hospital policy (since it is within the competence of the Länder to bear the investment costs for hospitals), emergency care, and regionally active sickness funds (Dent et al., 2004, p. 732). Thus, the Bundesrat represents a second institutional veto player in addition to the governing parties. Moreover, it can be argued that the degree of cohesion within the veto players is higher, at least as far as the parties are concerned. This is not because there are no intra-party conflicts in Germany, as a number of studies confirm (Debus & Bräuninger, 2008; Marx & Schumacher, 2013; Seeleib-Kaiser, 2010; Steiner & Mader, 2017), but because legislative processes in Germany are dominated by the parliamentary parties of the governing coalition, and MPs are rarely freed from the obligation to vote in accordance with parliamentary party policy. Even then, party membership exerts a significant influence on individual preferences (Engler & Dümig, 2017). Since the Bundesrat consists of representatives of the Länder governments, and a Land’s vote can only be cast unanimously, the Bundesrat is sometimes misused for party-political and intra-coalitional conflict (Auel, 2014, p. 425; Brunner & Debus, 2008; Lehmbruch, 1998).

The term divided government is also used to describe divergent party-political majorities in the Bundestag and Bundesrat, which is also said to have a negative impact on legislative productivity (Manow & Burkhart, 2011). Similar to France, a constitutional reform was designed to increase the power of parliament, but in the German case at the expense of the power of the Bundesrat (Stecker, 2016). In historical retrospect, though, Germany is characterized as a grand coalition state that permits governing only in coalitions between parties as well as between the federal and the state level (Schmidt, 2008, p. 90). Empirical testing of these contradictory views found evidence of dynamic patterns of policy-making and mixed results for both competitive and consensual policy-making with respect to the policy domain (Breunig, 2014, p. 144). When party coalition dominance differs in the Bundestag and Bundesrat, votes in the Bundesrat are sometimes won by concessions regarding state interests (Ganghof & Bräuninger, 2006, p. 533). Thus, divided government also contributes to a more consensus-oriented relationship between government and opposition in the Bundestag at the outset of legislative initiatives (Hohendorf et al., 2020). Summarizing the relationship between the Bundestag and the Bundesrat, then, one can conclude that both represent arenas that are heavily dominated by party groups and regional interests. The subnational level of government should not be neglected in the policy-making process.

The German government consists of the chancellor and his/her cabinet members, who are politicians appointed by the chancellor, who in turn sets guidelines and organizes the executive branch (Müller-Rommel, 1997, p. 172). Coalition governments at the federal level have always consisted of up to three parties, although the increasing success of the right-wing Alternative for Germany (Alternative für Deutschland, AfD) is challenging the existing party system and making two-party coalitions less likely (Bräuninger et al., 2019, p. 83). A central and, in recent years, increasing role in legislation at the local, regional, and national levels is attributed to the coalition agreement on which the government-forming parties agree (Gross & Krauss, 2019). In this context, the content of such coalition agreements often includes conflict issues on which the parties try to make as few concessions as possible to their respective coalition partners (Klüver & Bäck, 2019). However, this is not always the case, and even the most ideologically distant coalitions can govern cooperatively if the portfolio allocation and sites of conflict resolution are strategically chosen (Saalfeld et al., 2019). The head of state has no veto power and only a representative function, but becomes important in crisis situations such as a lack of parliamentary majority for chancellor and government formation (Hornung et al., 2020).

The systematic analysis of veto players in the two countries helps identify key positions that programmatic groups need to cover—or will aim to cover—to ensure minimal resistance to the implementation of their policy program. This mechanism is similar to that proposed by the veto control hypothesis in analyses of cabinet formation (Eppner & Ganghof, 2017). Thus, the PAF actually draws on and incorporates institutional factors that help or hinder the formation and success of “new alliances” (Hacker, 2004, p. 718). Horizontal collaboration at the local level may prove to be a necessary element in paving the way for major policy change and path dependencies (Hager & Hamagami, 2020). Given the above considerations, the importance of the subnational level of policy-making depends on the degree of federalism. As a consequence, it can be expected that:

Decentralization Hypothesis

In decentralized democracies, the involvement of subnational actors is a necessary condition for programmatic group formation. The involvement of subnational actors then also is necessary for programmatic groups’ success.

In addition to veto players, which can potentially prevent and block reforms, health policy decision-making processes in France and Germany are also shaped by other institutions that can directly influence decision-making. Unlike in the case of veto players, however, this influence is not mandatory. The following chapter therefore focuses on corporatist structures as distinct loci of decision-making, which are also most relevant to health policy.

3.2 Corporatism and Organized Interests

Mediation between collective interests and the state is usually subsumed under the research strands of pluralism, lobbyism, and corporatism (Klenk, 2018, p. 21). Schmitter (1974), in his widely cited state of the art article, defines corporatism as

a system of interest and/or attitude representation, a particular modal or ideal-typical institutional arrangement for linking the associationally organized interests of civil society with the decisional structures of the state. (Schmitter, 1974, p. 86)

In the course of further research, scholars have drawn a distinction between a structural understanding of corporatism (as defined above) and a processual understanding that places more emphasis on the process of policy-making from agenda setting and formulation to implementation. The latter is labeled concertation, while the combination of the two is called neo-corporatism (Baccaro, 2003, p. 685). Similarly, Lehmbruch (1977, p. 95) emphasizes that interests are not only linked and integrated into the policy process, but that interest associations are in effect given responsibility for specific decisions and their implementation, which emphasizes the reciprocal dimension of linkage. Critics argue that the de facto primacy of the state interest over sectoral interests persists and that this even diminishes the earlier competition between interest groups to lobby in the policy process (Cox, 1981, p. 79). Nonetheless, the degree to which interests are involved in the policy process in terms of ensuring compliance with the enacted reforms and providing tying knots for organized groups is an important feature for this study’s interest in explaining collective action in the policy process and programmatic change.

Corporatism has been shown to be a component of consensus democracy (Lijphart & Crepaz, 1991, p. 245) and, although this view has been challenged (Keman & Pennings, 1995, p. 279), has been incorporated in the comparative politics typology of democracies by Lijphart (2012). France and Germany differ most in terms of the relationship between the state and interest groups, both regarding regular exchange and systematic involvement in the processes of policy formulation and adoption. In Germany, recurrent forms of exchange between the state and interest groups in the form of consultations represent a new type of negotiation democracy (Czada, 2015). In France, on the other hand, corporatist arrangements, which are institutionalized forms of governance, are not realized even though there is a willingness on the part of the respective unions and associations to do so and even though there is a need to address financial pressures, including through pro-capital policies and the introduction of competitive elements (Lux, 2015). This is not to say that unions in France do not have some influence on the parliamentary process and that their strikes would not have a significant impact (Mazevet et al., 2018; Weßels, 2007, p. 110).

Many studies in corporatism have focused on the economic and labor market sectors (Heinze & Schmid, 1997; Parsons, 1988). When corporatism refers only to a sectoral alignment of interests and agreement, the term meso-corporatism is usually used (Cawson, 1985). Molina and Rhodes (2002, p. 326) argue for a process-oriented rather than a structure-oriented understanding of corporatism, focusing on actor networks rather than formal institutions. An important aspect for the development of actor networks from corporatist structures is the exclusivity with which associations have a monopoly position in negotiating agreements with the state (Klenk, 2018, p. 22).

With respect to the health sector, corporatist elements are strongly existent in Germany and also present—but to a different degree—in France. The French and German health care reforms in the 1990s had a structural and in some cases negative impact on the actors of self-governance and medical professions. In particular, the inherent goal of cost containment is not attractive to corporatist actors (Godt, 1987). However, the physicians’ associations in both countries were unable to prevent these reforms (Hassenteufel, 1997, pp. 8-9). Following on from the executive-legislative relations outlined earlier, the declining influence of these interest groups in France can be explained by the parliamentary backing of the government in the absence of cohabitation in 1995, the lack of which had previously provided opportunities for interest group influence (Immergut, 1990, p. 397). The strongest organized interest group in the French health care system is physicians, who value their elitist status, resulting from many years of top-level training and a numerus clausus (not abolished until 2019), and protect it against state interventions (Mendy, 2015, p. 9). Some scholars even argue that the group of physicians has political and economic resources that outweigh any formal institutional rules and grant bargaining power in any state (Marmor & Thomas, 1972, p. 438). While the lack of involvement of corporatist actors in the policy process in France allows for a more centralized state-led policy process, primarily through administrative, statist regulation, which has been positively associated with the effectiveness of cost containment measures (Wilsford, 1991), this also presents a downside for health policy-making in France. The high political costs and the manifest opposition, which, although it has fewer means of institutionalized decision-making power, represents a stiff resistance to reforms can only be addressed by centralized hierarchical administrative governance (Hassenteufel, 1996, p. 478). Medical entrepreneurs who aim to shape policy, however, become a grateful type of ally for state actors to drive new reforms (Hassenteufel et al., 2020, p. 54). Thus, it is not straightforward to derive a strength or weakness of the state from corporatist structures, as both views have their justification (von Beyme, 1983, p. 177).

In contrast, Germany has a tradition of corporatist structures and self-governing bodies in the health care system, which means that the welfare associations are assigned tasks and, in some cases, regulatory powers in a reciprocal working relationship with the state (Toens, 2008, p. 102). A key feature of the German health care system is the indirect role of elected officials in the financing and provision of health care services, the responsibility for which is shifted to the self-governance of health care actors (Saltman, 1997, p. S17). The success of a far-reaching health policy reform in Germany in 1992, after almost 50 years of blocking of major reforms by interest groups, can be attributed to a grand coalition compromise (Bandelow, 1998, p. 210). In parallel with the reforms in France, cost containment measures were enforced that had previously been in force but with which the associations of statutory health insurance physicians (Kassenärztliche Vereinigungen, KV) had not previously complied (Giaimo, 1995, p. 364f). Since then, the influence of interest groups has depended heavily on the substantive and discursive congruence of their interests with the politically dominant guiding principles (Bandelow, 2007a, p. 290). On the other hand, the reform also maintained and intentionally strengthened collective agreements negotiated between the social partners (Giaimo & Manow, 1999, p. 978). This can be seen as a first sign of the centralization of competences that progressed in the 2000s and led to a selective strengthening of the competences of certain bodies of self-governance and joint decision-making, while holistically it tended to erode established corporatist structures (Gerlinger, 2010, p. 127f). Recent analyses assess these changes as a dismantling of the established German welfare state model (Trampusch, 2020, p. 159f).

To summarize the institutional perspective of corporatism in terms of its influence on the core interest of this book, programmatic groups and policy programs, there are two concluding observations that will be considered in the following. First, it can be questioned whether corporatist arrangements generally lead to more or less legislative productivity or state effectiveness. Delegating central government decision-making authority to the interest association level offers the advantage of a more cooperative style of policy-making with less substantive and public resistance to reform, but at the same time requires the need for compromise. Following this observation, recent research recommends taking into account the specific institutional environment to assess its relevance (Giuliani, 2016, p. 38). From the perspective of the veto player control hypothesis, regardless of the exact degree of corporatism in a given country, it may be helpful to coordinate legislative proposals with interest associations to gain support and keep opposition low. To this end, and similar to what Bandelow calls discursive alignment (Bandelow, 2007a, p. 290), it is useful to choose a narrative or discursive frame for policy reforms that fits the interests of the relevant stakeholders. In line with the previous considerations on veto players, high delegation of power to self-governing bodies in the health care system could therefore be seen as creating sectoral veto players that can collectively block policies through delayed or misguided implementation. Corporatism also provides a complementary perspective to explain opportunities for group formation and to reveal actors potentially forming joint reform alliances, as meso-corporatism usually provides institutionalized opportunities for exchange (von Winter, 2014, p. 179). One would therefore expect that:

Self-Governance Hypothesis

When competencies are delegated to self-governance of sectoral actors, it is necessary for programmatic action to occur that programmatic actors either emerge from self-governance or that programmatic groups encompass actors from the self-governance to be successful .

Sectoral actors do this either by putting forward policy proposals themselves, forming alliances with or without state actors, or publicly opposing reforms and using narratives to generate acceptance for their views. The ways in which a political system introduces interests either through sectoral actors or the public thereby helps to examine the extent to which such structures influence the process of group formation and what types of groups emerge.

While citizens rarely have direct means of influence, the culture of protest in France is exceptionally strong, as demonstrated most recently by the protests of the so-called yellow vests (Chamorel, 2019). In addition to elections in which they vote for the president and the NA, citizens participate in political debates often through public demonstrations in major cities. One example of extra-parliamentary participation is the “Grand Débat” (great debate) initiated by then-president Emmanuel Macron in early 2019 in response to yellow vests protests against the government. While the government set the overarching themes of ecological transition, taxes/expenditure, state organization, and democracy/citizenship, health policy emerged as another issue on citizens’ side, especially regarding access to health care (Tranthimy, 2019). Thus, while French policy-makers do not face strong blockage of reforms by powerful corporate actors, they do face the challenge of legitimizing reforms to the French public, which, not least because of the party system, indicates a crisis of representation (Shields, 2005). This is not limited to France, as research on the impact of protests on legislative agendas in Western countries, including Germany, shows that protests have an impact, especially on social welfare issues. However, this effect is said to be stronger in majoritarian than in consensus democracies (Bernardi et al., 2020, p. 16f).

3.3 Bureaucratic Recruitment Systems and Elite Building

Elite studies have a long tradition of examining what makes individuals elites and what influence elites have on society and political decisions. Probably the most prominent works are by Mills (2000) and Robert Dahl (1961), whose work is also referred to by Genieys and Hassenteufel (2012). The principle derived from this work, that elites can be identified through a “decisional approach” because they are found in formal decision-making positions, is also the basis of the PAF. There is a general consensus on the definition of elites as political actors with a substantial influence on decision-making (Hoffmann-Lange, 2007, p. 910). But how do actors reach these positions? To assess the likelihood of programmatic group formation in different countries, it is essential to review the education systems of elite recruiting. Knowing which career paths lead to power (i.e., positions in the policy-making process) enables researchers to know where programmatic groups will form and which channels to monitor and merge to trigger programmatic action. The empirical study informed the answer to this question through systematic biographical analyses of programmatic actors. It has been shown that the elite system in France differs from that in Germany.

The elite system in France is essentially shaped by the institution of the ENA and its associated career paths. Actuaries trained in this institution are not only placed in the top positions of administration and politics, they also already form informal networks among themselves, which have a long-term impact on their work (Searls, 1978, p. 171). Thus, the structures in France are much more elite than in Germany; Suleiman Suleiman (1974) also refers to these key actors as the administrative elite. In health policy, Genieys and Smyrl (2008, p. 80) find these actors in the Grand Corps and IGAS, as well as in the central administration of sickness funds. While the path to these positions in France almost exclusively through the ENA, career paths in Germany are more diverse.

Since as early as the eighteenth century, the recruitment processes of senior civil servants in Germany have proceeded through university routes, originally with a strong focus on law studies, which later opened up to social sciences and economics as alternative entry levels into the senior civil service (Derlien, 1991, p. 392). In the case of German health policy, empirical analysis has linked the emergence of programmatic action to the rise of health economists in the system, which contributed essentially to the establishment of a network among these actors (Ulrich, 2012). At the same time, even below the level of politicians, partisanship is a relevant determinant of careers not only in the policy-making process (Bach & Veit, 2018; Kube, 2004). Research from other parliamentary countries such as Portugal shows that career paths that run through parties are more durable than those that run only through the civil service or parliament (Seixas & Costa, 2021). Parties should therefore not be neglected when it comes to the formation of elites. Although electoral programs and party manifestos are distinct from the concept of policy programs as defined by the PAF, one hypothesis could also be that parties in parliamentary systems take on the task of program generation, which rather makes nations with less stable party systems form programmatic groups as an alternative venue for idea generation.

In terms of integrating political elite studies into public policy research, there has been little but important research on how the recruitment of elites and their role in policy processes differs across countries, but how accounting for these differences can inform political science (Blondel & Müller-Rommel, 2007). This is not to say that elite studies in general have a young history; the work of Mills and Dahl and the seminal work of Putnam (1976) prove otherwise. They already saw a small group of elite individuals, characterized by professional and biographical backgrounds and sectoral specialization, as predisposing to influence in a policy area. Yet public policy research and policy processes have placed less emphasis on a systematic examination of individual elite actors and how the structures of elite recruitment as institutional predispositions shape pathways to policy change. This is where the PAF comes in and fills a gap by focusing on elite formation and the construction of elite groups. The 2018 Palgrave Handbook of Political Elites, which summarizes the current state of elite research worldwide (Best & Higley, 2018), provides fruitful insights in this regard. Summarizing research on “executive elites”, Verzichelli (2018) notes that each state is typically led by a small group of executive-elite actors who share a kind of history that binds them together and is characterized by specific relationships between those who come from the political field and those who come from the administrative field of work. The political-administrative relationship is not only a crucial aspect in defining elites, but also in their collaboration, as will become clearer in the following subchapter on the science-policy nexus.

Building on these findings of elite research in public policy, the question arises as to what makes individuals members of elite groups and how they can become programmatic actors. In this regard, the PAF provides a possible link between the often separately discussed background and behavior of elite actors (Rhodes et al., 2007, p. 5). Highlighting the role of the education systems and career paths institutionalized in the ENA as elite-forming institutions in France and the field of study and respective political and sectoral career paths as elite-forming institutions in Germany, the following hypothesis is formulated:

Biographical Identity Hypothesis

Programmatic action depends on the existence of institutions that foster shared biographies among actors and thus favor group formation.

Generalizing these institutional conditions, one can conclude that there are functional analogies in the institutions that produce top policy actors and that identifying them is an important step in finding potential programmatic actors. However, research also confirms that not only the professional background but also membership in relevant legislative bodies are predisposing factors for the generation of innovative ideas (Makse, 2020). The hypothesis includes the understanding that the institution relevant for programmatic action is the presence of elite formation structures that establish commonalities among actors, be it the same education, the same partisan ideology, or the same work environment (ministry, sickness fund, or other state-related institution). Structures of elite formation favor the formation of groups because they ensure a common background among policy actors.

3.4 Policy Advice and Scientific Impulses

If it can be assumed that social groups are formed on the basis of shared characteristics and that these groups influence the policy process over time, the question arises as to what drives their innovative ideas and the promotions during the policy process. In terms of PAF, ideas are compiled into what is called a policy program, and this program is said to reflect in part the policy preferences of the programmatic actors, but it cannot be seen as a mere translation of preexisting preferences or even beliefs into policy. But where do these ideas come from? In general, they can be either externally borrowed, that is, transferred from other political systems and following other national and international experiences (exogenous generation of policy ideas), or generated from within a policy subsystem, for example, by expert commissions, evaluations, or scientific results produced by think tanks and national universities (endogenous generation of scientific knowledge).

A focus on the exogenous generation of policy ideas profits from the review of literature on policy transfer and internationalization in explaining the adoption of certain reform strategies from other governments. This literature clearly distinguishes between policy innovation and policy invention, with the latter describing the original development of new policy instruments and the former referring to the adoption of policy instruments from other jurisdictions or systems (Berry & Berry, 2017, p. 253f). There are four most prominent mechanisms of policy diffusion (Shipan & Volden, 2008) organized around the three main dynamics of voluntary adoption of policies from other countries, regulatory race, and external pressure and coercion to adopt policies. The first dimension here subsumes two mechanisms that were originally separate in the literature, imitation, and learning, as these are difficult to disentangle empirically.

The first part of the policy transfer and diffusion models sheds light on the processes by which governments learn from the experiences of other countries by idealizing them as role models, imitating them, or deliberately following different paths. When transferring policies—where it is often not clear whether they imitate other countries to create an ingroup feeling or whether they have actually learned from the experiences of foreign governments—an important assessment relates to the integration of policies developed for and in different institutional contexts into the respective institutional setting (referred to as institutional fit) (Obinger et al., 2013, p. 122f). Whether policies fit institutional settings depends on policy-related and institutional characteristics. Thus, from which countries governments adopt policies depends on a number of factors (Maggetti & Gilardi, 2016, p. 93). Similar institutional settings can favor policy adoption (Strang & Meyer, 1993), as can the preexistence of certain cultural values (J. L. Jensen, 2003). Similarly, geographically and ideologically close (“neighboring”) systems are more likely to learn from each other (Grossback et al., 2004; Mertens et al., 2019, p. 14; Mitchell, 2018). The ideological argument also applies to the imitation of political parties by their transnational counterparts (Wolkenstein et al., 2019) and a party-based propensity to adopt ideologically different policies (Butler et al., 2017). A prominent example is the third way politics in labor parties worldwide, most notably promoted by Tony Blair (Johnson & Tonkiss, 2002).

Secondly, the competitive nature of policy transfer and diffusion is inherently trying to perform not just as good as other countries (as suggested by learning or imitation argument), but better than others, especially in relation to the global market (Marsh & Sharman, 2009, pp. 271-272). In doing so, governments may seek to make themselves more attractive in terms of tax policy (Burge & Rogers, 2016) and align their policies accordingly in the competition for businesses’ locations (Baybeck et al., 2011; Leiser, 2015). Whether these processes lead to higher or lower standards in comparative regulation is summarized under terms such as races to the top and races to the bottom. Following this debate, scholars have noted a race to the top in, for example, environmental standards (Bastiaens & Postnikov, 2017; Busch et al., 2005, p. 164; Saikawa, 2013) and a race to the bottom in higher education spending (Bailey et al., 2004, p. 72), as well as mixed results in labor rights (Mosley & Uno, 2007).

Thirdly, increasing processes of internationalization have led states to adopt policies because they would otherwise fear sanctions. Such coercive policy diffusion is only possible if international organizations or supranational institutions (such as the EU) have the necessary threat potential to force states to adopt policies. Alternatively, states can credibly exchange support or cooperation for policy adoption and attach policy adoption as a condition of action. Processes of policy adoption through coercion or pressure can be distinguished from harmonized intentions to converge policies and from more uncoordinated mechanisms of policy diffusion (Busch & Jörgens, 2005, p. 867). Among relatively equivalent states in terms of economic performance, demographics, and import/export relations, coercion is more likely to be exerted in the course of treaty formation (Morin & Gold, 2014, p. 784). Alternatively, there is political pressure under which public sector organizations reform to maintain the flow of support and resources (Andersen & Jakobsen, 2018, p. 217). A prominent example is the case of Greece, which was forced to implement austerity measures in exchange for financial aid during the financial and economic crisis (Obinger et al., 2013, p. 115).

In health policy, the explanatory power of policy transfer and diffusion models is tendentially rather limited (Starke & Tosun, 2019, p. 196). Following the argument of Makse and Volden (2011) that the characteristics of policies, particularly their combinability with prior practices and their complexity, influence whether they are diffused, it is not surprising that few policy diffusion studies focus on health policies. Finally, jurisdictions must also consider how policies will be implemented (Nicholson-Crotty & Carley, 2015). Given the often historically evolved and institutionally divergent structures of health care, diffusion in health policy is likely to occur only when institutions are newly established and given additional responsibilities, as in the case of evidence-based health agencies (Hassenteufel et al., 2017) or health technology assessment (HTA) (Löblová, 2016). In less federalized systems with shared responsibilities among health care organizations, subnational responsibilities can serve as opportunities for experimental evaluation of policy effectiveness (Volden, 2017). This research reemphasizes the need to consider institutional differences and predispositions in the study of health policy.

Moreover, the processes of harmonizing policies across Europe are less developed due to institutional challenges, and countries rarely want to transfer competences in this policy area to the supranational level (Martinsen et al., 2020). Consequently, internationalization, Europeanization, and diffusion models are expected to play a minor role in explaining health policy in France and Germany in recent decades.

However, even if limited, there is a certain international influence on national health policy programs. Joint membership in national or supranational bodies has been shown to drive the diffusion of policy knowledge so that, for example, changes in the hospital financing systems of a majority of OECD countries resulted in part from their positive effects on reducing health expenditures in other OECD countries (Gilardi et al., 2008, p. 568). Hassenteufel and Palier (2014) stress that the new economic and budgetary rules set by the European Union has increased the presence of deficit reduction instruments and arguments in the French political discourse. Although not yet driving French health policy reforms, the economic rules clearly put pressure on the national governments that hinders the adoption of rather leftist reforms involving an increased spending on health care and hazarding the consequences of augmented debt and deficit.

While the French health policy can be seen in part as a reaction to these international pressures, the neoliberal ideas put forward by Reagan and Thatcher translated into elements of competition prominently being introduced in Germany. These were borrowed from the experience of the Dutch reform in terms of regulated competition and the reference pricing system (Leiber et al., 2010), the US in terms of selective contracts (Cheng & Reinhardt, 2008), but not in France (Propper, 2018, p. 494), and Australia with regard to DRGs (Schreyögg et al., 2006). Although health care reforms in Germany strongly resembled the “Third Way” strategies as in the UK, one cannot speak of a true convergence of strategies and instruments, but rather of a similar way of legitimizing reforms (Bandelow, 2007b). Ultimately, the adoption of policy reforms from other countries depends heavily on existing institutions in the respective political systems and subsystems, especially in the health care sector, whose established structures make one-to-one policy transfer difficult (Verspohl, 2012, p. 308). Although the health care system in both France and Germany is based on a Bismarckian structure of contribution-financed SHI, and Bismarckian welfare states tend to converge (Hassenteufel & Palier, 2007), institutional differences regarding federalist, corporatist, and political structures make it difficult to copy many (at least structural) reforms.

While policy transfer and diffusion models obviously influenced the content of health care reforms in France and Germany, the question remains how the utilization of policy advice and scientific impulses are linked to biographies and thereby to actors’ biographies. The endogenous role of scientific knowledge and advice in the policy process can help shed light on this. It is divided into two main strands of literature:

The first is concerned with evidence-based policymaking, more specifically how research is or should be formally embedded in the policy process and how the relationship between researchers and policy-makers is or should be (French, 2019). The use of experts and evidence in deliberative processes of citizen participation also fall under this conceptualization (Roberts et al., 2020), but are not considered in this contribution due to their marginal impact on policy processes. However, the involvement of expert advice in decision-making structures varies and depends on institutional structures such as cabinet decision-making and government composition (Fleischer, 2009). It can be achieved by taking into account suggestions and expertise from sectoral actors, such as interest groups. In the area of health policy, for example, medical and epidemiological advice could be relevant information for decisions regarding which services are reimbursed by sickness funds. However, evidence from other studies suggests that the consideration of interest groups on highly politicized issues depends less on their expertise than on their social support (Willems, 2020). More to the point, interest groups may even be disempowered if expert advice is used more intensively to inform and legitimize policy-making (Cross et al., 2021). Expert advice can also come from political parties or, in a supposedly neutral way, from the involvement of scientists in decision-making bodies or councils.

Under the label of evidence-based policy-making, some scholars now argue for systematic consideration of science by policy-makers. However, it is increasingly difficult to make evidence available to decision-makers in a productive way. Part of this challenge lies in science communication, as both the recipients of expert advice in policy are unclearly perceived and it is unclear exactly what is required of scientists to support decisions (Saretzki, 2019). One solution to this challenge may be ongoing exchanges between policy actors and experts, ideally in committees through which science is expected to have its greatest impact on policy (Bogenschneider & Bogenschneider, 2020; Peterson, 2018). Another challenge lies in the selection of expertise for policy decisions, also with regard to potential conflicts of interests, although McComas et al. (2005) note that experts in advisory bodies value expertise over impartiality. In German health policy, this challenge has been addressed by the largely impartial proposals developed in the SVR-G and the Enquete Commission (Knieps, 2009). However, greater collaboration between politics and science may also prove fruitful for coalition-building (Cairney et al., 2016, p. 401). Ensuring that scientific advice is considered over time requires long-term strategies and collaboration, a new form of co-production (Cairney & Oliver, 2017, p. 4).

This leads us to the second understanding of scientific knowledge and advice as a more group-centered perspective on advisory committees. In this regard, the literature on policy advice has revealed a large influence of advisors on policy-making processes. At the science-policy interface, Hoppe (2010) distinguishes several types of boundary workers that fill in diverse functions. The concept of policy advice is understood here less as scientific impulses for innovative policy ideas than as political advice. The closest to the notion of programmatic action is mega-policy strategy, as programmatic actors are focused on major changes and strategic outlooks (Hoppe, 2010). In fact, programmatic actors represent a specific type of science-policy relationship—they can be civil servants, policy advisors, or administrative actors of sectoral bodies who liaise with scientists and are informed by the evidence provided to develop new ideas and compile them into a strategic vision for the policy sector.

The institutions of expert advice involvement also vary depending on the political institutions surrounding them. In countries where political parties play a significant role in formulating policies, recruiting political elites, and preparing legislative proposals through parliamentary committees, party-affiliated foundations can have a considerable impact on the generation of ideas. Public administration research also suggests that the civil service tends to become politicized when it must provide expert advice with equal attention to political considerations (Craft & Howlett, 2017). However, Shaw and Eichbaum (2020) show that politicization of civil servants is not driven solely by exogenous factors, but is inherent in civil servants’ sense of duty to their minister. They distinguish between different forms of politicization of the ministerial bureaucracy, focusing on the relationship between civil servants and political advisors and the extent to which political considerations moderate the provision of advice (Hustedt & Salomonsen, 2014, pp. 749-751). In partisan democracies, this is all the more relevant, because civil servants are selected in part on the basis of partisan ties and are thus prone to giving political rather than evidence-based advice.

On the other hand, expert advice may not only initiate new thinking, but also serve to confirm existing opinions or be instrumentalized in the struggle for “truth” (Kisby, 2010). Heikkila et al. (2020) show that scientific information almost always leads policy actors to corroborate their preferences and that this can only be broken if actor networks are diversified, actors are familiarized with science, or scientific knowledge is coupled with existing risk perceptions. Thus, for science to generate new policy ideas, actors must be brought together who ideally do not have preexisting policy preferences. This is even easier in policy sectors that have not yet been touched, although examples of French and German health policy show that this is possible even in highly contested policy sectors blocked by vested interests, as long as scientific uncertainty does not prevail and prevent far-reaching legislation (Bromley-Trujillo & Karch, 2019), and as long as the actors initiating change do not belong to these competing coalitions. As a consequence, the following hypothesis is postulated that sees the institutionalization of advice and ideas as a potential necessary condition for programmatic action.

Science Policy Hypothesis

Enabling the inclusion and institutionalization of scientific advice is a necessary condition for the generation of innovative ideas and inspires policy programs in the policy process. Such venues of intellectual reflection and discussion are necessary also to inform and support the formation of programmatic groups and successful programmatic action.

Taking the institutionalization of expert advice as a necessary institutional condition for programmatic action, it is interesting to see how this is shaped in different institutional settings. There are different forms and dimension of advice. A prominent view in public administration is on ministerial advisors, who are assigned to a single minister and are functionally equivalent across countries, acting between politics and bureaucracy (Hustedt et al., 2017). An example are the ministerial cabinets in France (De Lamothe, 1965) or Italy, where career patterns lead to links between sectoral experts in the ministerial bureaucracy and other sectoral bodies (Di Mascio & Natalini, 2013, p. 338). The empirical study has shown that these ministerial cabinets are also highly relevant for programmatic action. Their members have been trained (which is a link to the institution of elite formation) to provide expertise, and the networks they form influence policy-making and perpetuate themselves (Rouban, 2007, p. 492).

Although particular to the French political system, the functional equivalent found in other countries is the extent to which new ideas are formulated and driven by a group of actors close to the state apparatus (such as the minister). Thus, it is not ministerial cabinets per se that enable programmatic action, but the institutionalization of expert advice at a particular point in the policy process within groups that subsequently advance their jointly developed ideas. This mainly works through the formation of shared biographical trajectories through these expert bodies. In line with the literature on policy transfer and diffusion, it must be noted that these biographies can also be driven internationally.

3.5 Intermediary Conclusion: Institutional Conditions for Programmatic Action

The review of the literature strands outlined above served two purposes: First, to highlight the gap that the PAF fills as an additional lens on policy processes by identifying the distinctive features of existing theoretical frameworks and situating the PAF within these lines of research. Since these perspectives may also offer alternative explanations for the policies explained in the empirical analysis using the PAF, it is important to clarify in which situations these theoretical perspectives prove particularly useful—and in which they do not—which will be an important argument for the usefulness of the PAF in unfolding explanatory power. The extent to which they can serve as alternative explanations for the observed policy changes will be addressed later. The second goal was to use existing research on political institutions and public policy and in particular the ways in which institutions are incorporated in theoretical mechanisms explaining policy processes, to theorize the opportunities and constraints of programmatic action. If one pursues the goal of the PAF to be applicable and insightful across different political systems, it is imperative to work out the extent to which country-specific and comparative studies of the institutional settings within which policy-making takes place can help sharpen the PAF concepts and identify the relevant institutions for programmatic action. In this regard, it is particularly fruitful to incorporate the insights of established institutional analyses in explaining group formation and the emergence of ideas as well as their subsequent success. Indeed, the theoretical perspectives presented offer points of departure for PAF to develop theoretical arguments, particularly regarding the ways in which institutions may influence programmatic group formation and the success of programmatic groups and policy programs. While such an overview provides a starting point, only later empirical investigation will reveal which institutions have proven relevant to programmatic action.

To bring clarity and order back into the previously extensively discussed theoretical jungle, this chapter briefly summarizes the core explanation for policy change advocated by theoretical perspectives and justifies why and which elements of the PAF present add value to the theoretical state of the art. While some institutionalist arguments of existing theories also appear transferable to the PAF, such as corporatist and decentralized structures, the later analysis will also add undiscovered institutional drivers of programmatic action, which include institutional change in addition to institutions promoting elite formation and innovation of ideas.

Table 3.1 provides an overview of policy process frameworks, their foundations, core explanation, and understanding and relevance of institutions. The postulated hypotheses assume that these institutions observed in other theoretical perspectives are potentially also relevant to the PAF or, more specifically, to programmatic group formation and the success of programmatic groups and policy programs. However, the postulated hypotheses point toward a great importance of education and systems of elite formation, as well as of the institutionalization of policy advice, to programmatic action.

Table 3.1 Overview on scope of theories of policy change and stability

The core explanations for policy change vary depending on the theoretical perspectives: Policy change in response to problem pressures, external events, or situational factors such as opening policy windows represent the perspectives adopted by the MSF, which further specify the institutional conditions under which policy entrepreneurs are likely to succeed in their attempt to push through their policy proposal. Long-term processes of policy change, on the other hand, tend to be represented by network theories and the ACF, which propose to conceptualize policy change as a process that takes place over a decade or more. The PAF brings to this set of theoretical perspectives a view of shared biographies and resulting collaborations based on a policy program to include policy change over at least a decade to a generation of programmatic actors (30 years or more). The key innovation, then, is that it assumes career policy actors seeking greater authority in the state and that it assumes that these actors form programmatic groups based on shared experiences and pursue a policy program of their ideas to achieve their goals.

Under what institutional conditions both the formation of programmatic groups and the success of programmatic groups and their policy programs is achieved is the core question of this book. After looking at existing research on the interplay of actors and institutions in policy processes, it is possible to assess which institutions are potentially relevant for the course of programmatic action. This assessment is based on the previously elaborated Programmatic Action Framework. Table 3.1 therefore contains a column outlining the institutions taken from existing theoretical approaches and the ways in which they might influence the formation and success of programmatic groups. In the following paragraphs, this relationship is exemplified for each theoretical strand.

Two frequently addressed institutions in comparative politics that have an influence on policy-making and actors are corporatism and federalism. Corporatist structures can generally be seen as a facilitating factor for group formation. Their institutionalized venues for policy-making offer numerous opportunities for exchange and cooperation. Programmatic groups have a chance of success with their programs if they succeed in making strategic use of these corporatist structures and integrating the corporatist actors into their programmatic group. The same is true for varying levels of federalism. Subnational structures generally support the formation of groups and exchange among policy actors and therefore are necessary to include when programmatic groups seek success because they are relevant in policy-making processes.

External events, socioeconomic developments, and public crises are often integrated into policy process frameworks (as external shocks in the ACF or focusing events in the MSF) and thus always represent catalysts of processes, be it group formation or policy success (Saurugger & Terpan, 2016). Therefore, with respect to the PAF, it is equally reasonable to assume that these situational conditions act as accelerators of group formation and also represent opportunities for short-term ad hoc cooperation. Moreover, they can be useful and contribute to programmatic success if programmatic groups use these problem pressures and crises to advance their own policy program, and if the policy program is able and flexible enough to react to these developments and provide an adequate response to the emerging problems.

In actor-centered approaches to policy processes, namely MSF, ACF, and network theories, the relevant institutions are always those that provide resources to actors or opportunities for actors to behave in ways that favor policy change. These include some of the institutions of veto players, federalism, corporatism, parties, and path dependencies mentioned above. Network theories in particular emphasize the role that institutional opportunities play in cooperation and collective action. In sum, institutional opportunities that promote networking among policy actors and those that provide policy actors with the resources and strategies necessary to pursue their goals contribute to programmatic group formation. The subsequent success of these strategies then depends, in turn, on whether programmatic groups have the necessary resources to achieve their goals in the institutional settings and whether they use a coherent narrative that fits their strategy.

When talking about institutions, policy process frameworks basically mean the same type of political institutions introduced by prominent typologies such as those of Lijphart (2012), Tsebelis, or Hall and Soskice (2001). These are used in analyses that apply the PET (Fernández-i-Marín et al., 2019, p. 14; Kuhlmann & van der Heijden, 2018, p. 329), theories of coalition formation (Fischer, 2015, p. 246), and the MSF (Zohlnhöfer et al., 2016).

While Table 3.1 already contains an assessment of the potential relevance of political institutions for the PAF, formulating concrete expectations about the (directional) impact of well-known and widely recognized institutional settings on programmatic action seems less straightforward. In particular, this is because institutional conditions exert a different influence on programmatic action depending on whether the emerging programmatic group under analysis already occupies key positions in the policy sector (yes/no) and whether it faces an established programmatic group that already holds key positions (yes/no). It follows that institutional differences per country do not necessarily explain programmatic group formation and programmatic success, or at least have different effects depending on the characteristics and situation of the programmatic group. To illustrate the impact of each institutional factor on the different stages a programmatic group may be in Fig. 3.1 shows a coordinate system listing institutional influences on opportunities and constraints for programmatic group formation and programmatic group and policy program success, depending on whether it is an emerging programmatic group with or without key positions filled. Furthermore, the effects of institutions are visualized in cases where the emerging programmatic group faces an established programmatic group with or without key positions occupied. The chances for group formation and programmatic success are always evaluated for an emerging programmatic group.

Fig. 3.1

Institutional opportunities and constraints for programmatic action. Source: Own compilation

Depending on the characteristics and situation of an emerging programmatic group, how would institutions affect the formation of the programmatic group and its subsequent success? Four scenarios are conceivable here:

3.5.1 Emerging Programmatic Group, Not All Key Positions Occupied

Imagine many policy actors who are potential programmatic actors. This group of actors does not (yet) occupy all the necessary positions that would allow them to participate directly in all the arenas where policy ideas are debated, and decisions are made. In such cases, the degree to which group-forming institutions are institutionalized with the participation of relevant actors will contribute significantly to whether a programmatic group will form and succeed, because the group needs this kind of connection to gain the positions necessary to possess resources for success. Decentralized structures in weak federalism and corporatism provide a variety of opportunities and venues to connect and form groups. Given the decentralized or corporatist structures that the programmatic group then penetrates and given that the group then includes members across these structures, its resources increase and its success solidifies. Other networking opportunities and institutionalized biographical paths are even more conducive to interaction among policy actors, especially when no established programmatic group dominates such venues. However, in this case, the subsequent chances of success are not as good as those promised by corporatist or decentralized structures in the short term. Since networking events offer many points of contact, a programmatic group can also be attacked more easily. For an emerging programmatic group that does not occupy key positions, the necessary resources to adopt and implement its program are lacking. Accordingly, neither the formation nor the success of programmatic groups is expected.

3.5.2 Emerging Programmatic Group, with Key Positions Occupied

However, when considering an emerging programmatic group that already occupies central key positions, several institutions favor its formation and subsequent success. Federalism and corporatism, as well as institutionalized education systems and policy advice, favor the emergence of ideas and groups because they provide a pool of policy actors among whom are already actors in the necessary positions that can influence policy-making at many levels. Consequently, this favors both the formation and the subsequent success of programmatic groups because they benefit from multiple opportunities to collaborate, include all relevant actors in their group and program, and engage in intellectual discussion that they control through key positions.

3.5.3 Established Programmatic Group, Not All Key Positions Occupied

When the emerging programmatic group faces an already established programmatic group, the question of group formation and success depends on which key positions are still held by the established programmatic group. If it has already lost key positions, this provides some opportunities for the emerging programmatic group. Again, federalism and corporatism, as well as institutions that foster group formation through institutionalized career paths and advisory systems, favor group formation and provide an anchor for the emerging programmatic group to challenge the established programmatic group. If they can engage policy actors in the key positions not held by the existing dominant programmatic group in an inclusive policy program, this favors their success.

3.5.4 Established Programmatic Group, with Key Positions Occupied

Finally, the most uncomfortable situation for an emerging programmatic group is when there is an established programmatic group that occupies all the key positions. As Bandelow and Hornung (2020) show, this is the moment when the incumbent programmatic group will be strongest. Since all key positions are occupied by the dominant programmatic group, all institutions are under its control and thus the stabilizing institutions play into its hands. Federalist and corporatist structures are shaped by the policy program. Only the opportunities for networking and collaborative forums through elite formation and expert advice remain, offering the potential for programmatic groups to form, but they will not be successful because key positions belong to the established programmatic group. In such cases, policy actors might simply join the dominant programmatic group to pursue their goals, comparable to what is called engaging in mainstream politics in party politics (Schumacher & van Kersbergen, 2014).

What is also apparent from this coordinate system is that the institutional factors can be found in the upper right or lower left quadrant in each case. This means that the questions about the formation of programmatic groups and their success are theoretically (and possibly also empirically) very closely linked. Accordingly, the expected influence of institutions equally regards the formation and success of groups.

Summarizing the results of these theoretical considerations, it is clear that the effects of the institutions examined here on programmatic action are complex and also depend on the stage a programmatic group is at, at any given time. While some political institutions can shed light on one or another aspect of favorable circumstances for programmatic group formation and programmatic success, the institutions of programmatic action appear to have a different focus than the previously known and used in comparative politics and policy process research, namely a focus on the formation of groups and the involvement of innovative ideas for programs to be successful.