Introduction

This chapter outlines our theoretical framework, which is based on reconciling insights from three interconnected analytical and conceptual approaches. The first section presents the key dimensions of agenda setting, one of the most established analytical constructs in social sciences and political science in particular. The second section elaborates at greater length on advocacy coalition theory, which remains one of the most celebrated approaches in policy studies. We relate these analytical approaches to each other, reconcile them to see their combined power, and discuss their applicability in studies of European integration. Not least their lessons for modern policymakers explain why both approaches still resonate today.

We see these analytical approaches as closely intertwined and particularly well-suited to examining the strategies of the Europarties in relation to Treaty reform and the broader question of the future of Europe. Europarties are engaged in continual advocacy and agenda-setting about key European matters. In addition, the third part of the theoretical framework discusses the transnational partisan dimension of European integration, which we conceive of as a central mechanism through which the European Union (EU) evolves. These are the three analytical approaches that underpin the empirical analyses in Chapters 4 and 5, while the concluding chapter weaves together these strands of research to reflect on how Europarties shape EU politics.

Combining these approaches helps to achieve a deeper understanding of Europarties, their actions, and how they have shaped European integration—in the context of Treaty reforms, they are also helpful in uncovering the underlying reasons for their success (or lack of it). It means capturing the essence of what drives Europarties and why they matter, as well as highlighting the challenges they face both in mobilizing their networks and in maintaining momentum in the ever-changing process of integration. However, the approach in this chapter is deliberately conceptual while subsequent chapters develop the argument about networks and coalitions.

Agenda-Setting in EU Politics

Agenda-setting is a fundamentally important stage of politics. Starting with Cobb and Elder (1971), academic research has produced different typologies and approaches to studying agenda-setting. That literature often identifies three types of agendas: the public agenda includes issues that citizens find salient; the media agenda consists of issues that are covered by the media; and the political agenda includes issues that policymakers deal with. According to the so-called multiple streams framework (MSF) model (Ackrill et al., 2013; Béland & Howlett, 2016; Kingdon, 1984), policymaking processes consist of three streams: the problem stream consists of problem perceptions among policymakers; the solution stream consists of proposals for political decisions; and the politics stream consists of political activities and developments like lobby campaigns, or the political context in which decision-making occurs. The links between the three streams are made by issue entrepreneurs, individuals, or organizations that ‘are willing to invest their time and energy in promoting a particular issue’ (Elder & Cobb, 1984: 121). When these three streams meet, a ‘policy window’ opens and the issue moves to the agenda of decision-makers. Within MSF, ‘the analytical task is to specify the dynamic and complex interactions that generate specific policy outcomes’ (Ackrill et al., 2013: 872–873), but particularly in complex settings such as the EU, this can be inherently difficult.

As for the origins of issues on the agenda, they can come from the external environment or from the political actors themselves (Mansbach & Vasquez, 1981). The former approach sees political issues arising from the international environment. The latter category in turn includes issues that arise from the interests of the actual stakeholders, the political institutions, and actors within them. As argued by Princen (2007, 2009), in EU governance, the latter approach is normally more appropriate for understanding the sources of items on the agenda of the EU institutions, although major external developments such as terrorist attacks, military conflicts, refugee crisis, or climate change obviously feature high on the EU agenda. National governments or interest groups try to move issues to the Brussels agenda, and the European-level actors—Europarties and their EP political groups included—have their own strong reasons for having matters debated in EU institutions.

Agenda-setting success is often influenced by how problems are framed (Daviter, 2007, 2011). Issue entrepreneurs can refer to broadly shared fundamental values (for example, human rights, sustainable development, or democracy), or use an alternative strategy of ‘small steps’ whereby support is gradually built up through more low-key strategies, including behind-the-scenes processes and depoliticization of issues (Princen, 2011). A related tactic is issue bundling or what in MSF terminology is called coupling: ‘Apart from skills and resources, entrepreneurs pursue strategies to join together problems and policies into attractive packages, which are then “sold” to receptive policy-makers’ (Ackrill et al., 2013: 873). Considering the ‘distance’ between Brussels and average citizens, ‘agenda-setting strategies in the EU will be focused more exclusively on dynamics that take place within policy communities than on reaching out to larger audiences outside of those communities’ (Princen, 2011: 940). And, as Princen also points out, broadening the scope of participation entails the risk of creating controversy and opposition. For example, proposals such as transnational lists for EP elections are sure not to please the more Eurosceptical politicians.

Another key dimension concerns the ‘venue’ (Baumgartner & Jones, 1993), that is, where and by whom the issue is debated. Princen (2011) distinguishes between venue shopping and venue modification. ‘Venue shopping occurs when agenda-setters seek out a venue (among those available to them) that is most receptive to their cause. Within the EU, venue shopping may occur between EU institutions (horizontal venue shopping) and between the different “levels” in the multilevel system that the EU forms part of (vertical venue shopping)’ (Princen, 2011: 931). Venue shopping occurs among already existing venues, whereas venue modification means that ‘if a suitable venue is not available, actors may sometimes also be able to modify the range of available venues in order to create one that is better suited to their purposes’ (Princen, 2011: 933). For example, in EU governance environmental activists may prefer that environmental policies are on the agenda of actors that are likely to have more pro-environment positions. For Europarties, a particularly relevant question is the balance between supranational (Parliament, Commission) and more intergovernmental (Council, European Council) institutions.

Modern Europe is a multilevel polity that offers political actors various access points for influencing decision-making. The Commission enjoys the monopoly of legislative initiative, and more broadly as the ‘engine of integration’ it is commonly perceived as having a central role in setting the agenda in Brussels (e.g., Hartlapp et al., 2014; Koop et al., 2022; Pollack, 1997). Sometimes external shocks or unexpected events can have a strong impact on agendas, as has happened recently with the euro crisis, the refugee crisis, Brexit, COVID-19, and Putin’s war in Ukraine. Pollack (1997) distinguished between formal and informal agenda-setters in EU governance. The former includes the ‘big’ institutions like the Commission and the European Parliament (EP), whereas the latter are issue entrepreneurs. Europarties in a sense belong to both categories: they are independent civil society organizations, but strongly present in EU institutions. Key individuals inside Europarties and EP political groups are thus both policymakers and issue entrepreneurs. Overall, there is a broad range of actors from national and EU institutions to lobbyists and interest groups to public opinion that influence which issues receive the attention of EU decision-makers (e.g., Ackrill et al., 2013; Daviter, 2007, 2011; Princen, 2007, 2009, 2011; Tallberg, 2003). As national governments and EU institutions consist of party politicians, it is obvious that, essentially, all major integration milestones as well as normal EU laws and policies have been shaped by political ideologies and the programmes and positions of political parties.

The agenda-setting approach is thus helpful in understanding the emergence and framing of issues on the agenda of decision-makers. However, it usually emphasizes positive power and neglects negative power, that is the power to prevent other actors from devoting attention to specific issues. While we in this book focus on goals and priorities of the Europarties, it is equally relevant to acknowledge issues and solutions not promoted by these supranational partisan actors. But regardless of whether Europarties promote their favoured solutions or try to keep certain proposals off the agenda, they engage in advocacy, either alone or as coalitions of like-minded actors.

Advocacy Coalition Framework

Following the discussion of agenda setting, this section presents the key aspects of the Advocacy Coalition Framework (ACF).Footnote 1 It introduces its concepts and assumptions, trying to nail down a definition, and it also gleans some of the framework’s strengths and limitations. It discusses applications and describes how the framework can be further developed based on lessons from a broad set of empirical applications more generally and specifically in the EU context. Our ultimate consideration is the applicability of this framework to Europarties—outside or inside the European Parliament within the political groups and even across them—and how we can usefully build on the ACF to investigate partisan transnational advocacy networks or coalitions in the EU.

Key Concepts and Assumptions

In everyday use, advocacy, of course, means any action in support for or recommendation of a particular cause. How, then, do scholars define advocacy coalitions? In broad strokes, advocacy coalitions are groups of like-minded actors who share beliefs and engage in a ‘non-trivial’ degree of coordination to decisively impact policy outputs and change. Sabatier (1988: 139; see also Sabatier, 1987; Sabatier & Pelkey, 1987), who pioneered the ACF, suggests the following definition of advocacy coalitions: ‘These are people from a variety of positions (elected and agency officials, interest group leaders, researchers) who share a particular belief system—i.e., a set of basic values, causal assumptions, and problem perceptions—and who show a non-trivial degree of coordinated activity over time’. To Sabatier (1988: 133) advocacy coalitions are composed of ‘people from various organizations who share a set of normative and causal beliefs and who often act in concert’. The underlying assumption is that actors can be aggregated into several advocacy coalitions and that actors can be drawn from various governmental and private organizations.Footnote 2

Coalition dynamics, in other words, are crucial as the ACF assumes that the final policy output reflects the winning coalition’s beliefs (e.g., Pierce et al., 2020). Three meta-theoretical concepts underlie the ACF: coalitions, learning, and policy change (Henry et al., 2022; see also Nohrstedt et al., 2020; Weible, 2017). Furthermore, there are three cross-cutting or basic concepts within the ACF that serve as the backbone for analysing those three key components (coalitions, learning, and policy change): subsystem, actors, and beliefs (Henry et al., 2022).

The ACF is one of several theoretical approaches available for studying policymaking. It sought to provide an alternative to the understanding of the policy process as a policy cycle. In the words of Weible (2017):

The ACF aimed at gaining a better understanding of some of the most perplexing puzzles in public policy, including the formation and maintenance of coalitions, the propensity for learning and the role of science and technology in policy processes, and the factors associated with policy change over time.

As an approach to public policy as a field of study, and with hundreds of applications across the world, the ACF is one of the most established and most frequently applied approaches for studying policy processes.

The ACF can be understood as a policy process framework that has been developed to simplify the complexity of public policy (Weible et al., 2009). It is designed to deal with complex subsystems involving large numbers of actors and to understand policy changes over a period of a decade or more within a particular substantive domain/subsystem. As distinct from other theories of the policy process, the ACF provides a more cooperation-oriented approach and one that centres on policy change. It does so by emphasizing notions of learning and effects therefrom for policy outcomes. The ACF is best applied at the subsystem level and less within specific action situations (Weible et al., 2011: 357). Indeed, policy subsystem is the primary unit of analysis and the principal empirical and theoretical domain. Advocacy coalitions operate within policy subsystems and the ACF highlights the role of both formal and informal actors in such subsystems and their strategies and resources for furthering their policy objectives by shaping policy outputs and impacts (e.g., Cisneros, 2021; Jenkins-Smith et al., 2018; Sabatier, 1988; Weible, 2017; Weible et al., 2009, 2011, 2020).

ACF is based on seven foundational assumptionsFootnote 3:

  1. 1.

    Policy subsystems, characterized by a geographical area, an issue, and policy actors.

  2. 2.

    The subsystem actors include participants who attempt to influence policy. These may include government officials, private or non-private organizations, experts, scholars, consulting firms, think tanks, and media, among others.

  3. 3.

    The ACF assumes that policy actors are boundedly rational.Footnote 4

  4. 4.

    Subsystems aggregate actors into one or more coalitions. The ACF provides a lens to see the policy actors as members of coalitions. These coalitions are formed based on similarities and differences in core policy beliefs.

  5. 5.

    Policies often reflect and translate the beliefs of one or more coalitions.

  6. 6.

    Scientific and technical information is important for understanding subsystem affairs. Such information, besides the day-to-day experience of the policy actors, informs the causal patterns adopted by the belief systems.

  7. 7.

    To understand the policy process or policy change, the researchers must adopt a long-term perspective (e.g., ten years or more). Often, the debates among the coalitions last more than decades, and to understand the coalitions, learning, and policy change, one must understand all the past events.

The ACF attempts to understand and explain mainly three puzzles. These are, again, advocacy coalitions, learning, and policy change. They are considered puzzles because existing empirical research has produced mixed results.

Regarding the first puzzle, advocacy coalition, Weible (2017) points out that even when there is evidence showing the existence of coalitions and shared beliefs within those coalitions, there is no clarity if it is necessary to have shared core beliefs when forming coalitions or whether secondary beliefs are sufficient. He further notes that these diverse findings reflect different approaches when studying advocacy coalitions. And, additionally, some scholars bear in mind that other factors, such as shared interests, trust, and resources are important too in coalition formation, not only shared beliefs. The policy actors who are part of the advocacy coalition are those who are essential to the ‘coalition members’, and those who play a certain role within the coalition. They include brokers, who work to reach agreements among opponents; and entrepreneurs, who play a role in leading coalitions, facilitating learning, and producing policy change.Footnote 5

The second puzzle, learning, refers to the way in which individuals decide ‘to change their actions and way of thinking after having certain experiences and which are concerned with the accomplishment or revision of the guidelines of the belief system of each individual’ (Weible, 2017; see also Weible & Jenkins-Smith, 2016; Weible et al., 2011). Research has shown that learning does indeed occur within and between different coalitions, but it is not clear whether this learning process includes changing the core and secondary policy beliefs within the coalition, or whether the change in secondary beliefs can begin to generate this learning.Footnote 6 In addition, many researchers have emphasized that other factors, such as networks and science, have been shown to facilitate policy learning within and between the various coalitions.

Thirdly, policy change, refers to the changes that occur in policy, what generates these changes, and, subsequently, what these changes are (Weible, 2017; see also Weible et al., 2011). There are some changes in the core beliefs of the coalitions, named ‘major’ policy changes, and there are ‘minor’ changes, which occur in the secondary aspects of the policy subsystem. Research has shown that there are indeed changes in policy and that there are certain factors that lead to a change in policy. However, the process of understanding is complex because policy change is not the result of one event alone, but rather a combination of diverse dynamics that occur in one process over time.

Taken together, these three puzzles or themes—advocacy coalitions, (policy-oriented) learning, and policy change—capture the core areas of theoretical emphasis in ACF and structure empirical explorations within the framework. The ACF assumes that policy actors are primarily motivated by their belief system, which is partitioned into fundamental normative orientations called deep core beliefs, normative and empirical policy-related beliefs called policy core beliefs, and narrow and instrumental secondary beliefs (Weible et al., 2020). Among the assumptions, ACF thus explicitly identifies beliefs as the causal driver for political behaviour. But how to establish the cause–effect relationship? This question can be difficult to answer not least because of the complexity of advocacy coalitions and the potential impact of other factors. It may be that the ACF understates the complexity of the problem and overestimates the explanatory power of advocacy coalitions.

Empirical Applications and Questions About Generalizability

Since its emergence in the 1980s, the ACF has developed into a rich and varied research programme with a growing community of scholars applying the framework, testing and developing its hypotheses, and exploring new methods of data collection and analysis in political contexts that span the globe (Weible, 2017; Weible et al., 2020). At its core, the ACF is about understanding policy change and stability, and the role of policy-oriented learning within processes of policy change and within policy subsystems. This has been the subject of considerable empirical investigation in a variety of settings. Over time, ACF has been applied to cases involving airline deregulation, telecommunications regulation, drug policy, energy policy, environmental policy, forest policy, health policy, water policy, ocean policy, pollution control, climate change, and intelligence, to mention only a few. Thus, it has mainly found its way into empirical applications to policy processes. With the focus on policy applications, ACF has paid attention to those actors usually involved in such policy processes, mainly interest groups, while other sets of actors may go unnoticed. For example, ACF has hardly at all been applied to party-political actors. This is surprising, given that political parties work to influence public policy and are sources of advocacy.

Much of the first wave of scholarship received criticism for putting forward wide claims unsubstantiated by anything like convincing evidence. The ACF was criticized for a bias towards pluralistic political systems, such as the United States. But with a growing number of applications, across a wide range of policies, the later waves of scholarship have done a lot in terms of addressing claims about the policy impact of advocacy coalitions. In the 1990s, the framework was developed through revisions to some of its assumptions (Weible et al., 2009). There was an acknowledgement that more needs to be understood about the actual conditions under which change takes place, that external perturbations are a necessary but not sufficient cause of change in the core policy attributes, and that one potentially important factor in this change is the role of the policy entrepreneur. One of the major revisions to the ACF was summarized in Sabatier and Weible (2007), where the framework was reformulated to ease applications outside of the pluralist system in the United States to corporatist systems that generally are less open, more centralized, and restrict participation. The revision identified two additional paths to policy change; internal subsystem events and dispute resolution through negotiated agreements involving two or more coalitions (Sabatier & Weible, 2007: 204–207). Subsystems, of course, can vary in their development, some being more well-developed than others; and they can overlap, which is more likely to occur when they are well-developed.

Sabatier and Jenkins-Smith (1993) assessed the ACF based on six applications (see also Jenkins-Smith & Sabatier, 1994). Later, Weible et al. (2009) examined 80 applications of the ACF spanning nearly twenty years (1987–2006). By taking stock of the existing applications, they identified and discussed some of the strengths and weaknesses of the ACF and offered directions for future research. The reviews showed that the ACF is applicable to various substantive topics, across various geographical areas, and in combination with other policy process theories and frameworks. The most tested hypotheses involved policy change, learning, and coalition stability. Hypotheses tended to be confirmed, yet questions remained about the membership, stability, and defection of coalition members; about the causal mechanisms linking external events and policy change; and about the conditions that facilitate cross-coalition learning. Continuing or emerging areas of research deserving theoretical and empirical attention included the role of institutions and resource dependence in the framework, policy subsystem interdependencies, and coordination within and between coalitions. In addition, the reviews found that coalition membership was relatively stable over time and that policy core beliefs glued coalition members together, but defection was also common. Hence, what was needed was original theorizing and deliberate research designs that investigated both the stability and defection of coalition members over time.

Another review of research discussed applications of the ACF in the Philippines, China, India, and Kenya, and concluded with an argument for the continued application of the framework outside of Western Europe and North America (Henry et al., 2014). Pierce et al. (2017) catalogued and analysed 161 applications of ACF from 2007 to 2014, a plurality of which in terms of depth examined environment and energy, subsystems at the national level, and utilized qualitative methods. More recently, Pierce et al. (2020) explored how the ACF’s theory of policy change was applied to 148 policy processes in 67 journal articles from 2007 to 2014. One of their main findings was the large number of applications in the environment and energy policy domain.

Regarding coalitions, the ACF argues that the line-up of allies and opponents tends to be rather stable and that actors within an advocacy coalition will show substantial consensus on issues pertaining to the policy core (Weible & Jenkins-Smith, 2016). Typically, such advocacy coalitions do not only operate in a single forum but also at different levels of policymaking (e.g., Shanahan et al., 2011). What we see is the coming together of a variety of actors, institutions, policies, and practices developed in diverse communities to tackle (cross-level) problems. This reminds of the key themes addressed in research on global or European governance, including how to assess the effectiveness of institutional arrangements. We must also remember that ACF research tends to concern deeply political processes. Therefore, it is not just a theory about policy processes but also about politics and decision-making. It uncovers agreements based on important relationships (coalitions) that can be even decisive in ultimately shaping policy or legislation. That policy success and the sometimes hidden nature of these coalitions in turn raise crucial questions about transparency and accountability.

Despite this increasing range of research, crucial questions remain about ACF and its generalizability. The ACF suffers from limitations at least some of which are recognized within the research community itself. There has been justified criticism about the lack of terminological precision. What precisely is an advocacy coalition and how can we identify them? How do advocacy coalitions differ from traditional actor constellations involved in policymaking? How are advocacy coalitions formed and maintained? How do people act in advocacy coalitions? To what extent is there policy learning inside the coalition and between them? Key ACF components, not least learning, are not easily measured. Preferences can develop endogenously within the coalition or exogenously. Coalitions may be measured in different ways. According to Satoh et al. (2023), previous research has been inconsistent in defining and measuring coalitions, which has hampered comparative research and theory building.Footnote 7

The ACF has been criticized for not taking collective action problems, such as shirking, seriously (e.g., Schlager, 1995). This means, to realize its potential more fully, admitting the explanations of collective action from frameworks based on instrumental rationality. And to incorporate more in-depth accounts of how coalitions form and maintain themselves over times, and of the types of strategies the coalitions are likely to adopt to pursue their policy goals. Drawing on general coalition research, not least coalition formation in parliaments and governments may help to consolidate the ACF further. According to Weible et al. (2011: 355), however, progress has been made in response to ‘criticism about collective action in coalitions, in part, through the specialized empirical efforts by multiple scholars on developing theory about coalition stability and structure as the outcome variables. Similar efforts are needed for learning and policy change’. But important issues related to coalition stability remain. ACF emphasizes competition among stable coalitions, which may hold opposing normative policy beliefs, but these are generally expected to result over time in convergence and policy stability. There is thus a tendency towards consensus and stability which might exaggerate the level of consensus within these coalitions or networks. Outcomes may reflect dominant coalitions and power imbalances. Within political families—Europarties and political groups—there is a tendency to claim consensus, that decisions have been reached unanimously, even when there is a lack of cohesion and controversies behind the decisions have not been completely solved. Building and claiming consensus reflect distribution of power. Policymakers generally want to display unity.

In this vein, a final consideration is that ACF, both theoretically and empirically, tends to underestimate the existence of hierarchies and asymmetries in resources and power (here defined as the way people can influence others or make decisions).Footnote 8 However, within the ACF research programme, we also find categorizations of coalition resources and applications of these categories.Footnote 9 In brief, changes in the institutional framework conditions can (re-)shape the availability of power resources. Variation between actors also matter with, for example, elected officials holding policymaking authority and bureaucrats holding expertise bringing different types of resources for the coalitions (Nowlin et al., 2022). And some players look set to benefit the most. In this context, it is interesting to note that, like the agenda-setting approach discussed in the previous section, ACF usually deals only with positive power and neglects negative power; the power to prevent other actors from devoting attention to specific issues. Of immediate relevance for our research purposes is a question raised by Weible et al. (2011: 356): ‘How and to what extent do coalitions capitalize on new resources to achieve greater influence in policy subsystems?’

ACF in the EU Context

The discussion above shows the gradual development of the ACF both theoretically and empirically. It has been refined in response to criticism, and the ACF provides a useful set of analytical tools by which to study highly significant processes of policymaking. ACF scholars have created a core community that regularly synthetizes findings from applications of the framework, giving the ACF the form of a true research programme (Cisneros, 2021). Scholars have also noted its relevance in understanding EU policymaking. In line with insights from agenda-setting literature, a consideration in the ACF is that actors and coalitions seek to maximize their advantage by strategic venue shopping—which in the EU context could mean coalition activities and coordination at different levels (subnational, regional, and national, European) and related to different EU institutions (Ingold, 2022; Sabatier, 1998). As Sabatier (1998: 121) noted, the multilevel governance system of the EU offers plenty of venues to influence policy development and therefore ‘the ACF should apply well to the increasingly complex set of relationships evolving within the European Union’. Rozbicka (2013), focusing on interest groups in her overview of ACF studies in the EU context, considered the ACF one of the most promising theoretical approaches in the continuously growing field of EU policy studies.

The role of the ACF in EU (policy) studies was addressed comprehensively by Ingold (2022), according to whom by 2018 as many as around 150 applications existed in Europe (see Pierce et al., 2017; Nohrstedt et al., 2020). However, most of those concentrated on national and regional policymaking, with only around 15% of European ACF studies dedicated to the supranational level, meaning EU politics. Highlighting the possibilities, challenges, and opportunities for studying policy change, coalitions, and actors at the supranational level, Ingold (2022: 567) offered a welcome addition to previous overviews of the existing literature in the field. Ingold further observed that like the global situation, the empirical applications most often covered energy, environment, and health policies—policy fields most studied by the ACF (cf. Pierce et al., 2017). Prominent examples of empirical applications include policy areas such as steel industry (e.g., Dudley & Richardson, 1999), employment policy (Johansson, 1999), taxation policy (Radaelli, 1999), agricultural policy (Nedergaard, 2008), wind power (Szarka, 2010), ecological risk assessment of pesticides (Hunka et al., 2015), smoke-free policy and health advocacy (Weishaar et al., 2015, 2016), tobacco regulation (Smith et al., 2015), the European Union Force (EUFOR) Althea operation (Palm, 2017), genetically modified organisms (Tosun & Schaub, 2017), pharmaceutical/public health policy (Brooks, 2018), biofuels (Rietig, 2018), and forest policy (Sotirov et al., 2021).Footnote 10

But, overall, what is the specific advantage of the ACF in the EU context? There is no simple answer as the research is continuously evolving, but the relevance of the ACF is seen in the growing number of studies employing the framework. Three patterns emerge from this literature. First, research has proven that the ACF is applicable within and across various policy issues. Echoing Sabatier (1998), Ingold (2022) noted that the ACF provides a suitable framework for studying EU policymaking, as it can be utilized for understanding how different public and private actors belonging to different countries and acting at diverse levels coordinate their efforts in pursuit of a shared objective. Importantly for our argument, ACF guides scholars towards looking beyond mere intergovernmental relationships and formal EU institutions. The gradual consolidation of both the jurisdiction of the EU and its political system obviously provides the basic operating environment for advocacy coalitions, but coalitions and policy entrepreneurs inside them utilize multiple informal channels for advancing their objectives.

The second pattern is continuing differences among researchers regarding the stability (or not) of advocacy coalitions and their internal power dynamics. Studies applying the ACF to investigating ‘formal’ coalitions formed by interest groups and nongovernmental organizations (NGOs) during consultation procedures conclude that they are ad hoc and short-term (Pijnenburg, 1998; Rozbicka, 2013; Warleigh, 2000), and thus differ from advocacy coalitions that are based on stable and long-term core beliefs. Warleigh (2000: 239–240) found that policy coalitions are not stable but issue-specific and concluded that ‘EU policy-making probably yields evidence of both advocacy and policy coalitions’. In his view, like Pijnenburg (1998), this finding justifies the claim that the pragmatic search for advantage is the primary shaper of coalitions in EU policymaking. Dudley and Richardson (1999) observed how coalition dynamics and the balance of power between competing advocacy coalitions may shift because key members choose to defect from one coalition to another. Key actors within these coalitions might be pivotal in the policy process. This brings us back to agenda-setting and those individuals who are policy entrepreneurs—advocates for proposals or for the prominence of an idea who can play a significant entrepreneurial role in the process of policy change. That boils down to the generic question of who has the final say in policymaking. With reference to the study by Newell (2018), Ingold (2022: 572–573) points out:

And this is true: not all coalition members have the same level of (decisive or formal) power, and it is crucial to know ‘who has an opinion, preference, or belief’, and who are the ones who have a more or less direct impact on the decision and thus on policy outputs and change.

Third pattern concerns ideas and discourse. Nedergaard (2008) found that discursive coordination was the most frequently used form of coordination during the 2003 reform of the Common Agricultural Policy (CAP). Radaelli (1999) drew attention to the political power of policy narratives—and the fact that a policy narrative developed—in the context of the advocacy coalition framework. Through studying tobacco regulation, Smith et al. (2015) showed how the ACF can be applied to one specific issue or one actor—the regulated tobacco industry. They highlighted the role of advocacy coalitions and how policy entrepreneurs with sufficient resources such as large corporations shaped the membership and direction of advocacy coalitions. In their case, the ability to shape regulatory reform involved the deliberate construction of a vaguely defined idea that could be strategically adapted to appeal to diverse constituencies. The question of how ideas interact with interests is, of course, long-standing in social science research. Shared terminology and discourse are indeed particularly relevant also in the case of Europarties.

To conclude this section, ACF has become more widely used in EU studies, but almost exclusively in relation to policies and policy change. The complexity of modern European political architecture, with a large number of actors and ideas involved in the processes, makes it difficult to isolate the independent effect of coalitions. The empirical studies also show how challenging it can be to identify and follow actors and their activities across EU levels, institutions, and venues, especially as policy change is often incremental and occurs over a longer time period. These methodological considerations are relevant also when assessing the influence of transnational partisan coalitions. Clearly, there is potential and the need to extend ACF to networks and coalitions of politicians and their parties. That means more attention to politicians’ transnational networks, not least the more informal ones. To our best knowledge, ACF approach has not been applied to transnational settings encompassing partisan actors, particularly regarding how Europarties shape ‘history-making’ decisions such as Treaty reforms. In the next section, we combine the agenda-setting and advocacy approaches with the concept of transnational partisanship, particularly in relation to main constitutional and institutional developments in European integration.

Unpacking Transnational Partisan Advocacy Coalitions

We utilize the ACF and agenda-setting approaches for exploring the role of partisanship and Europarties in EU governance, especially in relation to Treaty negotiation and revision. In fact, the basic characteristics of the Europarties and their external environment suggest that advocacy and agenda-setting are fundamental for their influence. Existing research on ACF has analysed primarily policy change and has paid hardly any attention to political parties, whereas our main focus is on the actors themselves—the Europarties and their networks. We identify them as stable and institutionalized advocacy coalitions, but also recognize that their membership and internal power dynamics may vary over time and depend on the specific context of the bargaining process.

While the ACF has generally been applied to national policy processes, recent decades have witnessed a burgeoning literature on transnational advocacy networks (TANs), usually building on Keck and Sikkink (1998). Parts of this literature draw on ACF research to develop integrated frameworks for empirical applications. A few examples are Litfin (2000) dealing with advocacy coalitions in the case of globalization and Canadian climate change policy; Farquharson (2003) examining global tobacco advocacy networks; Pralle (2003)Footnote 11 on the internationalization of Canadian forest, supporting the theory of ‘venue shopping’, too; Zippel (2004)Footnote 12 on a TAN involving advocates and policy expertise regarding sexual harassment in the EU; Carpenter (2007) studying advocates in the human rights sector, asking why some issues but not others galvanize TANs; Novak (2020) applying the lens of transnational advocacy networks to human rights litigation; and Holzscheiter et al. (2021) who examine advocacy coalition constellations and norm collisions in international drug control, human trafficking, and child labour. Hence, the last three decades have seen a growth in transnational advocacy coalition or network studies. It reflects the real-world growth in transnational political mobilization by ‘non-state’ actors, as well as the existence and role of advocacy coalitions across countries. This implies that a wide range of advocacy coalitions is being globalized, which also has implications for our understanding of international and regional politics. The extent to which they make their impact felt will depend on their resources, mobilization, and political strategy for reaching policy goals.

However, most of this literature emanates from research on interest groups and social movements or public participation. Political parties, whether national or transnational, do not feature prominently, if at all, in these research programmes. Yet in our increasingly interdependent world parties too have incentives for transnational activism, for engagement in spheres of cross-border governance in its broadest terms. More specifically, we suggest that the growth in the jurisdiction of the EU has presented national parties with functional pressures for transnational engagement, not least through the changing ‘political opportunity structure’ related to European-level institutional development and policymaking. This is a general theme that will surface throughout the remaining chapters in this book. In other words, we suggest a return to traditional scholarship on political engagement emphasizing political parties, alongside interest groups and social movements, as ways to influence public policy. But some political movements are likelier than others to go transnational.

A counterintuitive example of transnational action is when nationalist (often regional) movements or parties work together across borders, that is, in transnational nationalist advocacy (Gupta, 2008). In a similar vein, the transnational, cross-national coalition-building of political parties classified as ‘radical right’ or ‘populist’, which tend to share anti-immigration positions, has something contradictory about it (e.g., McDonnell & Werner, 2019; Steven, 2020). Obviously, the incentives for such transnational activity are strong enough, not least in the EP with its various resources available to parties and individual members. At the same time, most of these nationalist parties are (so far) not involved in EU policymaking through the other main institutions: the Commission, the Council, and the European Council. (Ethno-)nationalist parties, whether populist or under any other label, provide a critical case: if these parties engage transnationally, we can expect parties belonging to all other party/ideological families to do so.

More broadly, transnationalism denotes all types of cross-national contacts if one relaxes the assumption that states are the only units or actors interacting across national boundaries. As argued by Keohane and Nye (1977: 24–25), multiple channels connect societies, including transnational relations, where transnational ‘applies when we relax the assumption that states are the only units’. We are dealing here with interactions across national boundaries: face-to-face and through various communication channels. To classify a relationship as ‘transnational’, researchers typically look for how non-state actors of different kinds interact across national boundaries. They include a wide range of actors which need to be differentiated into subcategories or subtypes. For example, political parties are fundamentally different from business firms. And the degree of institutionalization of transnational coalitions varies from loosely structured networks to joint transnational organizations which are actors and not just arenas.

In the EU context, a new scholarship on transnational relations emerged in the 1990s which paid attention to policy effects more broadly. In the words of Risse-Kappen (1996: 58):

As for the EU, one would assume that the increasingly dense network of transnational coalitions and organizations—from transnational interest groups […] to European party organizations—not only affects EU policies directly, but also the processes of national preference formation as mediated by the domestic structures of the Member States.

At that point, the evolving Europarty organizations or their forerunners had already existed for two decades. And since the 1990s the political and institutional environment in which the Europarties exist—and which they have purposefully shaped—has altered fundamentally.

In our opinion, the contribution of advocacy coalitions to policymaking goes further and much deeper than existing research on EU governance has accounted for. It extends also to long-term constitutional and institutional changes advocated through transnational relations between political parties. Such relationships and presence therein allow individuals to cooperate for the shared objective of promoting European unity or particular policies. This demonstrates how the European Union functions not only through the interplay between member states or the EU institutions, with their respective mandates, but also through continuous transnational partisan interaction.

Coalitions are of course a central feature of politics. For individual actors and collective of actors alike, coalition-building serves to pool resources and power and facilitates influence over outcomes. Coalitions help to simplify the process of decision-making. This basic demand for cooperation prompts actors to form coalitions. However, there are alternative theoretical expectations for coalition formation. The partisan hypothesis suggests that actors form coalitions primarily based on ideological affinity, as defined by party affiliation (see Chapter 4). By contrast, other theories point to power, interest, and culture as driving concerns.

When transferred to the politics of coalition formation at the transnational/supranational (EU) level, our starting point from the ACF is that actor alignments will reflect ideological proximities and divides, as defined by party affiliation. Transnational partisan coalitions are expected to be stable, and we also expect their networks and memberships to expand over time—both in response to their internal consolidation and to the gradual empowerment and enlargement of the EU. Coalition patterns are also likely to be stable across issue areas due to both ideological proximity within coalitions and the fact that actors within the coalition grow accustomed to formulating common positions. At the same time, coalitions may well cut across ideological divides. In the context of the EU, initiatives and outcomes are often the result of coalition-building across political families. These coalitions both compete against one another but also join forces in pursuit of common objectives. That said, ideological divisions are a constant in party politics whether nationally or transnationally.

In the study of EU politics, the notion of partisan coalitions has received extensive support in the empirical work on the European Parliament (e.g., Ahrens et al., 2022; Bressanelli, 2014; Hix et al., 2007). Less has been written about the other EU institutions in this respect, but there are some assessments of ideological or party affinity in the Council and particularly the European Council (e.g., Hix & Lord, 1997; Tallberg & Johansson, 2008; and the literature cited in chapters one, three, and four of this book). Translated to the context of European Council decision-making, the partisan hypothesis generates the expectation of a party-political divide, with participants (at the highest level the prime ministers or presidents of the member states) coordinating positions within the dominant transnational Europarties—particularly the three party families we examine in this book, the European People’s Party (EPP), the Party of European Socialists (PES), and Alliance of Liberals and Democrats for Europe (ALDE). European Council summits are preceded by transnational party caucuses, and there are gatherings of government ministers represented in the Council. It is also an empirical question to what extent such Europarty caucuses play an independent political role or are primarily used by national governments for advancing their own agendas. While the status of heads of government as representatives also of national political parties—and involved in Europarties one way or the other—generates expectations of partisan alignments also in the European Council, these same political leaders are first and foremost responsible to domestic constituencies. Not least therefore, there are important limits to the trend of party politicization in the EU and to the influence of Europarties.

The influence of transnational Europarties is also conditioned by their capacity to operate effectively. What matters is their mobilization capacity to shape outcomes and, even when enjoying numerical superiority, the ambition and capacity to coordinate positions. In the process, there may be a convergence of preferences and positions, both within coalitions and between them. Hence, the simple existence of transnational coalitions is no guarantee of their effectiveness. In our empirical analysis in Chapters 4 and 5, we highlight the mobilizing role of both organizations—Europarty headquarters, EP political groups, political foundations etc.—and individuals (entrepreneurs) that belong to the coalitions Europarties have built and enlarged over several decades. Important in our approach is the temporal dimension: while transnational partisan coalitions may not always achieve their goals, their advocacy is continuous and ever-present. It is therefore crucial to extend the empirical lens beyond the European Council summitry and specific Intergovernmental Conferences (IGC).

A rare insight into life inside a transnational partisan advocacy coalition, and one of the first studies in the EU context drawing on the ACF, was provided by Johansson (1999). It unpacked transnational advocacy coalitions related to the European employment initiative (EEI). The coalition included trade unions and political parties as well as governments and EU institutions, and individuals within them. Those constituting the advocacy coalition favoured an employment title in the revised Treaty, and stronger coordination between member states regarding employment policies. While national governments were involved and pivotal for the outcome, it was to a large extent driven by transnational advocacy (see also Ladrech, 2000; Lightfoot, 2005). PES, both as an actor and arena, championed the employment chapter or title in the Amsterdam Treaty that came into effect in 1999 (see also Johansson, 2017; Tallberg & Johansson, 2008). There were transnational exchanges throughout the process, notably within transnational party caucuses preceding European Council summits. There was a clear party-political pattern behind the support given by individual governments, and that case study also highlighted continual advocacy, emerging convergence, and the role of individuals as policy entrepreneurs—themes that we shall analyse in our empirical chapters. Johansson (1999: 97) was also careful in assessing the independent effect of the partisan coalition: ‘the transnational dimension of EU policy-making and treaty reform must be seen as a complement to rather than a substitute for what was happening at the intergovernmental level’. In similar vein, Johansson and Raunio (2005) in their analysis of Europarty funding explored cross-party coalitions capitalizing on ‘incomplete contracts’ ever since the opening created by the Party Article in the Maastricht Treaty and how they successfully and continuously argued for a stronger position for the Europarties.

Our approach carries implications for theoretical approaches in agenda-setting and advocacy coalitions, as well as for research on transnational relations and the ever-growing literature on European integration in its broadest terms. We suggest that this literature, with some exceptions, has overlooked the important form of interaction through transnational party networks. In contrast to accounts that continue to emphasize intergovernmental interaction, we specify and map an additional and central mechanism through which processes of EU decision-making are taking place and European integration evolves. We call this mechanism the transnational partisan dimension of European integration.

We argue that Europarties should be viewed primarily as transnational partisan actors that operate both at the intergovernmental (Treaty reforms, European Council) and supranational (European Parliament, Commission, legislative, and day-to-day policymaking) levels of EU politics. This is evident in the gradual development of the Europarties and in their organization and identity. We discuss how the multilevel nature of the EU polity provides several channels for advancing policy objectives, and how the Europarties have throughout the decades established their own networks and coalitions for shaping the future of Europe. Coalitions work on many levels and their impact is cumulative. We further argue that the transnational character of the Europarties, often seen as their weakness, has enabled them to influence European politics in ways that are hidden from public view. We will develop this argument in subsequent chapters. Moreover, we pay attention to the discourse and terminology employed by the Europarty entrepreneurs. Successful advocacy often depends on or is at least facilitated by framing of issues (Daviter, 2011). ‘Democracy’, ‘legitimacy’, ‘representation of citizens’, and ‘participation’ are examples of ‘big words’ traditionally employed by Europarties in their documents. Europarties may differ in their concrete objectives and programmatic priorities, but terminology is essentially similar.

We now proceed to the empirical analysis of Europarties and examine the core hypothesis that they have contributed to fundamental institutional change in the EU, by shaping the agenda of Treaty reforms and bringing together advocacy coalitions that continuously engage in agenda-setting about the broad theme of the ‘future of Europe’. We begin with a chapter on Europarties themselves and their networks and then turn to an analysis of observable patterns of Europarty attempts at influence in Treaty reforms in the past as well as in the 2021–2022 Conference on the Future of Europe and beyond.