This book is about the crisis management of the European Union (EU) and its member states during the refuge crisis of 2015–16 and its aftermath. We focus on crisis policymaking and crisis politics during this crisis, which reached its peak in 2015–16, but continued to occupy European policymakers for several additional years. This was not the first refugee crisis in Europe, and its coming was not entirely unexpected. The inflow of asylum seekers into the EU had already started to rise before 2015, but in the first half of 2015, the number of arrivals accelerated, and it virtually exploded in the fall of that year. The asylum seekers crossed the Mediterranean between Turkey and Greece in ever larger numbers, proceeded along the Balkan route, and arrived in Hungary, from where they continued their journey toward Austria, Germany, and the Scandinavian countries. The crisis’s emblematic event occurred on September 4, 2015, when thousands of asylum seekers decided to leave the central train station in Budapest, where they had been stuck for some time, and to march on along the Hungarian highways in pursuit of their stated goal of reaching German soil. The Hungarian government, all too pleased by the asylum seekers’ decision to move on, facilitated their arduous trek toward the Austrian border by sending buses to accommodate them and bring them to the border. Faced with the prospect of the approaching caravan, the Austrian government urgently sought the help of the German government. It was during the night of this Saturday in September 2015, under the immediate pressure of the refugees proceeding toward the Austrian–Hungarian border, that the German chancellor made the critical decision to suspend the Dublin III Regulation and to admit asylum seekers to Germany, although they had already passed through several other member states of the union. This decision was later to haunt her as she tried to find a joint solution to the crisis with her fellow heads of government in the EU. It proved to be very hard to come to a joint approach to the crisis, and it was impossible to share the burden among the EU’s member states.

The puzzle we are trying to elucidate in our study of the refugee crisis is why the German chancellor, Angela Merkel, had come to be trapped in such a desperate situation in early September 2015, and why she and her fellow heads of government together with EU agencies proved to be unable to reform the Common European Asylum System (CEAS). It is not as if the European policymakers did not see the crisis coming. But although they were aware of what was brewing, they did not jointly prepare to meet the inflow of asylum seekers in the short term. Nor did they, once the policy failure of the CEAS was there for everyone to see, get their act together to reform the system in the long term. They only came up with a stop-gap solution, which made them dependent on less-than-reliable third countries. Answers to this puzzle do not just speak to the refugee crisis 2015–16 (from now on referred to as “the refugee crisis”); the way the EU and its member states faced this crisis goes a long way toward clarifying how the EU works more generally.

In the two-year period 2015–16, the member states of the EU received no less than 2.5 million asylum applications, mainly – but not exclusively – from Syrian refugees who had fled the civil war in their country. Under the pressure of this exceptional inflow of asylum seekers, the prevailing EU asylum policy and the asylum policies in the member states were put under enormous pressure, and existing conflicts within and between member states relating to the management of refugee flows and asylum requests were exacerbated. The pressure varied, however, from one member state to another, with important implications for policymaking. The way the EU and its member states reacted to this pressure demonstrates how cooperation is difficult in a situation, where they are not all hit in the same way, and in a policy domain where the EU and its member states share competences. In asylum policy, cooperation is rendered even more difficult by the fact that it is highly contested in the member states themselves. Already before the refugee crisis 2015–16, the humanitarian imperative to accommodate asylum seekers had been challenged by the European radical right in the name of national sovereignty and the protection of national cultural traditions. The refugee crisis served to increase the salience of migration issues and to reinforce the resistance of the radical right to the reception and integration of refugees.

It is important to study the refugee crisis because it has been most salient among the European publics, as we found in a survey put into the field in summer 2021. Asked about the “most serious threat to the survival of the European Union” in the decade before the arrival of the Covid-19 pandemic,Footnote ¹ almost a third (32 percent) of the citizens from sixteen countries considered the refugee crisis to be the most important threat,Footnote ² outdistancing the other recent EU crises, such as the Euro area and Brexit crisis. Importantly, the assessment of the threat to the EU’s survival varied by region: It was particularly in the northwestern European member states where most asylum applications were registered and in the eastern European member states where resistance to joint burden-sharing was the most intense that the population deemed the refugee crisis to be the most threatening to the EU. By contrast, while the refugee crisis was ranked highly by a significant portion of the population there, too, southern Europeans considered the threat of the financial and economic crisis and of the poverty and employment crisis as considerably more important than the refugee crisis, and the citizens of the UK and Ireland perceived the Brexit crisis as the biggest threat.

As a matter of fact, the way the refugee crisis was managed has left behind conflicts between member states, which have been further exacerbated in subsequent crises and which are likely to haunt the EU in times to come. Moreover, against the background of the underlying integration–demarcation conflict in the national European party systems, asylum policy constitutes a latent time bomb that might explode at any moment if inflows of asylum seekers increase again and the issue becomes once again more salient. Asylum policy remains a potent means for electoral mobilization on the left and on the right. The large opposition to immigration in some member states is bound to constrain the future options available to policymakers, as it is likely to constitute a major obstacle to joint solutions.

At both the EU level and the level of the member states, we investigate the kind of conflicts that were triggered by the problem and political pressure the EU and its members were exposed to during the crisis, how these conflicts influenced the way they attempted to deal with the pressure, and the kinds of policy solutions they adopted in the short and longer term. At the EU level, cooperation between the member states was, if anything, even more demanding than at the national level, because of the fragmented competence structures in asylum policy and because both the intensity and the type of problem pressure varied significantly between the member states. While the member states that were directly hit by the crisis in one way or another sought the cooperation of the others, the more fortunate among the member states were not prepared to contribute to joint solutions, or at least not to lasting joint solutions. We investigate the attempts to overcome the initial unilateral scramble to the exit by the member states and ask what kind of transnational conflicts were exacerbated or newly created by these attempts and to what extent they prevented joint solutions. We pay particular attention to the interaction patterns between the national and the transnational conflicts in policymaking during the crisis.

As we shall see, conflicts within and between member states during the refugee crisis were very intense, and the prevailing EU asylum policy proved to be impossible to reform during the crisis. This does not mean that any joint solution was impossible. We demonstrate that the member state governments found provisional stop-gap solutions that did reduce the problem and political pressure in the short and medium term, even if they did not produce a long-term policy solution. As a result, asylum policy remains an unfinished construction site that constitutes a latent threat to the resilience of the EU polity to the date of writing.

To answer our key puzzle, we intend to embed the refugee crisis in a broader theoretical framework that allows us to situate crisis policymaking and crisis politics more generally in the EU polity and in Europe’s underlying conflict structures. In order to understand the difficulty of coming to joint decisions in asylum policy, we need to first grasp the fragmented and nontransparent decision-making structure in the multilevel EU polity in general and in EU asylum policy in particular. Second, we need to get a sense of the already existing fractures in the member states and between them – fractures that were then exacerbated in the crisis or complemented by newly created divides as a result of the way some member states attempted to come to terms with it.

A General Framework for the Analysis of Crisis Policymaking and Crisis Politics

At a first glance, the refugee crisis threatened at most the resilience of the Schengen area and the principle of free movement. Designating it as a “deep” crisis that threatened the survival of the polity as a whole might, therefore, seem somewhat overblown. However, we claim that it should be at least considered as such a crisis, because it revealed fundamental tensions undermining the resilience of the EU polity and its capacity for designing joint EU policy. To understand this, we build on Stein Rokkan’s structural approach to the formation of the European state system as it has been applied to the process of European integration by Stefano Bartolini (Reference Bartels2005). This approach has the advantage of being situated at the intersection of the literatures on European integration and comparative politics. We complement this macro-structural approach with insights from the grand theories on European integration and concepts of policy analysis, which will allow us to link the macro-structural context to policymaking in general and to policymaking under crisis conditions in particular.Footnote ³

Our framework is not generally applicable; rather, it is specifically focused on the context of the EU polity, since we are interested in how the refugee crisis was managed in Europe. As is well known, of course, the EU is quite an exceptional polity, which has important implications for the way the refugee crisis – or, for that matter, any Europe-wide crisis – is managed. The EU is composed of a set of heterogeneous member states that are constituted as nation-states – that is, polities characterized by the successful integration of their economic, cultural, administrative, and coercive boundaries (Bartolini Reference Bartels2005). Over a period covering several centuries, in each member state, the closure of external boundaries has created three processes of internal consolidation: center formation (the creation of authority structures), system maintenance (the creation of loyalty, identity, and solidarity among the locked-in population), and political structuring (the creation of organizations, movements, and institutional channels for the articulation of the population’s voice). The combination of boundary building (bounding), center formation (binding), and system maintenance (bonding) – the three B’s of the “polity approach” to the EU integration process (Ferrera, Kriesi, and Schelkle Reference Ferrera, Kriesi and Schelkle2023) – has provided the member states with an idiosyncratic structure of opportunities and constraints for the internal political structuring.

In the nation-state, external closure and internal structuring (voice) are intimately linked, as are opening and destructuring (exit)Footnote ⁴: As the people in a given territory can no longer escape the binding decisions of the political authorities at the center, they demand participation in the political process and organize collectively in order to make their claims known and to impose themselves against opposing claims. The external closure induces social interactions among the locked-in actors, which increases the likelihood of collective action among them, “domesticates” the actors’ strategies, and focuses them on central elites (forcing them to become responsive to pressures from below). Political structuring within the nation-states results from the strategic interaction of collective actors and the stabilization of these interaction patterns, which produce national policies. Importantly, this structuring has occurred in a way that is specific to each nation-state and has focused policymaking and politics on the national center.

Compared to the nation-state, the EU and its member states constitute a new type of polity with a rather unique character that we attempt to capture by the notion of the “compound polity of nation-states” (Ferrera et al. Reference Ferrera, Kriesi and Schelkle2023). At its core is a fundamental tension that the European integration process has introduced in the European system of nation-states (Bartolini Reference Bartels2005: 368, 375), a tension that is exacerbated by the fact that it is the governments of the nation-states that are the drivers of the integration process. On the one hand, the process of European economic (and other forms of) integration is predicated upon the removal of boundaries between the European nation-states. On the other hand, the national, democratic, and welfare features of the union’s member states (the features that were left outside the initial integration project) are predicated upon the continued control over redistributive capacities, cultural symbols, and political authority by the member states. The integration project progressively represents a direct challenge to these other features of the member states. The integration process breaks up the three-layered coherence between identities, practices, and institutions; dismantles the coincidence among the different types of state boundaries; and leads to the dedifferentiation of European nation-states after five centuries of a progressive differentiation in their legal and administrative systems, social practices and cultural and linguistic codes, economic transactions and market regulation, and social and political institutions. As Bartolini (Reference Bartels2005) points out, the integration process is causing the destructuring of national polities without sufficient restructuring at the EU level.

This was never more evident than in the period of the refugee crisis. The fundamental tension between the integration process and the destructuring of the national polities becomes particularly critical in crisis situations, above all in a policy field like asylum policy, where some, albeit not all, member states are jealously defending their national sovereignty against the encroachment of European integration. Routine policies in established polities (such as nation-states) have only marginal implications for the maintenance of the polity itself. However, the combination of the lack of a joint policy on border control, outdated asylum policies that were concocted at a different juncture, the ability to follow beggar-thy-neighbor approaches, isolated national policies, and finally a resistance to share the common burden meant that what should have been a routine policy problem challenged the bounding, the binding, and ultimately the bonding of the EU member states, revealing the fundamental tensions in the EU’s architecture. In other words, policymaking in crisis situations is more likely to impinge on the maintenance of the polity as such, and this applies in particular for a compound polity like the EU, where a stable underlying structure has not (yet) been established. As a compound polity, the EU is constantly testing new modes of combining its three constitutive elements, that is, boundaries, binding authority, and bonding ties.

Taking this into account, Figure 1.1 (taken from Kriesi, Ferrera, and Schelkle Reference Kriesi and Hutter2021) presents the five building blocks of our general analytical framework. The three B’s and the preceding discussion are located as the initial “block” of our model and structure the policy space afforded to European policymakers. The actual policymaking, which lies at the heart of our analysis, is constrained by this “compound” EU structure and the conflicts it generates and, furthermore, by the policy heritage begotten by this structure, that is, the lackluster border control coordination and the semifunctional joint asylum framework, and also by the immediate problem and political pressure. In turn, the crisis policymaking reshapes the bounding, binding, and bonding status quo as new institutions and actions attempt to face the crisis, contributing to or hindering polity maintenance and eventually leading to one of the outcomes indicated in our final building block.

Figure 1.1 The analytical building blocks of the theoretical framework

The challenge of the refugee crisis focused on bounding, that is, on the internal and external bordering of the EU, with important implications for binding and bonding. In the EU, the master tension is exacerbated by the fact that the integration process breaks down internal borders without, at the same time, providing for commensurate joint external border controls. Accordingly, migration governance currently has two components in the EU: free movement internally, and a common migration and asylum policy with regard to third country nationals (TCNs). Put simply, the EU has an open borders framework internally (the Schengen area) but external migration restrictions (Geddes and Scholten Reference Geddes2016). However, while EU member states have little control over internal movements,Footnote ⁵ they remain in charge of regulating admission of TCNs, a prominent group among whom have been asylum seekers.Footnote ⁶ Though matters of asylum are notionally a shared competence between the EU and national governments (article 4 of the Treaty on the Functioning of the European Union [TFEU]), at the end of the day, it is the member states themselves that determine access to their territory and whether and how they will abide by international norms (Schain Reference Schain2009), the amount of resources they are willing to invest in the assessment of asylum claims, policing efforts against irregular migration, deportation procedures, and the integration of successful asylum applicants. Moreover, the ability of the EU to control its external borders extends only as far as the capacity of the member states at its external borders to fulfill this task. As a result of insufficient control of external borders, the refugee crisis was first an instance of the breakdown of external borders in the southern European border countries most exposed to the inflow of refugees. Greece, in particular, had border control issues, which created tensions that jeopardized the Schengen area’s continued existence.

As they struggled to regain control, decision-makers both in the EU supranational institutions and in the member states, particularly those most affected by the refugee crisis due to their country’s exposure, implemented a set of measures that amounted to what Schimmelfennig (Reference Schimmelfennig2021: 314) calls “defensive integration,” that is, a combination of measures of mainly internal rebordering (the resurrection of barriers between member states or their exit from common policies or the EU altogether) with external rebordering, that is, the creation and guarding of “joint” external EU borders, policed partially by a common armed force, that are institutionally recognized as the union’s borders in treaties and agreements with third countries. Combined with internal debordering, external rebordering contributes to “effective integration” (Schimmelfennig Reference Schimmelfennig2021: 314), as the bounding process of the EU acquires meaningfulness at the expense of the national bounding. By contrast, the combination of internal and external debordering would lead to an outcome of “disintegration.” From the perspective of the European integration process, “defensive integration” appears as a second-best solution that is basically one step forward, one step backward – or a “failing forward” (Jones, Kelemen, and Meunier 2016, Reference Jones, Kelemen and Meunier2021; Lavenex Reference Lavenex2018) – approach with regard to integration, an outcome that combines elements of stagnation and adaptation in our framework. While our description of the outcome of the crisis is in line with the failing forward approach, we focus on the policymaking process, which is given short shrift by this approach.

Our Argument in Brief

Our focus on the policymaking process puts the making of binding decisions at the center of the analysis. Our basic argument is that, against the background of the underlying conflict structures at the EU and the national levels, the policy-specific institutional context within the compound polity (the competence distribution in the policy domain and the institutionalized decision-making procedures governing crisis interventions) and the characteristics of the crisis situation (the intensity and distribution of the problem and political pressure among member states) jointly determine to a large extent the way policymakers attempt to come to terms with the crisis.

Generally, the crisis-induced distribution of problem and political pressure may be more or less symmetrical. Crucially, in the refugee crisis, the incidence of the crisis across EU member states was asymmetric. Some member states were hit hard by the crisis, while others hardly experienced any problem pressure at all. Uneven exposure to a crisis creates a differential burden of adjustment, which increases the salience of national identities and limits transnational solidarity. In other words, an asymmetric crisis activates the underlying integration–demarcation conflict. In the case of the refugee crisis, the activation of this conflict was enhanced by the fact that it concerned, above all, external and internal boundaries. By contrast, the presence of a common, symmetrical threat experienced by all the member states of the EU multilevel polity is likely to be a powerful driver of expanded solidarity between member states. As in the Covid-19 crisis, the shared experience of a crisis may reduce the salience of constraints imposed by national identities and facilitate an extension of transnational solidarity. The uneven incidence of the refugee crisis among the member states makes for a complex configuration of transnational interests and facilitates the creation of “circles of bonding,” that is, coalitions of member states that are strengthened by the crisis and that lead to divisive bonding instead of systemic bonding that enhances the integration process.

In the absence of a joint approach to the looming threat of the crisis, unilateral actions on the part of some member states become more likely, with individual member states reacting to their specific crisis situation and relying on their own policy legacies. In the compound EU polity, such unilateral actions lead to externalities or spillover effects for other member states. Because of the dysfunctionality of the CEAS and the interlocking of EU and national policymaking in European asylum policy, the refugee crisis has engendered a large number of such spillover effects, giving rise to numerous cross-level and transnational interactions and conflicts, which, in turn, have rendered policymaking not only more complex but also more vulnerable to obstruction by some member states.

With respect to the institutional context of policymaking, we highlight four aspects. First, we take into account the policy-specific distribution of competence in the EU polity. In policy areas where the EU has high competence, it is more likely for European institutions to be situated at the heart of the crisis resolution process. Instead, where EU competences are low, European institutions lack the capacity to make an independent impact on crisis management. In the asylum policy domain, the EU has rather low competences and depends heavily on intergovernmental coordination among member states. In this domain, responsibility is shared between the EU and the member states, and the mixture of member-state interdependence and independence imposes reciprocal constraints on policymakers at each level of the EU polity. The limited competence of the EU in the asylum domain posed a great challenge for policymaking in the crisis, a challenge that was enhanced by the diversity of policy heritage as well as by the uneven incidence of the crisis in the various member states.

Second, we consider the institutional power hierarchy between member states. Depending on their size and resources, member states have more or less institutional power in the EU and are expected to contribute a larger share to the common public good. Moreover, informally, large states may also provide leadership for crisis resolution. This more or less institutionalized power hierarchy may be reinforced (as in the Euro area [EA] crisis), but also undermined (as in the refugee crisis), by crisis-induced power relations, which depend, in turn, on the distribution of the crisis incidence. Thus, Germany, the most powerful member state, was unable to play the role of a stabilizing hegemonic power in the refugee crisis because its institutionally strong position was undermined by the joint effect of the EU’s limited policy-specific competences and the crisis-induced spillover processes between member states.

Third, as regards the decision-making mode, we insist on the importance of what we call executive decision-making. Building on new intergovernmentalism, which stresses that intergovernmental coordination has become the key decision-making mode in the EU in general and in crisis situations in particular, we focus our attention on executive decision-making in the crisis. In the EU, this decision-making mode involves the heads of governments of the member states in a dual role – that of head of state or government representing a country in European negotiations and that of member of the European Council representing Europe back home. As a result of this dual role, the chief executives of the member states become the pivotal actors in the two-level game linking domestic politics to EU decision-making. Accordingly, we expect the governments of the member states and their key executives to play a pivotal role not only in domestic policymaking but also in policymaking at the EU level.

Last, but certainly not least, the focus on heads of member state governments crucially introduces partisan contestation into the management of the refugee crisis, since, at the level of the member states, the national governments are exposed to party competition. Building on postfunctionalism, we argue that the refugee crisis lent itself to the activation of the integration–demarcation divide in national party competition, providing a golden opportunity for the radical right to mobilize against the governments’ and the EU’s asylum policies. Exploiting the political explosiveness of asylum policy, in some member states, not only the radical right opposition but even government parties seized the opportunity of the crisis situation to create divisive coalitions of member states, which rendered the search for joint solutions extremely difficult in the refugee crisis. This final point of our argument indicates that we do not exclude the possibility of endogenous political sources of a crisis. But we maintain that strategies of “crisisification” (Rhinard Reference Riker2019; Boin, ’t Hart, and McConnell Reference Boin, Hart and McConnell2009; Rauh Reference Rauh2022) are not at the origin of great crises such as the refugee crisis, even if they can exploit such crises once they have come about. Such crises have largely exogenous origins that create a situation of urgency and uncertainty for policymakers, who are taken by surprise – although they might have seen the crisis coming. Policymaking in the crisis situation takes place under great pressure and produces policy-specific conflict configurations, constraints, and opportunities that may have consequences for the maintenance of the polity itself.

The Focus on Policy Episodes

For our analysis of the policymaking processes, we break down the management of the refugee crisis into a set of key policymaking episodes, which are triggered by salient policy proposals at both the EU and the national levels. Overall, we consider six EU-level episodes and five episodes each in eight member states. A policy episode covers the entire policy debate surrounding specific policy proposals that governments put forward, from the moment the proposal enters the public realm to the moment it is implemented and/or the related debate peters out. We carefully select the most important policy episodes during the refugee crisis at the EU level and in the individual member states. Episodes constitute more or less clearly delimited political developments that allow for “a disciplined and limited kind of dynamic research,” which, as Lazarsfeld, Berelson, and Gaudet (Reference Lipset, Rokkan, Martin Lipset and Rokkan1968[1944]: xxi) suggested in the preface to the second edition of their classic study The People’s Choice, holds the greatest promise for the future development of the social sciences.Footnote ⁷ Episodes are composed of actions by a stylized set of individual and collective actors. The focus on policy episodes makes it possible to systematically analyze and compare the policymaking process across levels and countries during the crisis. We have developed a new method – policy process analysis (PPA) – that is specifically suited to the comparative analysis of such episodes.

For the analysis of political conflicts within the episodes, we use some key concepts, which we briefly introduce here: political structuring, politicization, and conflict intensity. As conceptualized by Bartolini (Reference Bartels2005: 37), political structuring refers to the structural preconditions that allow the expression of voice. Bartolini uses this term “to point to the formation of those institutional channels, political organizations, and networks of relationships that allow for individual voice to achieve systemic relevance.” Conflicts are politicized within such structural preconditions. In addition, we adopt the broadly shared understanding of the concept of politicization (e.g., de Wilde, Leupold, and Schmidtke Reference Zaller2016; Hoeglinger Reference Höhmann and Sieberer2016; Hutter and Grande Reference Hutter and Kriesi2014; Rauh Reference Rauh2016; Statham and Trenz Reference Steenbergen, Scott, Marks and Steenbergen2013), which builds on Schattschneider’s (Reference Schimmelfennig1975) notion of the “expansion of the scope of conflict within a political system” (Hutter and Grande Reference Hutter and Kriesi2014: 1003). More specifically, we distinguish between two conceptual dimensions that jointly operationalize the concept of politicization: salience (visibility) and actor polarization (conflict, direction). Conflicts are politicized to the extent that they are both salient and polarized. Politicization can characterize an entire episode or the actor-specific contribution to the politicization of the episode. Conflict intensity provides an additional aspect of the conflict in a given episode. While the polarization measure does take into account the direction of the actors’ position on the policy in question, it says little about the nature and intensity of the conflict. The conflict intensity concept takes these additional aspects into account.

In relying on the concept of politicization in particular, we assume that, under contemporary conditions of “audience democracy” (Manin Reference Marsaglia, Tsang and Wang1997), policymaking is generally taking place under the close scrutiny of the media and of the attentive public. In addition, we assume that public scrutiny is particularly close in instances where policymaking is no longer confined to policy-specific subsystems but becomes the object of “macro-politics” (Baumgartner and Jones Reference Jones and Baumgartner2002), as is typically the case in crisis situations. Even under such conditions, however, not all policymaking is equally likely to become the object of the expansion of the scope of conflict in the public sphere. Some policymaking remains in the realm of “quiet politics” (Culpepper Reference Culpepper2011), confined to experts and technocratic problem solvers, and sometimes even top brass politicians succeed in avoiding the limelight of the public, at least for a few decisive moments. To be sure, we consider only key episodes of policymaking during the crisis, which are particularly likely to get politicized. But, as we show, even within this highly selective set of episodes, there is great variation in the extent to which they have become politicized. We inquire into the factors determining the level of episode-specific politicization.

Overview of the Volume

The volume is divided into four parts. The first introduces our theoretical and empirical approach in more detail and presents the context of the crisis – the crisis situation and the variety of episodes of policymaking to which it gave rise. Part II covers the actors and conflict structures at the two levels, while Part III analyzes the dynamics of policymaking and pays particular attention to the interaction between the two levels. Part IV addresses two types of political outcomes – the public opinion with respect to key policies in the asylum policy domain in the aftermath of the crisis, and the electoral consequences of the crisis. It also draws some conclusions from our findings.

Part I includes four more chapters. The next two chapters present our theoretical and empirical approach. Chapter 2 introduces our theoretical framework. Chapter 3 presents the eight countries we are focusing on and the forty-six episodes that we are studying in detail. In addition, it provides an introduction to our main tool for the analysis of crisis policymaking and crisis politics in these episodes – policy process analysis – and to the complementary methods of analysis, which we apply to the study of electoral outcomes and outcomes in terms of policy-specific public opinion. Chapter 4 introduces the three aspects of the crisis situation – policy heritage, problem pressure, and political pressure. It shows that at the EU level, as a result of the lack of harmonization of minimum standards between member states and of the deficient capacity of some national systems, the asylum policy rested on an “organized hypocrisy” (Krasner Reference Krasner1999), which, predictably, led to the breakdown of this policy in the course of the crisis. At the domestic level, the details of the heritage of the eight member states of our study serve to justify their categorization into four distinct types, as does the country-specific variation in the problem pressure in the crisis situation. Chapter 5, which concludes Part I, turns to the details of the policy episodes. It presents their timing and their substantive content. The association between politicization and the two types of pressure proves to be less close than expected – a finding that is discussed in terms of endogenous political dynamics during the crisis. As for the substantive content of the policy responses, the chapter documents that continuity prevailed – the crisis did not prove to be an opportunity to reform the existing system. Instead, failure to reform at the EU level and retrenchment at the national level characterized the predominant responses.

Against the general background characterizing the crisis situation and the policy responses adopted during the crisis, Parts II and III analyze in detail the actor configurations, conflict structures, and political dynamics of policymaking during the crisis. In these chapters, we combine quantitative characterizations of the various aspects of crisis management by the EU and its member states with qualitative narratives illustrating our more general points with specific cases. This strategy results in a rather long account, but we hope that the reader will appreciate our attempt to make the complex policymaking processes come alive.

Part II includes four chapters. Chapter 6 focuses on the actors and conflict structures at the national level, while Chapter 7 turns to the actors and conflict configurations at the EU level. At the national level, partisan and international conflicts were most common. Mainstream opposition parties emerged as the most important adversaries of national governments, although on occasion they were aided by challenger opposition from the left and, especially, from the right. At the EU level, member states and their key executives played a crucial role in the two-level game of EU crisis management. In terms of conflict configurations, the analysis shows that, at this level, international conflicts prevailed – vertical conflicts between the EU and its member states, transnational conflicts between member states, and externalization conflicts between the EU/member states and third countries.

Chapter 8 zooms in on the relationship between national governments and opposition, whereas Chapter 9 goes one step further and examines the right-wing discourse related to the refugee crisis. Chapter 8 highlights the importance of government composition in explaining the nature of domestic conflict in the refugee crisis. The analysis focuses on two aspects of government composition – fragmentation (as in coalition governments) and ideology. Fragmentation is associated with intragovernmental conflicts, while ideological distance accounts for the intensity of the partisan conflict between government and opposition. Tracing how right-wing actors responded to the crisis, Chapter 9 tries to uncover the elements that allowed these actors to become the main beneficiaries of this crisis (as shown in Chapter 14). The analysis shows that the right-wing parties tried to shift attention away from the initial humanitarian response to the crisis by framing it as a security issue. Concurrently, themes of perversity, jeopardy, and calls to tighten border and asylum policies dominated across the right-wing spectrum.

Part III, on the dynamics of policymaking, starts with Chapter 10, which seeks to uncover the determinants of elite support – broadly understood – behind government policies. The analyses build on the results of Chapter 8 and show how the governments’ opponents systematically responded to each other’s expressed level of support to the government’s initiatives. The results indicate that far from the elite groups closing ranks behind government proposals – as the “rally-around-the-flag” perspective would lead us to expect – they, depending on the context, used the strategic opportunity offered by mounting problem pressure to signal opposition to these proposals and to governments.

Chapters 11 and 12 address the dynamics of policymaking across the levels of the multilevel polity. Chapter 11 takes a closer look at cross-level episodes, which involve an important amount of interaction between the two levels. They include roughly half of the national episodes of our study. This is a remarkably high share, which indicates that national asylum policymaking is taking place in the shadow of EU policymaking. Chapter 12 studies the different ways in which the most important episode of our study – the EU–Turkey agreement – was linked to national policymaking. In this chapter, we ask, based on the EU–Turkey agreement, to what extent the debate on EU policymaking has been domesticated and to what extent the conflict configuration at the EU level is transformed in the national debate about an EU policymaking process.

In Part IV, Chapter 13 looks at the transnational and domestic conflict configurations among the citizen publics of sixteen member states in the aftermath of the refugee crisis. In terms of transnational conflicts, we find the expected opposition between the frontline states (Greece and Italy) and the Visegrad 4 (V4) countries (augmented by eastern European bystander states). At the domestic level, we find the equally expected opposition between nationalists and cosmopolitans that is politically articulated by the radical right and some nationalist-conservative parties on the one side, and by the left and some parties of the mainstream right on the other side. The domestic polarization appears to be more intense than the transnational one.

Chapter 14 examines the electoral repercussions of the refugee crisis in seven member states. As the refugee crisis wanes in memory, it has left some important and lasting marks in the European political landscape. However, the legacy of this crisis was not a wholesale transformation of party systems in some countries, as in the case of the Euro area crisis. Instead, it served as an opportunity for parties mainly from the right, which were able to strategically exploit the crisis for their own electoral purposes. Finally, Chapter 15 summarizes and concludes. To reiterate the general point we are trying to make: The refugee crisis constrained European policymakers, who tried to come to terms with it in ways that induced them to adopt short-term, stop-gap responses and prevented them from coming up with long-term, joint solutions. If, in a certain sense, we confirm the failing-forward assessment of the crisis outcome, we provide a much more specific account of the policymaking process in the crisis that allows us to pinpoint the crisis-specific conditions that, combined, led to this outcome. There is nothing inherent in the integration process that led to the outcome of the crisis. Instead, there is a lot of crisis-specific conditioning that, however, has path-dependent effects that will outlast this specific crisis.

Our approach to the management of the refugee crisis of 2015–16 builds on the polity approach to the EU integration process (Ferrera, Kriesi, and Schelkle Reference Ferrera, Kriesi and Schelkle2023) and attempts to elaborate it in various ways by making use of insights from the grand theories of European integration in combination with concepts and ideas from comparative politics and policy analysis. This has the advantage of tying the supranational and national policymaking during the crisis together within one and the same theoretical and empirical framework. Such a combination allows one to systematically link policymaking at the two levels of the EU polity and to consistently focus on the prevailing conflict configurations at each level individually and jointly at both levels.

The challenge of the refugee crisis focused on bounding, that is, on the internal and external bordering of the EU, with important implications for binding and bonding. In a certain sense, bounding is the precondition for binding and bonding. Without the creation of external closure, it is hard to develop internal feelings of community and to create a center of political authority able to take binding decisions for the entire community. As observed by Schimmelfennig (Reference Schimmelfennig2021), open boundaries not only weaken the community’s capacity to protect itself against outside intervention (e.g., military attack, terrorism, crime), they also tend to weaken internal communal ties. The weakening of bonds of identity, in turn, may undermine the willingness of individuals to contribute to the public good and engage in social sharing. Solidarity may suffer both from the opportunities to exit (e.g., tax evasion, capital flight, and brain drain) and from the opportunities to enter (e.g., when those who enter benefit from the public goods without ever having contributed to them). Weak identity and solidarity undermine the consensus that constitutes the social foundations of democracy (Dahl Reference Dahl1956). By contrast, higher and better-enforced barriers and congruent external boundaries reduce exit and entry opportunities. Schimmelfennig (Reference Schimmelfennig2021: 323): “Locking in actors and resources helps to preserve the cultural homogeneity and identity of the people living inside the territory, strengthen institutions of social sharing, protect the territory from outside threats to security – and thereby build the social foundations of democracy” (Bartolini Reference Bartolini2005: 36–53; Rokkan Reference Roubini and Sachs1974: 49).

Schimmelfennig (Reference Schimmelfennig2021: 324) expects community deficits – such as threats to national identity, rising inequality, or an increase in crime and military vulnerability – to lead to the politicization of boundaries and to rebordering pressures. The question is whether such pressures lead to more internal or external rebordering. Schimmelfennig expects that, for reasons of efficiency, such pressures increase demands for external rather than internal rebordering: Internal rebordering would constrain the benefits of increasing scale in the EU. However, efficiency considerations clash with community considerations. Thus, exogenous shocks, such as the refugee crisis, which render both external and internal boundaries highly salient, tend to activate the underlying integration–demarcation conflict and mobilize partisan contestation at the level of the member states in the name of defending the national community. This mechanism is likely to enhance internal rebordering, even if the member states are closer to each other than to non–member states. As a result of these contradictory influences (see Chapter 1), the refugee crisis has given rise to what Schimmelfennig (Reference Schimmelfennig2021) calls defensive integration, that is, a combination of measures of mainly internal rebordering (the resurrection of barriers between member states or their exit from common policies or the EU altogether) with external rebordering (the creation and guarding of “joint” external EU borders).

In this chapter, we elaborate our argument to account for this outcome. This argument, as we have already pointed out in the introductory chapter, focuses on the policymaking process. In other words, it is the binding component of the polity approach that constitutes the center of our theoretical attention. The crisis led to the politicization of the EU’s boundaries – internal and external, both at the EU level and at the level of the member states. We shall try to explain why it was “defensive integration” rather than “dilutive integration,” full integration, or disintegration that was the chosen outcome of this politicization process.

We divide the presentation of our theoretical framework into three parts: First, we discuss the underlying conflict structure in the EU’s compound polity of nation-states. Then we turn to the politicization of policymaking during the crisis, which is a function of both the specific characteristics of the crisis situation and some key characteristics of the compound polity. Finally, we discuss possible outcomes and their determinants in terms of policy (“defensive integration”) and in terms of polity (the underlying conflicts and their political structuration).

The Underlying Political Conflicts of the Refugee Crisis

“Defensive integration” constitutes a limited, minimum common denominator solution to the refugee crisis, an outcome predicated on the combination of conflicts between member states, between the central EU authorities and member states, and finally within member states. Our analysis will be guided by the key notion that the outcome of the policymaking process fundamentally depends on the political structuring and politicization of the underlying conflicts in the crisis situation and the political dynamics unleashed by it.

We start from the observation that the EU polity has a two-level structure that invites political structuring at both the supranational level of the EU and the national level of the member states. Similar to coming-together federations, in the compound polity of the EU, the conflict structure at the EU level is dominated by the territorial dimension. This dimension produces two lines of conflict: a vertical one, focused on the powers of the polity center vis-à-vis those of the member states, and a horizontal one, revolving around the specific interests of the member states. Throughout the twentieth century, functional conflicts became increasingly important in the nation-states. Thus, territorial structuring was complemented by partisan/ideological structuring. This facilitated central consolidation – the formation of a center capable of speaking directly to “the people” and of advancing system building. In the EU, however, the conflict structure is still dominated by the territorial dimension. Given the strength and direct legitimation of national centers, the territorial channel of representation (via the European Council and Council of Ministers) has remained more important than the corresponding functional channel (via the European Parliament [EP]).

Accordingly, the main political fault lines at the EU level run between member states and between member states and the EU agencies. Only recently have party-based conflicts gained some visibility and salience in the EP arena. Interstate conflict is by definition horizontal and pitches (coalitions of) member states against each other based on material and normative interests. Conflicts between member states have many triggers and targets. In the refugee crisis, they led to the politicization of internal and external boundaries and of national communities. The dividing lines between member states that emerged during the crisis were above all the result of the differential incidence of the crisis: The immediate problem pressure differed from one state to the next depending on the policies in place, the state’s geographical location, and its attractiveness as a destination state for asylum seekers. In our subsequent analyses, we shall distinguish between five types of member states that developed very distinct interests during the crisis: frontline states (Greece and Italy); transit states (Austria and Hungary); and destination states, which are further divided into two subsets – restrictive (France and the UK) and open destination states (Germany and Sweden), depending on their institutional and political openness toward incoming refugees. In addition, there is the category of the bystander states, which were not directly concerned with the crisis but which played an important role in its management nevertheless.

According to the Dublin regulation, border states are responsible for any asylum seeker entering the EU (i.e., the Schengen area) through their territory. In the refugee crisis, this regulation shifted the obligation of accepting and integrating asylum seekers to the southern European frontline states, where they first arrived in the EU. But, as we know, the bulk of the asylum seekers did not stay in these frontline states but continued their journey toward the north of Europe. On their way, they traveled across the transit states such as Hungary and Austria. The classification of Austria as a transit state instead of a destination state might be contested, since Austria received a comparatively large number of asylum seekers, too. But as we shall see, the data point toward Austria having been above all a transit state. Our distinction between two types of destination states is partly informed by the member states’ policies and border control practices during the refugee crisis (more open versus more restrictive) and partly by their prior policy regimes (IMPIC dataset, Helbling Reference Hellmann2016). Germany and Sweden were the principal destination states during the refugee crisis, while countries like France and the UK remained largely untouched by the inflow of asylum seekers. In addition to these four types of member states, there is a fifth category – the bystander states, a category that was not at all directly concerned with the crisis. Among these states were several eastern European countries, as well as countries like Ireland and Portugal.

Based on their common preferences, member states often form transnational coalitions. New intergovernmentalist scholars have provided evidence that national preference formation in the EU has become an inherently transnational process that involves governments of member states (Kassim, Saurugger, and Puetter Reference Kitschelt and McGann2020; Fontan and Saurugger Reference Fontan and Saurugger2020; Kyriazi Reference Lahav and Messina2023). Moreover, under crisis situations where uncertainty and urgency prevail, national preference formation and European-level bargaining tend to become simultaneous processes, with policymakers being involved and negotiating at the national and the EU level at the same time (Crespy and Schramm Reference Crespy and Schramm2021). At this bargaining stage at the European level, transnational coalition formation is a crucial part of policymaking (see Wasserfallen et al. Reference Webber2019 for the Euro area crisis).

The vertical component of the territorial conflict configuration at the EU level refers to the relationship between the EU agencies and the member states. The supranational institutions may be pitted against (coalitions of) member states. Once decisions are made, they become collectively binding and directly enforced by the ECJ (European Court of Justice). A given member state can thus feel dominated by the center when its interests are defeated and undesired policies are implemented. In this case, conflict may indeed take on a vertical drift, turning into opposition against the EU as such. As Mair (Reference Majtényi, Kopper and Susánszky2013) has argued, it is the lack of a government–opposition nexus at the EU level that opens the door to opposition in principle against the polity – to Euroscepticism and to populist reactions against the loss of control at the domestic level.

At the national level, the European integration process has, indeed, given rise to a nationalist reaction to European integration in the party systems, which is part and parcel of a larger conflict opposing cosmopolitans-universalists and nationalists-communitarians that has by now been restructuring domestic European party systems for decades. This new structuring conflict raises fundamental issues of rule and belonging and taps into various sources of conflicts about national identity, sovereignty, and solidarity. Importantly, in addition to the European integration process, migration has become the most important issue that has been politicized by this conflict. The conflict is structurally rooted, opposing the “losers of globalization” or the “left behind” against the “winners of globalization” or the “cosmopolitan elites” (Kriesi et al. Reference Kriesi2006).Footnote ¹ While the mainstream parties have mainly taken the position of the “cosmopolitan elites,” the preferences of the “losers” have above all been articulated by the new challenger parties of the radical right.

The radical right had become a most vocal and visible opposition in the party systems of most northwestern European countries (except for Germany) before the advent of the refugee crisis, while it had not been as present yet in the party systems of southern and eastern Europe. Already before the refugee crisis, the new divide had initiated a break with the period of a permissive consensus, and conflicts over Europe had been transferred from the backrooms of political decision-making to the public sphere. As argued by postfunctionalists, with the increasing importance of this conflict, identity politics have become more important for decision-making at the EU and national levels (Deutschmann et al. Reference Deutschmann, Delhey, Verbalyte and Aplowski2018; Kuhn Reference Kuhn2019). In the refugee crisis, joint action was constrained, and conflicts between member states were reinforced by the domestic politicization of national identities produced by the uneven distribution of crisis pressures within the EU polity. Consistent with the predictions of postfunctionalist theory, the tension between the uneven distribution of costs and benefits of crisis resolution at the international level and the limited scope of community feelings at the national level has made opposition to EU policy proposals more vocal. As a matter of fact, for the radical right, the refugee crisis constituted a golden opportunity to mobilize its nationalist constituencies against the admission and integration of refugees in its own country, including opposition to any joint schemes of international burden-sharing that would have increased the number of refugees to be admitted on the national territory. The decision-makers at both levels of the EU polity were exposed to the political pressure exerted by the radical right at the national level and had to come to terms with it.

However, the conflict structure at the domestic level of the member states cannot be reduced to the conflict between the nationalist radical right and cosmopolitan and pro-European forces. As a matter of fact, we face a much more complex reality domestically. If the ultimate source of partisan conflict is the radical right opposition, the pressure on government more often is likely to come from the mainstream opposition that tries to pin the government into a corner by accusing it either of doing too little in coming to terms with asylum seeker flows or of excesses and inhumane treatment of asylum seekers. As a matter of fact, the electoral success of the radical right parties has prompted mainstream parties to engage in strategic responses to fend off this electoral threat, often by shifting their own programmatic position toward a more restrictive stance on immigration (Abou-Chadi and Krause Reference Abou-Chadi and Krause2018; Abou-Chadi et al. Reference Abou-Chadi, Green-Pedersen and Mortensen2020). In the extreme, such strategic positioning can play out within the government itself in the case of coalitions, and especially grand coalitions (Engler et al. Reference Engler, Bauer-Blaschkowski and Zohlnhöfer2019; Höhmann and Sieberer Reference Hooghe and Marks2020), where coalition partners compete not only with the radical right but also with each other in an effort to send credible signals to voters that their concerns are heard. In this context, center right parties face the dilemma of whether an anti-immigration stance will advantage them in the electoral competition or whether it will play into the hands of the radical right.

The dilemma for the center-left parties is that they are trapped between the principle-based expectations of a left-liberal electorate and the threat of an exodus to the radical right of its traditional working-class voters. As center-left parties shy away from outright humanitarian positions, especially if they are part of a coalition government, nonpartisan actors are likely to enter to fill the void. The most likely candidates for such a role are political actors who are driven less by electoral considerations than by humanitarian and legal principles, such as NGO groups; intellectuals; church actors; and more broadly speaking, civil society actors. At the national level, we expect a more general conflict to emerge between governments and such civil society actors as a result of the parties’ turn to more restrictive policy positions on immigration. In addition to such a domestic conflict, a similar conflict is likely to emerge at the EU level, too, given that the EU and its member states adopted a realist strategy of “defensive integration.” At the EU level, the humanitarian position is also likely to be defended by civil society actors, together with supranational organizations charged with a humanitarian task, such as the UNHCR. Chapters 6 and 7 present the conflict structures at the national and the EU levels, and Chapter 9 provides a closer look at the framing of the refugee crisis by parties from the right.

Finally, in the compound EU polity, the national government is involved in the inter- and transnational conflicts that play out at the EU level. The existence of these parallel conflict lines is perhaps the most important feature of the structural political preconditions at the domestic level during the refugee crisis. Throughout this crisis, governments were involved in a two-level game, with their bargaining power in the European arena conditioned by the type and the intensity of conflict they faced from domestic stakeholders.

Policymaking in the EU Polity under Crisis Conditions

The crisis situation is first of all policy domain specific. It corresponds to the extraordinary moment of urgency and uncertainty that poses an immediate threat to the proper functioning of the policy domain challenged by the crisis, not necessarily to the polity as such. We claim that whether joint action at the EU level is forthcoming depends above all on two sets of factors – the policy-specific institutional context within the compound polity and the characteristics of the crisis situation. The policy-specific institutional context refers to the competence distribution in the policy domain at the moment the crisis intervenes and to the institutionalized decision-making procedures that govern the crisis interventions, while the characteristics of the crisis situation refer to the intensity and distribution of the problem and political pressure, that is, the crisis incidence, among the member states. Although their impact is hard to separate, we shall consider these two sets of factors in two separate sections and begin with the crisis situation. Chapter 4 will present the details of the crisis situation.

Crises Outcomes

Crises often act as “windows of opportunity” for the introduction of new joint solutions. However, in the refugee crisis, the joint presence of intergovernmental crisis management and heightened politicization of national identities has acted as a powerful constraint on the crisis policymaking process, increasing the likelihood of minimum common denominator solutions based on narrowly defined member states’ preferences and making joint policy initiatives harder to achieve (Ferrara and Kriesi Reference Ferrara and Kriesi2021). As we know, the breakdown of the EU’s asylum system in the 2015–16 crisis has mainly triggered the same kind of response as in past crises – namely, a shift of responsibility outward and a reinforcement of border control at the EU level (Guiraudon Reference Guiraudon2018). There was a lack of a push for more integrative solutions. At the national level, we also witness continuity with past legacies: The crisis led to the reintroduction of border controls at the domestic borders and to a further retrenchment of asylum policy across the member states but not to any fundamental changes of policy. In general, the measures introduced during the crisis were consistent with an approach at the national and EU levels that can be traced back for more than two decades (Geddes, Hadj Abdou, and Brumat Reference Genschel and Jachtenfuchs2020). Chapter 5 will present an overview of the policy responses at both the EU and the national level.

Reform of the dysfunctional EU asylum policy proved to be impossible. We expect that two factors mainly contributed to this outcome: on the one hand, the early policy failure (relocation scheme) at the EU level, and on the other hand, the stop-gap externalization solution (EU–Turkey agreement) that was adopted at the peak of the crisis. The early policy failure has undermined mutual trust among the member states and has lastingly poisoned the mutual relationships between them. The successful stop-gap solution has taken off the pressure for more far-reaching reforms. As a result, both capacity and motivation to reform declined, and the can was kicked down the road in a series of non-decision-making episodes. The early policy failure, in turn, has to be interpreted in terms of the underlying master conflict between integration and demarcation: As we shall show, it is the result of the mobilization of national identities by nationalist-conservative governments that deliberately used the issue to radicalize their opposition to joint solutions. Such mobilization processes against joint solutions by member state governments are, of course, most likely in member states that are not directly affected by the crisis. Moreover, we would expect that such mobilization processes occur especially if the potential beneficiaries of joint solutions are widely perceived as undeserving because of earlier domestic policy failures (such as Germany) or as untrustworthy because of endemic structural incapacities (such as Greece).

In spite of the great threat to EU survival perceived in the citizen public (see survey results reported in the previous chapter), no disintegrative dynamic developed at the elite level to threaten the survival of the EU polity during the refugee crisis. It has been argued that not only were national measures and externalization sufficiently effective, but supranational integration among member states was actually not functionally necessary in this crisis (Schimmelfennig Reference Schimmelfennig2022; Genschel and Jachtenfuchs Reference George and Bennett2018). This kind of argument downplays the dysfunctionality of the existing system of European asylum policy and also neglects the indirect consequences of the unresolved issues of asylum policy for subsequent crises in the EU. Given the importance of the integration–demarcation conflict in the European party systems, asylum policy remains a potent means for electoral mobilization on the left and on the right. The large opposition to immigration in some member states is bound to constrain the options available to policymakers because it is likely to constitute a major obstacle to joint solutions. Chapters 13 and 14 address the explosive potential of migration-related issues among the voters, as well as the electoral consequences of the refugee crisis.

More specifically, we should not only consider the consequences of the crisis for policymaking at the EU level. The problem pressure in the destination states may be such that it constitutes a fundamental threat to the survival of political regimes, governments, political parties, and their leaders. We would argue that the refugee crisis provided the crucial impetus for the emergence of the illiberal democracies in Hungary and Poland: Political entrepreneurs in both countries seized the opportunity to transform their political regimes and thereby created the rule-of-law crisis. In terms of threats to governments, parties, and their leaders, in the Euro area crisis, this concerned mainly the southern European member states (see Hutter and Kriesi Reference Ignazi2019). In the refugee crisis, we expect that this danger loomed large above all in the northwestern European destination states, and especially in Germany, where a grand coalition dominated by a center right party (the CDU-CSU) was held responsible for the large inflow of asylum seekers. As has been already pointed out, the German government, and especially the dominant center right party, was caught by surprise and found itself trapped by the incoming flow of asylum seekers. It had to adopt policies that were obviously unpopular with large parts of the dominant party’s electorate. We expect leaders of governments who are trapped in this way to look for the EU to come to their rescue by adopting EU measures that alleviate the pressure they are facing. Whether such support will be forthcoming depends, as we have already pointed out, on the distribution of the pressure among the member states and on the support such a government finds among the EU authorities.

However, contrary to the Euro area crisis, we do not expect the refugee crisis to have triggered a wholesale transformation of party systems in some member states. Except for some open destination states, we suggest that the refugee crisis provided much more room for strategic choices by parties, since it was cumulative and expected and, overall, posed less of a threat to the individual governments. The parties could anticipate the potential political impact of the issue and either shield against it or try to exploit it more or less successfully, depending on the case at hand. For individual parties, the crisis provided opportunities to benefit from the increased salience attributed to the immigration issue by the mainstream media and European electorates. We expect that right-wing actors who were persistent on their anti-immigration message and “owned” the issue enjoyed electoral gains at the expense of their proximate party families and the left. We do not suggest, however, that the drivers of the politicization and those who reaped benefits from this right-wing drift were necessarily the same in every country. Instead, we expect the beneficiaries to vary depending on the country-specific context of party competition.

Conclusion

To summarize our main expectations: We expect the management of the refugee crisis to be heavily shaped by the underlying political conflicts in the compound EU polity of nation-states, by the crisis situation that prevailed as a result of the policy-specific heritage, and by the combination of problem and political pressures at both levels of this polity in interaction with a set of particular characteristics of the EU polity. The vertical and horizontal territorial conflicts that are typical of this compound polity are expected to have been exacerbated by two aspects of the crisis situation in particular – the limited number of competences of the EU in the policy domain of asylum policy and the asymmetrical incidence of the refugee crisis among the member states. Finally, we formulated some expectations with regard to the crisis outcomes at the two levels of the polity. At both levels, previous assessments argue for more continuity than change – in terms of both policy and conflict structures – and limited spillovers from policy to polity change. However, we argue that the implications for the maintenance of the compound polity created by the way the crisis was managed may have been more problematic than meets the eye at first sight.

Our approach is compatible with the “failing-forward” framework as far as the outcome of the crisis is concerned. But this framework lacks concepts for the analysis of the policymaking process that we provide. At the same time, our framework is also compatible with the neofunctionalist approach as far as the importance of spillover processes is concerned. Contrary to neofunctionalism, we insist, however, that these spillover processes do not necessarily contribute to further integration but might, instead, undermine such integration by creating externalities for fellow member states that induce the latter to adopt internal rebordering measures and to create “circles of bonding” that may prove to be highly divisive for the future of the EU polity. Our framework also borrows from intergovernmentalism, whether in its liberal or its renewed version. The refugee crisis was primarily managed by intergovernmental coordination, in close interaction with the EU authorities, most notably with the Commission. The crisis-induced power relations between member states are as expected by liberal intergovernmentalism, and the details of executive decision-making are precisely in line with the expectations of new intergovernmentalism. However, contrary to liberal intergovernmentalism, we do not consider interest groups to be of prime importance for the management of a crisis like the refugee crisis. In this crisis, where identity issues loom large and are activated by partisan contestation in the member states, the political pressure exerted by party competition is much more important, in line with the expectations of postfunctionalism.

Introduction

In Chapter 2, we described our theoretical approach for studying the refugee crisis in a multilevel polity. We have also already introduced the outline of our empirical design. In this chapter, we describe the main elements of this empirical design, including our case selection strategies and the types of data used.

In the first part of the chapter, we describe our case selection, essentially delimiting the empirical scope of our study. As our theoretical approach is based on the perspective of the EU as a multilevel polity involving asymmetrical and interdependent relations between member states, our empirical universe consists of the unfolding of the crisis at both the EU level and the level of the member states. Within this empirical universe, our case selection strategy involves two steps. In the first step, taking into consideration the variation in policy heritage of European countries in the immigration domain, the immediate crisis situation they were facing, but also their centrality in the unfolding of the refugee crisis, we classify EU member states into four main types: frontline, transit, open destination, or closed destination states. In addition, we consider a fifth type, bystander states, which we, however, do not study in detail.

In the second step, within these selected countries, but also at the EU level, we study the crisis by breaking it down into a set of key policymaking episodes, which are triggered by salient policy proposals. Some of the policies we have chosen are legislative acts, such as reforms to the countries’ asylum systems, while others are administrative decisions and novel practices by state institutions, such as the reimposition of border controls in a period of heightened problem pressure. In the next section, we describe our episode selection strategy based on systematic media and secondary source analysis.

In the second part of this chapter, we focus on the empirical approaches we employ for studying the different stages and elements of the crisis. As our theoretical framework involves an ambitious design that aims to study the interplay of both supply-side and demand-side dynamics, our book draws upon a variety of original datasets involving various methods of data collection. While many of these methods are mixed throughout the forthcoming chapters depending on the elements of the crisis on which we zoom in (e.g., the crisis situation, policymaking during the crisis, political competition dynamics), the central dataset upon which the book is based uses policy process analysis (PPA), a method that relies on the systematic coding of media data for capturing the policymaking and politics surrounding policy debates. Drawing upon political claims analysis (PCA) (Koopmans and Statham Reference Koopmans and Statham1999), our original PPA dataset incorporates into a single framework information about all the major components of an empirically delimited policy episode in a country of interest. PPA is complemented with core-sentence analysis (CSA) for studying political competition dynamics in election campaigns, survey data for capturing public opinion on immigration, and speech analysis for studying rhetorical devices employed by key right and radical right actors during the crisis. In the following text, we detail the methodology behind these empirical approaches, and we point to the various parts of the book where they are employed.

Selection of Countries

Our theoretical approach is based on the perspective of the EU as a multilevel polity involving both dynamics at the EU level and asymmetrical and interdependent relations between member states, and domestic dynamics that shape the available policy options and outputs. Therefore, we study how the refugee crisis is unfolding in its various aspects at both the EU level, and in the various EU member states. By complementing a within-country perspective with an EU level perspective, we aim to provide a comprehensive account of the European refugee crisis’s origins, ongoing developments, and consequences.

For breaking down the variety of EU member states and the role they played in the crisis, we categorize these states into the four main types we already mentioned: frontline, transit, open destination, and closed destination. The fifth type, bystander states, was hardly affected by the crisis and therefore played a marginal role in its unfolding. While not studied in depth, we do mention these bystander states when zooming out on broader aspects such as the salience of the immigration issue in the public across member states or when they get involved in any political dynamics in our countries of interest or at the EU level. This country typology is guided by several criteria related to the policy heritage in the immigration domain of these countries; the immediate crisis situation they were faced with; but also, more generally, the migration trajectories in Europe. We selected two countries per type based on their centrality in the unfolding of the crisis: Greece and Italy as frontline states, Austria and Hungary as transit states, France and the UK as closed destination states, and Germany and Sweden as open destination states. It is in these eight countries and at the EU level that we study the specifics of policymaking and political dynamics during the crisis, while in the rest of the member states we adopt a more birds-eye view. In the following text, we describe our two main classification criteria: the crisis situation and the asylum policy heritage.

The first criterion on which we base our classification is the crisis situation. In this respect, the incidence of the crisis across EU member states was asymmetric, with countries experiencing different types and levels of problem pressure with regard to the number of entries and asylum requests. These asymmetries mainly result from the countries’ geographical location and their attractiveness as destination states for asylum seekers. Countries that are geographic points of entry into the EU are frontline states, countries that are desirable destinations for migrants are destination states, while countries situated along migration trajectories are transit states.

The second criteria behind our classification refers to the immigration policy heritage and the nature of the prevailing asylum regime. First, central to the asylum policy is the Dublin principle, according to which countries that are the first point of arrival for an asylum applicant are responsible for processing their claim. This principle shifted the burden of accepting and integrating refugees to the EU border states, which became the frontline states in the refugee crisis. Second, while the Common European Asylum System (CEAS) aimed at setting common minimum standard for asylum across EU member states,Footnote ¹ asylum regimes remain largely unharmonized (Kriesi, Altiparmakis, Bojar, and Oana Reference Kriesi, Ferrera and Schelkle2021; Scipioni Reference Siegel2018; Niemann and Zaun Reference Zincone, Zincone, Penninx and Borkert2018). Differences in these asylum regimes existed even before the crisis struck, as will become more apparent in Chapter 4. In order to obtain an idea of how the national asylum regimes actually worked in the past, we propose examining the rejection rate of asylum seekers prior to the crisis (2010–14). For our eight countries, the first column in Table 3.1 presents these rates for asylum seekers from the five countries (Syria, Afghanistan, Iraq, Pakistan, and Nigeria) that due to either political instability or sheer population size presented national asylum authorities with the greatest burden during the refugee crisis. We notice here that there are wide differences among the countries that are regularly considered as destination states for migrants. While Sweden and Germany had been open destination states for asylum seekers prior to the crisis, having rather low rejection rates, France and the UK were already more closed before the crisis. Consequently, we split the group of destination states into two categories: open and closed. The differences between these groups of countries will be studied in more depth and will become more apparent in Chapter 4, where we further inquire into their crisis situation in terms of policy heritage, political pressure, and problem pressure. While there is also some variation among the frontline and transit states, we do not divide them any further but do take into account these differences when studying individual countries.

Moreover, the capacity of national asylum systems to deal with asylum requests also varies considerably between member states. Unfortunately, there are no longitudinal data available for this aspect, but the figures in the second column of Table 3.1 provide a snapshot of the financial resources available for the determining authorities. The ordering of countries is closely aligned with the rejection rates, except that the UK has somewhat more resources and Austria a lot less resources than the rejection rates would lead us to expect. As these numbers suggest, the Greek, Hungarian, and French systems fall far short of what would have been required for proper functioning. The Greek asylum system had already been judged to be dysfunctional by the European Court of Human Rights (ECHR) and the European Court of Justice (ECJ) as of 2011, and in 2012, the UNHCR arrived at the same assessment for the Hungarian asylum system (Trauner Reference Trauner and Ripoll Servent2016: 314). In other words, the national asylum systems of precisely those countries that were supposed to take care of the massive refugee inflows in the refugee crisis were least prepared to do so. Admittedly, annual budgetary appropriations are only one aspect of how effectively a given country’s asylum system functions. However, in the context of a sudden spike in requests, the available resources of the system are an important indicator of its capacity to satisfy the country’s CEAS obligations. These resources further reinforce our split of the destination states into an open and a closed type, while still pointing to significant differences in the asylum regimes of the other types of countries that will be studied in the upcoming chapters.

Selection of Episodes

Within these selected countries, but also at the EU level, we study the refugee crisis by breaking it down into a set of key policymaking episodes, which are triggered by salient policy proposals. Some of the policies we have chosen are legislative acts, such as reforms to the countries’ asylum systems, while others are administrative decisions and novel practices by state institutions, such as the reimposition of border controls in a heightened period of problem pressure. A policy episode in our framework comprises the whole policy debate surrounding these specific policy proposals that governments put forward, from the moment the proposal enters the public debate to the moment the proposal is implemented and/or discussion surrounding it is no longer salient.

Our approach of focusing on specific policymaking episodes, rather than studying the refugee crisis as a whole, brings several advantages to the analysis. First, adopting an episode-based strategy enables the systematic comparison of our countries of interest by allowing us to compare policies of a similar type across countries (e.g., asylum reforms, border control). Second, by breaking down the crisis into policy proposals and by focusing on periods of heightened salience of the immigration issue in the public debate, we can limit the resources required for an in-depth study of all the actors involved, the actions they undertake, the issues they address, and their interactions in a systematic manner. Lastly, our episode-based strategy does not preclude, but rather complements, the strategy of studying the crisis as a whole. Different aspects of the crisis are better suited to be studied by one or the other strategy; for example, general trends in salience are better studied throughout the crisis as a whole, whereas policymaking and political dynamics surrounding specific policy proposals are preferably studied in a bounded episode. Accordingly, we adopt an encompassing analytical strategy when looking at problem and political pressure (Chapter 4), at conflict configurations on the demand side (Chapter 13), and at electoral outcomes (Chapter 14). Conversely, we focus on episodes when studying the variety of policy responses to the crisis (Chapter 5), the actors and conflict structures in policymaking (Chapters 6–9), and the dynamics of policymaking (Chapters 10–12).

To systematically select these episodes, we have resorted to a two-step strategy. In the first step, we analyzed a variety of international press sources using a broad timeframe (starting in 2013 and ending in 2020) covering the crisis so as to make sure that we capture policy processes starting before or continuing after the peak of the crisis in 2015–16. We used the international press at this initial stage based on the idea that the proposals that make it into the international news are publicly most relevant and impactful. We constructed a corpus of news articles based on general migration-related keyword and performed an initial round of in-house, manual coding for identifying policy proposals. Based on the number of times a selected proposal appeared in the media, we delimited an initial set of particularly relevant proposals. In the second step, we cross-validated this initial set of episodes by using secondary sources (various publications of think-tanks and NGOs such as the Migration Policy Institute, European Migration Network, and Asylum Information Database) and by performing similar searches in the national press with the aid of native-language-speaking coders. Finally, we ended up with five key policy episodes in each of the eight countries and six policy episodes at the EU level. Additionally, to obtain a better grasp on the interactions between the EU and the domestic levels, we also studied one of the salient EU-level policy proposals – the EU–Turkey Deal – and the debate surrounding it in four member states representing our four country types: Greece, Hungary, Germany, and the UK.

A similar process was adopted for establishing the more specific timeline of these episodes, with a few important additional steps. The initial episode timeline that was established based on the two steps described above was further refined in close collaboration with a team of native-language speakers who helped us identify episode-specific keywords that were iteratively tested and applied to national news sources. Episode timelines are, therefore, exclusively based on the characteristics of the individual episodes. We have not harmonized their duration, as we are interested in how the episodes unfolded in their entirety. The process of timeline selection is further described in the following section on policy process analysis, where we also describe the data we have collected on these episodes.

Table 3.2 summarizes the episodes we have coded via the short labels we assigned to them together with their timeline. We can classify these episodes into two main types according to their substantive scope: (1) asylum-related policy reforms (including rules of burden sharing between member states, the retrenchment of asylum law, and the introduction of integration laws and laws on return in the member states) and (2) external border control measures (including the externalization of refugee protection). Not only does the substantive focus of policymaking vary across member states and phases of the crisis (as we show in Chapter 4), but it also plays a role in how political dynamics develop in these countries. For example, in Chapter 5, we show that the dominant types of actor conflict vary by episode type: societal conflicts are prevalent across all episode types, while intragovernmental conflict is prevalent mostly in border-related episodes.

Data Collection and Analysis

Policy Process Analysis (PPA)

The main method we rely on for studying the political dynamics during the crisis and the variety of policy responses across our selected EU and country episodes is policy process analysis (PPA) (Bojar et al. Reference Bojar, Kyriazi, Oana and Truchlewski2021a). PPA intends to be a comprehensive method for the data collection and analysis of policymaking debates. As such, PPA aims at capturing the public face of policymaking, that is, the subset of actions in a policymaking process that are presented to the general public through the mass media. The method relies on analyzing media data based on systematic hand-coding of indicators related to the actors involved in the policy debate, the forms of action they engage in, the arena where the actions take place, the issues addressed, and the frames used to address these issues. The resulting dataset allows for the construction of more aggregate indicators at different levels of analysis (at the episode level, at the actor level, at different time units) for studying the policymaking debate and the political dynamics surrounding it from multiple angles, both statically and over time.

In its design, PPA is a specific form of political claims analysis (PCA) (see Koopmans and Statham Reference Koopmans and Statham1999) and also incorporates elements from other methods previously employed to study protest events (protest event analysis [PEA]) (see Hutter Reference Hutter and Grande2014; Kriesi et al. Reference Kriesi2020) or contentious politics (contentious episodes analysis [CEA]) (see Kriesi, Hutter, and Bojar Reference Bojar2019; Bojar et al. Reference Bojar, Kyriazi, Oana and Truchlewski2023) by making use of the systematic coding of media data. At its core, PPA is also an event-based methodology that focuses on identifying distinct actions undertaken by a variety of actors addressing particular issues and how they unfold over time. However, while PEA and CEA are usually limited to identifying either actions in the form of protest events or actions initiated by a limited set of actors (government versus challengers) to study mostly contentious politics, PPA enlarges the empirical scope to the study of entire policy debates.

Its encompassing scope and event-based focus make PPA a specific form of political claims analysis (PCA) (Koopmans and Statham Reference Koopmans and Statham1999). Starting from a critique of protest event and political discourse newspaper analysis as being too “protest-centric” and focused primarily on nonroutine protest actions, PCA argues for the need to include events that occur outside the context of reported protest but that are important for understanding conflict. As such, PCA extends event coding to including actions that take on institutional forms, such as legal actions, and including public and institutional actors beyond social movements. Our PPA methodology takes this critique seriously by also enlarging the empirical scope of the analysis to include both institutional and noninstitutional actions and actors. Where PPA departs from PCA is in its focus. PCA originated as a method primarily focused on studying the demand side of politics by taking as its starting point claims making (“strategic demands made by collective actors within a specific contested issue field”) (Koopmans and Statham Reference Koopmans and Statham1999: 206) and attempting to enlarge the study of contentious politics and placing it into its wider context. By contrast, our PPA methodology is essentially supply-side focused by having as a starting point policymaking processes and specific policy debates while attempting to place these in their wider political context. It is this supply-side focus that drives our strategy of analyzing selected empirically delimited policy episodes rather than general contested issue fields and studying the policy debate surrounding them in a systematic fashion.

In its supply-side focus, PPA also comes close to another approach to the study of policymaking processes – the comparative policy agendas (CPA) project (Baumgartner, Green-Pedersen, and Jones Reference Jones and Baumgartner2006). However, rather than focusing particularly on the agenda-setting phase of policymaking as the CPA does, PPA systematically incorporates into a single framework information about all the major components of a policy debate. Therefore, as further detailed in the section below, PPA documents actions ranging from formal steps in the policymaking process to administrative and nonstate actions but also protest events and even single verbal claims. Due to its goal of studying both the politics and the policymaking surrounding a particular episode, the actors documented in PPA are not restricted to solely governments; rather, they include all actors involved in the debate – political parties, civil society actors, supranational actors, and third-country actors.

Similar to CEA, PPA attempts to capture the middle ground between a qualitative narrative approach and a quantitative approach to describing the chronology of policy processes (see Kriesi et al. Reference Kriesi2019). By including extensive string descriptions of each action, PPA provides a rich body of qualitative information on the politics and policymaking surrounding policy debates. At the same time, by coding specific action characteristics, it allows for the construction of systematic, comparative indicators at various levels: countries, episodes, and actor types. In its qualitative inclination, one could think of this approach as being related to process tracing in that we seek to document all the various chains of actions involved in a policy debate as they unfold over time in a systematic fashion. However, process-tracing is aimed mainly at single-case inferences about the intervening causal process, that is, on the causal mechanisms that link a given cause to an outcome in a single case (George and Bennett Reference Georgiou and Zaborowski2005: 206–207; Beach and Pedersen Reference Beach and Pedersen2016: 4–5). In contrast, our method is aimed at combining such single-case inferences with cross-case inferences, making it essentially a comparative method. We therefore analyze the refugee crisis by comparing the variety of countries and episodes based on a combination of both qualitative evidence on the sequences of events in the form of descriptive narratives and systematic, quantitative indicators measuring relevant characteristics of these episodes (e.g., politicization and conflict intensity, as further detailed in the next section). In doing so, we aim to study within-country policy processes and how they evolve from problem pressure via domestic actor constellations and conflicts all the way to policy outcomes but also to compare such policy processes in a systematic fashion.

To sum up, PPA is well suited for our research goals in studying the refugee crisis for two main reasons. First, its broad empirical scope allows us to focus on a wide variety of actions and actors in systematically reconstructing the various policy debates both at the EU level and at the level of the member states. We can therefore use PPA for identifying the policymaking repertoires employed at these different levels, as well as for systematically studying the wide variety of actors involved in these debates, their configurations, and their discursive strategies. Second, by aiming at the middle ground between quantitative and qualitative approaches, PPA allows us to combine systematic, comparative indicators of the various aspects of politics and policymaking surrounding policy debates with the reconstruction of the narrative chronology of these policy debates by the use of a rich body of qualitative evidence.

Having set up our empirical universe as bounded segments of the policy debate that take the form of distinct policy episodes embedded in the broader context of the 2015–16 refugee crisis, the first step in constructing our PPA dataset was defining and gathering the media corpus to be analyzed. Therefore, the first decision we were confronted with was source selection, that is, the selection of news media to be studied. Depending on the level of policymaking, we selected either international news sources (for the EU level) or national news sources (for the level of the member states). We used the news aggregator platform Factiva for document retrieval, as it provided us access to a large number of media outlets, which allowed for systematic multicountry comparison together with transparent and replicable selection criteria on the source.

Following good practice standards in working with media data from methods such as protest event analysis (Hutter Reference Hutter and Grande2014; Kriesi et al. Reference Kriesi2019), we also tried to engage with issues of selection bias (e.g., Earl et al. Reference Earl, Martin, McCarthy and Soule2004; Ortiz et al. Reference Pardos-Prado2006), that is, with the biases associated with news source selection and their coverage of debates, actions, or events. In order to mitigate such biases, we adopted several strategies. First, we relied on a wide variety of media sources, rather than a single source, in order to be able to capture as many aspects of the policy debates as possible. Second, as just mentioned, in order to mitigate biases related to newsworthiness and proximity, we selected news sources that are proximate to the level of analysis: For EU debates, we focused on large news agencies (Agence France Presse, Associated Press, Reuters, Financial Times, Euronews, ANSA, BBC, MTI), while for national debates, we relied on national media. Third, in order to mitigate biases related to the political motives of the various sources and their potential impact on news coverage, we selected news sources on different sides of the political spectrum. Consequently, for each of the eight selected sources, we selected one major newspaper left of center and one right of center in terms of ideological leaning (with some minor exceptions related to data availability).

After selecting the news sources, the second decision related to corpus construction consisted of the identification of the keywords used for the retrieval of articles related to a particular episode. One of the main considerations at this step was achieving a balanced relevance ratio – the ratio between false positives (irrelevant articles that the keyword combination retrieved as positive hits) and false negatives (relevant articles that the keyword combination filtered out as negative hits). Since our data is manually coded, we aimed for a relatively slim but robust corpus. That is, our corpus needed to be manageable in terms of the number of articles identified so as to not make the coding process too cumbersome and resource intensive, but it still allowed us to capture the full range of actions in a given policy episode without missing relevant articles filtered out by a too restrictive keyword combination.

In practical terms, the selection of keywords related to each of the EU and country episodes was performed by the authors of the book in close collaboration with a team of native-language-speaking coders (mostly comprised of political science PhD students who were also knowledgeable about the subject at hand – the refugee crisis). At this stage, we took advantage of the capabilities of the news aggregator Factiva, which allowed us to construct complex search strings using Boolean algebra and its standard logical operators. For each episode, we chose an initial set of episode-specific keywords based on secondary sources (policy reports, secondary scientific literature, etc.) and initial search queries in the national press. We then further refined this initial keyword selection through an iterative process of going back and forth between the selection and the corpus obtained. We selected those keyword combinations that passed the initial reading of the selected articles and achieved a satisfactory balance between the size of the corpus and the number of events filtered out.

After having constructed the corpus, the last step in the PPA coding process consisted of action coding. As already mentioned, PPA is an event-based methodology and hence the unit of observation at the level at which the data is collected is an action. An action in our framework is defined as “an act, or a claim by an actor with a prominent role in the political world that has a direct or indirect relevance for the policy debate” (Bojar et al. Reference Bojar, Kyriazi, Oana and Truchlewski2023). Therefore, within our framework, actions can be steps in the policymaking process, verbal claims, episode-related protest events, and other types of actions that we outline in the coding scheme below. This definition is rather open-ended because the relevance of an action is contingent on the specificities of the actual policy debate at hand.

Note that while the lowest level of observation is an action, the unit of analysis at which we draw conclusions can be pitched at any level of aggregation (actor types, issue categories, entire episodes, types of countries, etc.) depending on the research question, as will become apparent in the following chapters. In order to measure the various features of actions, action coding is based on a common core of variables that are coded for each of the actions in each episode: the arena where the action takes place, its (procedural) form, its (substantive) type of engagement with the policy, its overall direction vis-à-vis the policy, its direction vis-à-vis target actors, the organizational characteristics of the actor undertaking them, the organizational characteristics of the target actor, the issues it engages with, and the normative frames used by actors to present their positions to the public (Bojar et al. Reference Bojar, Kyriazi, Oana and Truchlewski2023).

Based on initial trial rounds of action coding, we refined these major characteristics of an action with specific categories relevant for the refugee crisis. This resulted in a detailed codebook with hierarchically organized categories at various levels of specificity. The codebook was complemented by a dedicated coding spreadsheet that was provided to the coders to make the data collection process as systematic and comparable throughout country episodes as possible. At the end of the coding process, at the national level, our team identified 6,338 codable actions for the 40 episodes, yielding 157 actions per episode on average. However, there is considerable variation in how eventful the individual episodes are, ranging from 48 actions in the Residence Permits episode in Sweden to no less than 363 actions during the quota referendum in Hungary. In fact, Hungary has proven to be the most eventful of our eight countries with 1,204 actions, followed by Greece with a total of 1,086 actions. At the EU level, we have coded 1,257 actions in the six episodes, with the EU–Turkey Deal being the most eventful one (437 actions), while the EU–Libya episode had the lowest number of actions (62). These two datasets are complemented by the EU–Turkey Deal episode and the debate surrounding it in four member states containing an additional 1,138 actions. In the following text, we describe how each action characteristic was implemented in our data collection effort.

The first set of characteristics for each action that we have identified is related to the arena where it takes place. Arena choice is an important aspect of the policy debate because it can shape the type of actors that gain access to policymaking, the size and type of audiences that participants can address, and the type of policy options on the table as a function of the gate-keeping role of agenda setters (Timmermans Reference Timmermans2001; Lowi and Nicholson Reference Lowi and Nicholson2009; Princen Reference Princen2011). Arenas are also important because procedural forms of action depend on where they take place. We identify nine types of arenas in our codebook (see Figure 3.1) varying from decision-making institutional arenas such as the national governments to less institutionalized arena types such as protest or society more generally. Furthermore, for each of these nine arenas, we also identified specific forms of the action. For some of the arenas, the set of action forms was based upon long-standing traditions in the pertinent literature, such as the set of action repertoires in the protest arena (Traugott Reference Trauner1995; Della Porta Reference Della Porta, Snow, della Porta, Klandermans and McAdam2013), while for others, such as the media arena, it was decided inductively based on our trial coding.

Figure 3.1 Policy action arenas at the national and EU levels

Figure 3.1 shows that most of the actions in our dataset at both the EU level and the national level take place in the media area (these usually come in the form of statements, press conferences, interviews, etc.). Unsurprisingly, the next most prominent arenas in our dataset at the national level are national governments and parliaments, while European institutions and the cross-national arena prevail at the EU level. Beyond these arenas, our data collection effort also captured actions taking place in the electoral arena, in the protest arena, and at the level of society more generally, thereby providing us with a multifaceted picture of the policy debate not only in venues mostly dedicated to supply-side actors but also in venues where demand-side actors such as civil society organizations most often operate.

After settling the “where” of the action, the next set of characteristics refers to the type of action that actors undertake with reference to the policy proposal and to other actors involved in the policymaking process. In this respect, we included a wide action repertoire, distinguishing between policymaking steps, policy claims, administrative state actions, and nonstate actions. It is at this level that our PPA methodology is distinguished from other methods dedicated to analyzing policymaking processes such as the comparative policy agendas (CPA) project (Baumgartner et al. Reference Baumgartner, Green-Pedersen and Jones2006). Rather than only studying formal steps in the policymaking process, our dataset also includes verbal claims and statements made by a variety of actors in the policy process. In fact, as Figure 3.2 reveals, the most prominent policy action forms at both the national and EU level are precisely policy claims (these usually include actions such as full verbal support/opposition of the policy, clarifications, apologies, and verbal commitments to further action).

Figure 3.2 Policy action types at the national and EU levels

Distinguishing between policy claims and formal policy steps provides us with a more nuanced picture of how the policy debates unfolded, as these substantive types of action most often also indicate whether the action implies a broad level of agreement or disagreement with the underlying policy on the table. In addition to the substantive action types, we also use a general policy direction code (positive, negative, or neutral) as an indicator of the actor’s position regarding the issue at stake. Finally, since in most of the episodes we follow up on the implementation of the policy in question and also include actions undertaken by nonstate actors (such as policy evaluations, NGOs involved in the implementation of a particular policy), we also consider administrative actions performed by state and nonstate actions.

Beyond characteristics of the action itself, the actors involved in a particular policy debate are of particular interest to us, as is shown in Part II of our book. By studying actors and the actions they undertake, we are able to analyze conflict structures and dynamics of coalition formation at both the national and EU levels, which is crucial in the negotiation stages of these policy episodes. Note that at this stage, we try to identify not only which actors undertake a particular action but also whether that particular action is targeted at other actor(s) in the policy debate. We therefore take into account two types of actions: monadic actions, which only have an initiator actor who addresses an issue, and dyadic actions, which have initiator actors who address not only an issue but also a target actor. For the dyadic actions, similar to the policy direction code, we introduce an actor direction code (negative, positive, or neutral) that captures the actor’s relational position vis-à-vis the target actor regardless of how they relate to the policy as such.

The actor characteristics that we identify in our PPA data collection are organized hierarchically at four levels: their nationalities; their broad institutional affiliations (such as the national government); their narrow institutional affiliation (a particular ministry); and, in the case of individual actors, their position within the institution’s hierarchy (executive, subexecutive, or lower rank). This hierarchical organization allows us to study actor configurations at various levels of specificity, identify both domestic coalitions and cross-national coalitions, as well as capture dominant decision-making modes such as executive decision-making or partisan contestation.

In Figure 3.3, we present the share of actors involved in policy debates at the EU and national levels according to their broad institutional affiliations. National governments are the central actors in our domestic policy debates, with more than 30 percent of the actions being initiated by them. In contrast, inter- and supranational actors are the initiators of most actions (more than 80 percent) at the EU level. Despite these two categories unsurprisingly taking center stage at their respective levels, we can see that other national institutions (e.g., regional authorities), political parties both in government and in opposition, as well as interest groups and civil society actors have nontrivial shares of actions, especially at the domestic level.

Figure 3.3 Initiator actor types at the national and EU levels

Although in all the episodes we select actions relating to a particular policy proposal, most of the time the debates tend to revolve around more than one issue. Many of these proposals are in fact policy packages containing multiple issues that collectively make up the policy debate. Moreover, many actions do not directly relate to the policy but are important nevertheless because they have the potential to influence the future course of the debate. Differentiating between specific issues allows us to capture the more fine-grained thematic crisis responses, which are discussed in more depth in Chapter 5. We therefore introduce a set of issue codes organized in such a way as to capture a broad categorization of migration-related policy areas as reflected in the organization of asylum and migration policies in the EU member states.

Figure 3.4 presents the broad categorization of issues our episodes involve.Footnote ² We can see that at both the EU and the national level, asylum issues and border control issues dominate the agenda. As some actions are directed toward a whole policy package (i.e., an episode), we introduce this as a specific issue. As we are interested also in some actions that do not directly relate to migration policy but are relevant for the policy debate at hand, we have complemented this with an “others” category to capture impactful actions and/or events in our episodes, such as issues pertaining to diplomatic relations between countries or humanitarian tragedies.

Figure 3.4 Issues at the national and EU levels

Finally, an important characteristic we include in our study relates to the discursive frames actors use. This essentially refers to the ways in which actors justify their action or interpret the political problem at hand. Such discursive framing is important because it can shape other actors’ attitudes and behavior (Koopmans and Statham Reference Koopmans and Statham1999; Rucht and Neidhardt Reference Rutz1999). The frames employed give us an overview of the communication strategies employed by the various actors involved in the policy debate and can be used to identify discursive coalitions in the political process.

Figure 3.5 presents the main frames identified in the refugee crisis grouped into four major categories. These broad categories were constructed inductively based on several rounds of trial coding and were further adapted though the data collection process. We observe that while humanitarian and democratic frames appear to be important at both the EU and national levels, there is still a wide discrepancy between the discursive strategies that actors employ at the two levels regarding other framing categories. At the national level, sovereignty, security, and identity frames dominate the discourse, but at the EU level, international solidarity arguments take a more central place. Chapter 9 will further delve into the issue of framing, looking at the role of discursive coalitions within the refugee crisis.

Figure 3.5 Frames at the national and EU levels

If above we have presented general descriptions of the major characteristics of actions captured by our PPA dataset, these characteristics also stand behind the formation of systematic, comparative indicators used across our country episodes. One example of such an indicator that is used extensively in the following chapters is politicization (De Wilde Reference De Wilde2011; Hooghe and Marks Reference Hooghe, Marks, Jones, Menon and Weatherill2012; Hutter and Kriesi Reference Bremer, Schulte-Cloos, Kriesi and Hutter2019b). Politicization allows us to capture the expansion of the scope of conflict within the political system (Hutter and Grande Reference Hutter and Kriesi2014: 1003). We conceive of politicization as a multifaceted concept involving a dimension of salience (the number of actions occurring in a particular episode in a particular time frame) and a dimension of polarization (the share of positive and negative actions in that timeframe), both of which are captured by our PPA dataset.Footnote ³ Another indicator based on our PPA dataset that we use in Chapter 6 is conflict intensity, which is designed to capture the conflictual nature of the policy actions undertaken by actors. We define conflict intensity as a combination of the type of policy action that the actor undertakes and the direction of their actions vis-à-vis their target actors. While politicization allows us to capture the expansion of the scope of conflict, conflict intensity allows us to capture its nature, as some policy actions in our dataset are more conflictual than others (e.g., threats and denigrating opponents are more conflictual than simply proposing a new policy or negotiating) and as actions can be negative, positive, or neutral toward target actors. Accordingly, each action in our dataset is assigned an ordinal conflict intensity score on a 5-point scale based on a classification of policy actions and direction toward the target actor.

Conclusion

In this chapter, we have introduced the main building blocks of our empirical design for studying the refugee crisis. In order to meet our ambitious goals of studying the refugee crisis in all its stages, taking into account both its policymaking and political developments, involving both supply-side and demand-side dynamics, we have set up an equally ambitious empirical strategy.

First, this strategy relies on both a categorization of the type of countries in the refugee crisis as well as a selection of key policymaking episodes together with the political debates surrounding them. We focus on eight countries of four different types: Greece and Italy as frontline states, Austria and Hungary as transit states, France and the UK as closed destination states, and Germany and Sweden as open destination states. Most chapters in Part II, III, and IV study the refugee crisis in these eight countries by breaking it down into a set of key policymaking episodes involving salient policy debates. Additionally, since we look at the refugee crisis as taking place in a multilevel polity, EU-level dynamics are also included and studied following our episode approach both on their own (in Chapter 7) and in interaction with the domestic level (e.g., in Chapters 11 and 12).

Second, we have described our data collection and analysis strategies, which rely on several novel methods. Central to our book, we have introduced policy process analysis (PPA), a method that allows us to study these policymaking episodes in a multifaceted fashion by taking into account the actions undertaken, the fine-grained issues touched upon in the episode, the actors involved in the debate, as well as their substantive positions toward the policy at hand and their discursive framing strategies. While these data capture central aspects of the episodes we have selected, we do combine them with a variety of additional original datasets in order to capture those elements of the refugee crisis the PPA fails to measure. In particular, we have introduced core-sentence analysis (CSA), survey analysis, and speech analysis data collection efforts, which enable us to further zoom in on the collective mobilization dynamics and the political party election campaign strategies throughout the refugee crisis. The building blocks of the methods introduced in this chapter are essential for understanding the specific indicator construction and usage in the chapters to follow.

Introduction

In this chapter, we present the crisis situation – the policy heritage in the relevant policy domain of asylum policy, as well as the immediate problem and political pressure at the EU level and in the eight countries. The refugee crisis of 2015–16 was not the first refugee crisis in Europe. Other such crises preceded this one and have shaped the policy heritage at both the EU and the national level, which in turn was what the decision-makers relied upon when the problem pressure and the political pressure kept mounting during the summer and early fall of 2015. With increasing numbers of refugees arriving in Europe, the crisis pressure has been building up continuously. We track the mounting pressure in terms of the number of asylum seekers (problem pressure) Europe-wide, the salience of immigration issues in the national publics, and the strength of the radical right (political pressure) in the different countries. The crisis situation is expected to set the stage for the policymaking patterns as the crisis evolves.

In the first place, policy responds to the consequences of policy legacies (Heclo Reference Helbling, Gary and Mirilovic1974). Past policies create a situation of path dependence that limits the available choices for policymakers in the crisis situation. Policy legacies generate institutional routines and procedures that constrain decision-making. In particular, they constrain the range of available options (Pierson Reference Poggi2004). In the multilevel polity of the EU, the heritage of past policies refers both to the EU and the domestic level. Importantly, in the EU polity, the supranational level is not just another level at which international agreements are negotiated to be transposed nationally later on, nor is the EU a full-fledged federal system. In this “compound polity,” as a result of market integration and the more or less extensive pooling of core state powers, the EU member states are highly interdependent.

In the domain of asylum policy, responsibility is shared between the EU and the member states. While the latter have retained core competences, their policymaking still depends on the common Schengen–Dublin framework. Moreover, the policy-specific legislative framework is embedded in the overall institutional structure of EU decision-making. In asylum policy, the mixture of interdependence and independence of the member states imposes reciprocal constraints on the decision-makers at each level of the EU polity: While the interdependence imposes constraints on the policy response of national policymakers, the independence national policymakers have retained constrains the decision-making in asylum policy at the EU level. The limited competence of the EU in the asylum domain poses a great challenge for policymaking in a crisis, a challenge that is enhanced by the diversity of the policy heritage as well as the uneven incidence of the crisis in the various member states.

The immediate problem pressure is crisis-specific, as is the distribution of the incidence across the member states. The refugee crisis represents a specific type of crisis in terms of its problem structure and in terms of the distribution of its incidence across the EU member states. Crucially, the problem structure of this crisis implied a high degree of urgency but only a limited degree of uncertainty. Given the previous experience with refugee crises, one could have seen this crisis coming, and, as we shall see in the next chapter, the EU Commission was, indeed, preparing for its advent. But when the crisis arrived, it still hit the member states unprepared and required responses under conditions of high urgency. Crucially as well, the incidence of the crisis across EU member states was asymmetric. Some member states were hit hard by the crisis, while others hardly experienced any problem pressure at all. As we have already seen in the previous chapter, in addition to the problem pressure, the capacity to deal with the problem also varied considerably between member states, as some were more resourceful than others. We shall argue that the asymmetrical distribution of problem pressure and problem-solving capacity across member states, combined with the independence member states have retained in asylum policymaking, made joint responses particularly difficult. Political pressure added to this predicament in a number of key member states.

Policy Heritage

As already mentioned, the refugee crisis of 2015–16 was not the first refugee crisis in Europe. The most important previous crisis was the one linked to the Balkan wars in the early 1990s. At the end of the Cold War, between 1989 and 1994, the break-up of the former Yugoslavia led to the inflow of roughly 1.5 million refugees into the EU and in particular into Germany (see Figure 4.1). Germany not only managed hundreds of thousands of refugees at the time, but it also received 1.1 million ethnic German “Aussiedler” from central–eastern Europe and the former Soviet bloc. In reaction to this influx, the German government sought to export the crisis to the EU level. As Schuster (Reference Scipioni2000: 120) has already observed in her comparative analysis of the asylum policy in seven western European member states, six of which are part of our selection: “[P]erhaps the most remarkable about all the countries discussed in this volume is that, in spite of their different histories and experiences of granting asylum, asylum policy in each state has been so reactive. Asylum policy is developed and changed in response to particular crises.” Thus, although more limited in quantitative terms than the 2015–16 crisis, the previous crises have been very important in shaping the thinking of policymakers in the field. It forged the EU’s policy heritage, or, in Andrew Geddes’s (Reference Geddes2021) terms, its repertoire of migration governance. The Dublin Convention, which became the centerpiece of European asylum policy, was adopted in 1992, at the height of the earlier crisis. It was to determine which member states would have jurisdiction in matters of asylum – fatefully, responsibility was attributed to the member state in which the refugees arrived. As Geddes (Reference Geddes2021) argues, the past experiences of migration policymakers with crises generally shape their representations of what is normal about migration. Perceptions of normality, in turn, define what they know how to do and what they think they are expected to do next. Core perceptions and beliefs, once established, are hard to change and prove to be rather stable over time, even in new crisis situations.Footnote ¹

Figure 4.1 Refugee crises in Europe: number of asylum requests in the EU and in Germany, 1982–2020, in thousands

Sources: 1982–1997: UNHCR Statistical Yearbook 2001, Table C1; 1998–2020: Eurostat

The repertoire of EU migration governance has two components: free movement internally and a common migration and asylum policy with regard to third country nationals (TCNs). Put simply, the EU has an open borders framework internally, but external migration restrictions (Geddes and Scholten Reference Geddes2016). EU member states cannot control internal movement, but they are in charge of regulating admission of TCNs. While none of the EU laws govern admission, there are EU laws covering asylum, the return/expulsion of TCNs, family migration, the rights of TCNs who are long-term residents, highly qualified migrant workers, seasonal migrant workers, and a single permit directive linking work and residence. Added to this are a lot of other activities. Overall, EU asylum policy is partial in that it covers some, but not all, aspects of policy, and it is differential in that its effects have been more strongly felt in some member states than in others. Crucially, as argued by Geddes (Reference Geddes2021), while the numbers of TCNs in general and asylum seekers in particular to be admitted and their “integration” remain matters for member states, EU measures on migration and asylum are primarily oriented toward stemming “unwanted” flows at the external borders.

Problem Pressure

It was the external shock of mass displacements that created the crisis situation, that is, the urgency for decision-makers at the national and EU levels. This shock came to a head in the summer and fall of 2015, as is illustrated by Figure 4.1, which presents the development of the overall monthly number of asylum requests in the EU and in Germany, the country that received the largest number of such requests (as in the previous crisis). This figure also shows that, measured by the number of asylum requests, the problem pressure in 2015–16 crisis was considerably larger than in the previous crisis in the early 1990s. Given the accumulated experience with refugee crises, one might have expected that the uncertainty linked to the new crisis would be rather more limited and that the decision-makers were better prepared for this crisis. However, as we have already pointed out, this was not the case. EU asylum policy proved to be quite inadequate for dealing with the crisis shock, and the individual member states were, at least at first, left to their own devices. Past policy failures exacerbated the problem pressure that was mounting in the summer and early fall of 2015.

Importantly, however, the shock varied enormously from one member state to another, as is shown in Figure 4.2, which presents the monthly submissions of asylum requests as a share of the population (a proxy for problem pressure). It is in Austria, Hungary, Germany, and Sweden that the number of asylum applications peaked in the crisis situation of fall 2015 (indicated by the vertical solid line). Relative to the population, the peaks were most important in Hungary and Sweden, followed by Germany. In absolute terms (see Figure 4.1), Germany received by far the largest number of applications. While Germany and Sweden became the key destination states, Hungary and, to a lesser extent, Austria remained transit states, even as they also faced increasing numbers of asylum requests. In Hungary and Austria, asylum requests had already increased in the years preceding the crisis, and they peaked in 2015 at the height of the crisis. Thus, from a couple of thousand in a year, the number of claims submitted in Hungary increased to 19,000 in 2013 and 43,000 in 2014, and they reached a peak of 177,000 in the crisis year of 2015. Even so, only a minor percentage of the refugees submitted their asylum request in Hungary, as most refugees arriving in Hungary wanted to reach the destination states in northwestern Europe. This is most dramatically illustrated by the events of September 4, 2015, when thousands of asylum seekers marched on a Hungarian highway in their stated goal to reach German soil (Than and Preisinger Reference Than and Preisinger2015). Moreover, most asylum requests were subsequently rejected by the Hungarian authorities. Similarly, Austria also waved through most of the arrivals to Germany and destinations farther north.

Figure 4.2 Monthly submissions of asylum requests in 2010–2019 as a percentage of the population

Source: Asylum requests: Eurostat: asylums statistics; for Germany: EASY registrations (Erstverteilung Asylbegehrende, initial distribution of asylum seekers)

By contrast, France and the UK were mostly spared by the crisis in summer and fall 2015. These potential destination states were not accessible for refugees due to the strict regulatory regime, border control practices, and geographical location – which confirmed their status as restrictive destination countries. The inflow of refugees increased only slightly in France and was essentially nonexistent in the UK. France experienced an almost linear increase in asylum requests after the peak of the refugee crisis of 2015–16, but it was at such a low level that it is hardly noticeable at all in Figure 4.2. The UK’s geographically remote position with its maritime borders coupled with a restrictive immigration regime that had already made the “hostile environment” a reality on the ground by the time it was officially declared ensured that it would never face the kind of immediate problem pressure that open destination countries such as Germany and Sweden had to deal with.

Finally, the problem pressure measured by the number of asylum requests was also rather limited in the frontline states, in spite of the fact that one of them – Greece – was at the epicenter of the crisis because the inflow of refugees into the EU mainly passed through Greece. If measured by the number of arrivals, the problem pressure was most important in Greece, as is illustrated by Figure 4.3. In March 2015, the number of arrivals started to climb. They increased throughout the summer and autumn of 2015 and peaked at 211,000 in October 2015, before gradually declining to below previous levels in March 2016, when the deal between the EU and Turkey was signed. While Greece was overwhelmed by arrivals in summer 2015, most of the refugees arriving in Greece made their way farther north and did not register themselves with the Greek authorities, who were unable to process large numbers of applications anyway. Of the 800,000 arrivals in 2015, the EU Commission estimates that only 60,000 remained in Greece (Dimitriadi and Sarantaki Reference Dimitriadi and Sarantaki2019).

Figure 4.3 Asylum seekers and arrivals in Greece as a percentage of the population (arrivals are three-month rolling averages)

However, once the Balkan route started closing in early 2016 and once the EU–Turkey agreement was signed in March 2016, the number of asylum applications started to climb as some of those stuck in Greece took their chances with applying for asylum there. From that point onward, asylum applications and arrivals have tended to evolve together, which serves to document not only the advanced control provided by the joint EU–Greece hotspot approach but also the fact that the main movement corridors were shut down, allowing Greece time to process the backlog of asylum requests.

Italy faced a different type of crisis than Greece: Rather than a sudden and explosive shock, its type of crisis was characterized by small but reoccurring shocks, which had already started before the refugee crisis of 2015–16 and which persisted during 2017 and 2018, as shown in Figure 4.4. It is only after the Italian–Libyan deals and the port closures that Italian arrival and asylum patterns displayed a steady declining trend. Moreover, the Italian migration pattern had seasonal characteristics: A lull in the winter is followed by an increase in sea arrivals in Sicily during the spring and summer – a scenario that, until 2018, played out each year in a similar fashion. In Italy’s case, too, a large percentage of these arrivals did not register in the country, as is made clear by the two lines in the graph: Until 2018, only a fraction chose to apply for asylum in Italy, with the rest instead probably pursuing their journey toward other European countries without being registered.

Figure 4.4 Number of asylum seekers and arrivals in Italy as a percentage of the population

The two trends are largely uncorrelated up to mid-2017, when their relationship goes into reverse and becomes more tightly aligned: The arrivals drop significantly, and the number of asylum applications exceeds the number of arrivals. At this point, most probably as a result of the increasing difficulties related to pursuing their journey to other European countries, a larger number of refugees decided to register with Italian authorities.

Political Pressure

Political pressure contributes to problem pressure and the urgency perceived by policymakers in two ways: On the one hand, political pressure is exerted by the issue in question becoming more salient in the general public, and on the other hand, political pressure results from the issue being picked up by challengers in the party system who “own” it or by social movements from outside of the party system that mobilize on an “ad hoc” basis. We use as indicators for political pressure the issue of salience in public opinion, which is measured by a Google trends search for topics related to immigration and refugees; the issue salience according to Eurobarometer data; and the presence of a radical right challenger party – that is, the party that “owns” the immigration issue – at the outset of the crisis.

In terms of the salience of the issue in public opinion, political pressure was added to problem pressure in precisely those member states where problem pressure was greatest. Figure 4.5 presents the public salience of immigration and refugees as measured by Google trends and by the share of respondents to the Eurobarometer who considered immigration to be one of the most important problems at the time of the interview. As is shown by this figure, in the open destination and transit states, the public salience of immigration and refugees spiked at precisely the moment of greatest problem pressure at the peak of the crisis. In Germany, the salience of the migration issue shoots up in summer 2015, peaks in September 2015, and then drops off in two steps – first in November 2015 and then in early 2016. When measured by the second salience indicator – the most important problems mentioned by Eurobarometer survey respondents, the salience of migration similarly shoots up in the second half of 2015 – from 29 percent who consider immigration as one of the three most important problems facing the country in early July to an incredible 75 percent in September 2015. The salience of immigration declines more slowly according to this measure and remains at a high level of more than 30 percent up to the end of 2018. In the Swedish public, not only is the salience of migration issues closely related to the refugee crisis (with a peak in fall 2015), but the salience of immigration declines more slowly according to this measure and remains at a high level of more than 30 percent up to the end of 2018. In the Swedish public, the salience of migration issues is closely related not only to the refugee crisis (with a peak in fall 2015) but also to the national elections (witness the secondary peaks in September 2014 and September 2018), which saw the rise of the Sweden Democrats (SD), the radical right party in Sweden (see below). In fall 2015, no less than 44 percent of Swedes considered immigration to be among the most important problems, up from 21 percent in the first half of 2012. Similar developments can be observed for transit states – Austria and Hungary. In Hungary, 34 percent of the public considered immigration to be one of the most important problems facing the country in fall 2015, up from only 8 percent in 2013. Similarly, in fall 2015, the immigration issue was most important to 56 percent of Austrians, up from 19 percent in the first half of 2015. In these countries, too, after the peak of the crisis, the public salience in terms of the most important problem did not fall off as rapidly as the salience measured by Google trends.

Figure 4.5 Salience of immigration in national publics: Google trends and share of Eurobarometer respondents who consider immigration to be one of the most important issues

Public salience of immigration and refugees increased in the frontline and closed destination states at the peak of the crisis, too, but to a much more limited extent. France is the extreme case, where the public hardly registered the crisis at all. In the UK, the other closed destination state, the immigration issue had already been salient in the public before the crisis. Its salience in the British public did spike in fall 2015, but in terms of Google trends, it peaked only in June 2016, at the time of the Brexit referendum, when related concerns featured in the campaign, such as the largely unfounded claim that Turkey was about to join the EU. Greece, in turn, was in the thrall of the Eurozone crisis when the inflow reached its peak in fall 2015, and the crisis tended to crowd out any other public concern. The fact that most of the arrivals pursued their journey to the north also explains the relative lack of public salience of the issue in Greece, as does the fact that the migration flows at that point were concentrated on five islands, where only a small percentage of the Greek population lives. As the number of stranded asylum seekers grew, however, and the opposition’s leadership changed in late 2015, the perception of the issue became much more salient in the public sphere, as is evidenced by the Google salience trends in Figure 4.5, which reached a peak in Greece in March 2016, at the time of the EU–Turkey agreement and the closure of the Balkan route.

In Italy, finally, the salience of the issue of immigration seems to have a reverse relationship with the actual problem: The issue rises in importance according to Eurobarometer data, reaches its peak by 2017, and then recedes in importance. But the salience according to Google trends remains comparatively high throughout 2018 and 2019, although the number of arrivals and asylum applicants was in full decline (Figure 4.4). The case of Italy illustrates that problem pressure and political pressure do not necessarily rise and fall in lock-step, even if they did so in the open destination and transit states during the refugee crisis. Importantly, political pressure may actually be constructed by political entrepreneurs for their own purposes, and it may serve as a substitute for problem pressure. In other words, immigration issues may be rendered salient by the operation and effects of politics and the wider socioeconomic context within which they are embedded (Hadj-Abdou, Bale, and Geddes Reference Hagelund2022). Party strategies play an important role in this context (Abou-Chadi, Cohen, and Wagner Reference Abou-Chadi, Cohen and Wagner2022).

This brings us to the more direct pressure exerted by political challengers. Figure 4.6 presents the monthly vote intentions for the radical right party in the different countries. We can first distinguish between countries that already had a strong radical right before the crisis (Austria, Hungary, Italy, and France) and countries that had a comparatively weak radical right before the crisis (Germany, Sweden, Greece, and the UK).

Figure 4.6 Political pressure: radical right vote shares by country, monthly vote intentions

Among the strong radical right challengers, with the exception of France, all have been reinforced by the experience of the crisis. Only the French radical right did not benefit at all from the crisis, which is not surprising, given that France hardly experienced any crisis shock at all. Among the other three countries, the rise due to the crisis was temporary in the two transit countries, followed by a decline for similar reasons: Both the Austrian Freedom Party (FPÖ) and the Hungarian Jobbik were outcompeted by their mainstream center right competitors – People’s Party (ÖVP) and Fidesz, respectively. It is important to recognize, however, that at the time the crisis hit, both in Austria and in Hungary, the government was under pressure from a strong radical right. In Hungary, Jobbik had crossed the 20 percent threshold in the 2014 elections, becoming the main opposition party, and by the spring of 2015 it was polling above 25 percent. Well-positioned to capitalize on the influx of asylum seekers, Jobbik’s rise prompted Fidesz’s Victor Orbán to shift gears and to outcompete Jobbik on its own terrain – immigration. Contrary to the main thrust in the party competition literature that highlights mainstream parties’ difficulties in coopting the radical right vote (Meguid Reference Meguid2005; Pardos-Prado Reference Pardos-Prado, Lancee and Sagarzazu2015), Fidesz’s shift to the right on immigration has proved surprisingly successful. As shown by Figure 4.5, after its peak in the spring of 2015, Jobbik’s electoral strength began a steady decline, leaving Fidesz in a dominant position on the right by the end of the refugee crisis.

In Austria, the FPÖ had already been on the rise before the refugee crisis and had obtained 20.6 percent of the vote in the 2013 elections. As the crisis hit, the party clearly was one of its beneficiaries: At the peak of the crisis in fall 2015, its vote intentions reached 32 percent, and in summer 2016, it was 34 percent. In the local elections in Vienna, which took place at the peak of the crisis in October 2015, the issues of immigration and security were most salient. The dominant SPÖ defended an open border policy, while the FPÖ, its main challenger, called for a more restrictive policy. The SPÖ lost roughly 5 percent (from 44.4 to 39.6 percent), while the FPÖ gained roughly the same share (from 25.8 to 30.8 percent). Most importantly, in the presidential elections, which took place in April 2016, just after the EU–Turkey agreement had been signed and in the midst of heated debate on the asylum law, the candidates of both the SPÖ and the ÖVP failed to reach the run-off, where the candidates of the FPÖ and the Greens faced each other. This was an important reason for the then SPÖ chancellor to step down.

Importantly, the pressure from the rise of the FPÖ was felt not only by the SPÖ but maybe even more so by the ÖVP, which took a sharp right turn early on in the refugee crisis. The ÖVP leader also stepped down before the next elections and was replaced by Sebastian Kurz. The 2017 elections essentially turned into a battle over the meaning of the developments in migration policy since 2015 – a battle that was won by the mainstream conservative camp. From a traditionally pro-business party, the ÖVP transformed itself into a party focused on “law and order” in migration policy, largely adopting the respective positions of the FPÖ (Bodlos and Plescia Reference Bodlos and Plescia2018: 1357). In 2019, the FPÖ ended up being almost destroyed by a huge scandal involving its leader in the so-called Ibiza affair.

Italy is a special case because, as we have already seen, problem and political pressure were not aligned. Accordingly, the Lega and Fratelli d’Italia, the two parties of the radical right, increased their vote share not only when the crisis hit in fall 2015 but above all in later phases of the crisis when it became largely politically constructed by the very parties that benefited the most from its construction.

The four countries with an originally comparatively weak radical right can be divided into the two open destination states (Germany and Sweden), where the radical right rose as a result of the crisis and became a stable element of the party system, and Greece and the UK, where the radical right hardly benefited from the crisis and ended up essentially disappearing for reasons that were highly idiosyncratic – prosecution of Golden Dawn as a criminal organization in Greece and the aftermath of the Brexit referendum in the UK. However, as in the case of the transit countries, in both of these countries, too, competition from the mainstream right played a non-negligible role in the disappearance of the radical right. In Greece, New Democracy’s right-wing faction, emboldened by its role in helping elect the then leader of the opposition, Mitsotakis, had acquired crucial influence in the party. Launching rhetorical attacks on the government, its members managed to turn New Democracy into the largest beneficiary of the widespread public perception that Syriza’s “pro-immigration policies” had been one of the causes of the refugee inflow (Dimitriadi and Sarantaki Reference Dimitriadi and Sarantaki2018). This contributed to the party’s sudden climb in voting intentions and deprived the far right of its electoral support. In the UK, the radical right challenger (UKIP) was originally disadvantaged by the first past the post electoral system. In addition, similar to Greece, the Conservatives succeeded in outcompeting UKIP by becoming a hard-Brexit party. As we have argued in the section on British policy heritage, in this particular case, freedom of movement within the EU had become closely linked to the Brexit issue, and it was on this issue that the main bout of competition between the radical right challenger and the mainstream party from the right took place in the UK.

Finally, the cases of open destination states are very interesting because in both of these countries, the radical right experienced its breakthrough belatedly compared to other northwestern European countries and strongly benefited from the crisis. For different reasons – the long prevalence of class politics in Sweden and the fascist heritage in Germany – the rise of the radical right was delayed in these two countries. The breakthrough of the Sweden Democrats (SD) took place in 2010, when they obtained 5.7 percent of the vote, and they more than doubled their vote share (to 12.9 percent) in 2014. At the peak of the crisis in 2015, they polled 23 percent, becoming the largest opposition party. They have maintained an average of around 20 percent ever since. Their rise is part of the decline of class politics in Sweden, of the growing salience of sociocultural politics and in particular of the politicization of immigration, of the increasing convergence between the major mainstream parties (Social-Democratic and Conservative Parties), and of the radical right’s moderation (Rydgren and van der Meiden Reference Saunders2019; Jungar Reference Kalogeraki2015). Predictably, the Sweden Democrats exploited the refugee crisis and mobilized against asylum seekers coming to Sweden. Among other things, they praised Orbán’s hardline immigration policy in Hungary, organized an information campaign in foreign media to discourage asylum seekers from heading to Sweden, and even called for Sweden’s withdrawal from the EU if that was the price they had to pay for ending free movement.

Similarly, the German AfD rose belatedly. It had originally experienced a first breakthrough in 2013 thanks to its opposition to the Eurozone’s bailout operations, but eventually it benefited enormously from the refugee crisis. While it had not crossed the electoral threshold in the 2013 elections, when it received only 4.7 percent of the vote, it gained additional ground in the European elections of 2014 and in the subsequent German state elections, transforming itself from a neoliberal elitist party to a prototypical populist radical right party (Bremer and Schulte-Cloos Reference Bremer, Schulte-Cloos, Kriesi and Hutter2019a). It had fallen in a trough by summer 2015 (with only 4.7 percent of vote intentions) but rose rapidly during the peak months of the refugee crisis. By the end of 2015, it had reached 10 percent. After a new setback in summer 2017, it obtained 12.7 percent in the fall 2017 elections and has been able to maintain this level of support ever since.

As we have already seen in the cases of Austria, Hungary, Greece, and the UK, and as we shall see in the subsequent chapters, political pressure on the governments during the refugee crisis did not only or, depending on the country, not even mainly come from the radical right challengers. Transformed parties of the mainstream right, whether in opposition or in government, have become key protagonists of opposition to the reception of asylum seekers and of tightening asylum policies during the refugee crisis.

Conclusion

The configuration of the crisis situations among the member states makes for a complex configuration of transnational interests. Given the cumulation of both problem and political pressure in the open destination and transit states, we expect these states to become the major protagonists not only in the national responses to the pressure but also in the search for a joint EU policy response to the crisis. For these states, stopping the inflow of refugees and sharing the burden of accommodating the refugees who had already arrived was a priority. In the short run, the two types of states shared a common interest, which aligned them with the frontline states but put them in opposition to the restrictive destination states and the bystander states, as we argued in Chapter 2. While the interests of the transit states were clearly in line with those of the open destination and frontline states with respect to the inflows, the position of transit states with regard to accommodation was more ambiguous, since they clearly benefited from the secondary movements of the refugees within the EU. Moreover, the interests of the frontline and destination states differed with regard to the reform of the CEAS: Together with the other member states, open destination states were in favor of restoring the Dublin regulation, while the frontline states wanted to reform the CEAS in such a way that they would no longer have to assume the entire responsibility for accommodating the flood of new arrivals. Among the hard-hit open destination states, Germany is a special case. Even if it shared the most explosive combination of problem and political pressure with some other member states, the combined pressure became particularly important in its case – because of its size and influence, which enabled it to take the lead in common initiatives.

Let us finally add that the political dynamics that develop based on the country-specific conditions in the crisis situation are hard to predict. They depend not only on the exogenous pressure in the crisis situation but also on endogenous political dynamics – the actor configurations in the respective countries and the strategies of the respective political actors, which do not follow general expectations. Thus, with the benefit of hindsight, we know that the two transition states played an outsized role in managing the crisis. But we are hard put to formulate some general expectations in their respect. In both of these countries, center right parties radicalized, outflanked their radical right competitors, and proceeded to become the dominant governing parties. Both of these governments adopted policy positions designed to limit the number of successful asylum seekers, and both of them became highly influential in shaping the European response to the refugee crisis. The Hungarian government, together with its allies among the eastern European bystander states, actually became the most important opponent of European burden sharing in asylum policy. We shall now turn to the characterization of the individual episodes to get a better idea of how these political dynamics evolved during the crisis.

In Chapter 3, we introduced the policy episodes during which the policymakers elaborated their multidimensional response to the crisis at both levels – six episodes of policymaking at the EU level and forty episodes at the national level. In this chapter, we present these episodes and their exogenous and endogenous drivers in more detail to lay the groundwork for the subsequent analysis of the way policymakers reacted to the crisis. In a first step, we show how the overall politicization of the crisis response developed over time.Footnote ¹ This will allow us to characterize the timing of the policymaking during the crisis in a summary way. To be sure, we consider only key episodes of policymaking that are particularly likely to get politicized. But even within this highly selective set of episodes, there is great variation in terms of the extent to which they have become politicized, as we intend to show in the first section of this chapter. In addition, in this part of the chapter, we shall also discuss the episodes in terms of their key drivers, which we have introduced in the previous chapter – problem pressure and political pressure. As we shall see, in addition to these forces, endogenous factors also played a considerable role in determining the timing of the episodes.

In presenting the development of the politicization of the policy response over the course of the crisis, we shall distinguish between three periods: the precrisis period, which starts in early 2013 with the initiation of the first episode in our set and lasts until August 2015, when the crisis situation becomes acute; the peak period, lasting from September 2015 until the adoption of the EU–Turkey agreement in March 2016; and the postpeak period, which extends over several years from April 2016 up to the end of February 2020. We shall show that the politicization of the policy response at the EU level and at the level of the member states reached its apex during the peak period of the crisis. A closer look at the episodes involved at the national level will reveal, however, that not only the level of politicization but also its timing varied greatly across member states.

In the next two steps, we shall zoom in on the individual episodes at the two levels, briefly indicating their politics and substantive policy content, although space constraints will not allow us to go into much detail. While the timing and the details of the policymaking process are hard to predict and are sometimes rather surprising, in substantive terms, the policy responses did not stray very far from the well-known policy heritage in the asylum policy domain. In the 2015–16 refugee crisis, EU asylum policymaking remained prone to continuity rather than change (Ripoll Servent and Zaun Reference Zincone, Zincone, Penninx and Borkert2020), and the same can be said of national policymaking. Despite crises often acting as “windows of opportunity,” the breakdown of the EU’s asylum system in the 2015–16 crisis has triggered the same kind of response as in past crises – namely, a shift of responsibility outward and a reinforcement of border control at the EU level (Guiraudon Reference Guiraudon2018). At the national level, it led to the reintroduction of border controls at domestic borders and to a further retrenchment of asylum policy across the member states. In general, the measures introduced during the crisis were consistent with an approach at the national and EU levels that can be traced back for more than two decades (Geddes, Hadj Abdou, and Brumat Reference Genschel and Jachtenfuchs2020).

The Overall Politicization of the Policy Response during the Crisis

We have measured the monthly politicization of the policymaking processes during the refugee crisis at the EU level and across all eight countries at the national level.Footnote ² The two graphs in Figure 5.1 present the development of politicization over the three periods of the crisis for the EU and for the eight member states as a whole. The two graphs differ with respect to the indicator for politicization at the national level – the cumulation of the national politicization across the eight member states (graph a), as opposed to the average national politicization in a given member state (graph b). The two vertical lines in the graphs refer to the quickening of the crisis in September 2015 and to the adoption of the EU–Turkey agreement in March 2016 – the key moments that divide the crisis into its three periods. As is immediately apparent, the politicization of the crisis reaches its apex during the peak period, at both levels. For the EU, politicization is single peaked at the time of the EU–Turkey agreement; for the member states, there are two peaks, one at the moment the crisis explodes in September 2015 and another at the time of the adoption of the EU–Turkey agreement. But note that politicization does not subside in the aftermath of the adoption of the EU–Turkey agreement. More limited peaks follow in the third phase at the level of the member states in particular.

(a) Sum of national politicization: smoothed curves^a;

(b) average of national politicization: smoothed curves^a

^aThe first vertical line refers to the beginning of the crisis in September 2015, and the second one refers to the adoption of the EU–Turkey agreement in March 2016. The graph presents three-month running averages.

Figure 5.1 Development of politicization of the policy response during the refugee crisis at the EU and national levels.

Figure 5.1 also shows that the politicization of the asylum policymaking process had already started before the crisis situation became intolerable in September 2015, although it stayed at a low level. As we shall see below, it was above all the restrictive destination states that had already, before the crisis shock in September 2015, taken measures to restrict asylum seekers’ access to their countries. Finally, graph a shows that the cumulated politicization of the crisis at the member state level far outreaches its politicization at the EU level. If taken together, a lot more was going on in the member states than at the EU level. Indeed, the attentive public that follows quality news sources may have gotten this impression, given that such news sources report on a variety of countries. However, national policymakers are responsible only for what is going on in their own country. Thus, it might be more accurate to juxtapose the politicization of the crisis at the EU level to the average politicization of policymaking in the eight member states. Graph b provides this information. Viewed from this perspective, the development of national politicization is much flatter and far outclassed at its peaks in September 2015 and March 2016 by the politicization at the EU level. Compared to the politicization of the crisis in any individual member state during the peak of the crisis, but not before and after the peak, the EU-level politicization was most impressive.

As we have seen in the previous chapter, at the peak of the crisis, the problem pressure and the political pressure (measured in terms of public salience) were at their maximum in the transit and open destination states, and to some extent also in the frontline states (at least in Greece if measured by the number of arrivals). We expect the politicization of the crisis to be a direct response to the pressure exerted by the crisis situation on policymakers in the respective countries. The problem pressure in the crisis situation is bound to focus the governments’ attention on the policy domain that is hit by the crisis shock. Theories of the policy process stress the importance of attention to policy domains and the limited attention spans of governments (Jones Reference Jones and Baumgartner1994; Baumgartner and Jones Reference Jones and Baumgartner2002; Jones and Baumgartner Reference Jones, Kelemen and Meunier2005). More specifically, the crisis situation is likely to cause a so-called serial shift in policy processing, that is, a shift from parallel processing in policy-specific subsystems to serial processing in the “macro-politics” of top executives. At the same time, the crisis situation also concentrates the mind of the public on the policy domain in question. Just like the top brass political decision-makers, the public is focusing serially on one thing or at most a few things at a time (Simon Reference Sitaropoulos1983), given its limited attention span and the limited capacity of the media (Hilgartner and Bosk Reference Hinnfors, Spehar and Bucken-Knapp1988). The increased public attention on the policy in question is likely to reinforce the pressure on the government to act.

Table 5.1 provides a straightforward measure of the relationship between the pressure on the policymakers and the politicization of their crisis response: the correlation between politicization on the one hand and the three indicators for pressure that we introduced in the last chapter on the other hand. These correlations do not inform us about causal relationships, but they give us a rough idea of the strength of the association between the variables involved. Summing over all eight member states, the correlations are quite high, varying between r = 0.59 and r = 0.75. In other words, in line with expectations, the pressure exerted by the crisis is rather closely associated with the politicization of policymaking in response to the crisis.

However, if we go to the level of the individual member states, the association turns out to be close only in the two open destination states and in Austria, one of the transit states, and only for two of the three indicators for pressure. Even in the key open destination state, Germany, the association with problem pressure is relatively modest (r = 0.46). In the frontline states, politicization is associated only with public salience, but not with problem pressure (in Italy, the corresponding correlation is even negative), and in the closed destination states and Hungary, all associations are quite weak. While we would have expected such weak associations for the closed destination states, which were not directly hit by the crisis, the low associations in the case of Hungary are somewhat unexpected. For the third indicator, political pressure as measured by the radical right vote share, correlations are, with the exception of Hungary, generally low or even negative.

If this shows that policymakers reacted to the combined problem pressure and political pressure at the peak of the crisis by launching policy episodes in the most heavily hit countries, the associations between politicization and pressure are not as strong as we might have expected. The reason is that policy episodes were also politicized by factors endogenous to politics: The anticipating reactions of policymakers, the strategies of political entrepreneurs, key events, the legislative cycle, and the endogenous dynamics of policy reactions to the crisis once they have been set in motion all contributed to the politicization of the crisis, too. We can get an idea of the importance of such endogenous factors by inspecting the timing of the individual episodes at the EU and the national level.

To start with the EU level, the EU Commission responded to the rising tide of refugees in anticipation of things to come. In May 2015, it had presented the European Agenda for Migration, which sought to formulate a comprehensive EU approach to the surge in Mediterranean arrivals. The agenda-setting by the Commission rested on four pillars (Geddes Reference Geddes2018):

• Strengthening the common asylum policy with a reform of the Dublin regulation

• Improving control of the external border (through solidarity with border countries such as Greece and Italy and strengthening the mandate of Frontex)

• Reducing incentives for irregular migration (addressing the root causes of such migration in countries of origin, dismantling smuggling and trafficking networks, and better application of return policies)

• A new policy on legal migration

Based on this agenda, the Commission launched four of the five policy episodes we cover in this study in spring or summer 2015, that is, before the peak of the crisis. But the Commission’s proposals were not yet followed up by the European Council. Thus, the Commission had proposed to use, for the first time, the emergency response mechanism under Article 78(3) to set up a temporary relocation scheme (for a total of 40,000 persons in need of international protection) based on mandatory country quota to relieve the frontline states, Greece and Italy. The number of persons to be relocated seemed quite small, given the dimensions of the inflow of persons in need. But even this very limited measure was watered down by the July 20, 2015, European Council meeting: Participation in the scheme was to remain voluntary rather than mandatory as proposed by the Commission. At the EU level, the policymakers saw the crisis coming, but they did not yet react decisively.

For the national level, Figure 5.2 presents a systematic overview over the starting points of the national episodes by type of member state against the background of the developments of problem pressure (number of asylum requests) and political pressure (public salience of immigration and asylum as measured by Google trends). The vertical dashed lines in this figure indicate the starting points of the episodes, with grey lines referring to border control measures and black lines to modifications of asylum rules. The figure shows how the timing of the episodes varied depending on the type of member state. Thus, in the closed destination states (France and the UK), most episodes were initiated before the advent of the crisis and do not seem to be directly related to increases in problem pressure (which was comparatively low anyway) or to public salience of refugee and migration issues. These states had preventively taken measures to close their borders and to retrench their asylum policy. By contrast, in the frontline states (Greece and Italy), we observe a clustering of episodes rather late in the day – their starting points are only partially related to the development of pressure. It is in the transit and open destination states that the episode triggers are clustered just before the peak or during the peak of the crisis, when problem pressure and political pressure in the respective countries were at their maximum. With the exception of Austria, however, even in these countries, some of the episodes intervened only in later stages of the crisis.

(a) Frontline states;

(b) transit states;

(c) open destination states;

(d) closed destination states (smoothed curves).

^aVertical lines indicate the beginning of an episode. Gray lines refer to Border Control episodes, while black lines refer to asylum rules’ episodes. For a given type of member state, both problem pressure and political pressure are displayed on the same scale.

Figure 5.2 Starting dates of the episodes in relation to problem pressure (asylum requests) and political pressure (public salience).

As a matter of fact, there are several instances of episodes launched by political entrepreneurs. As we have argued in the previous chapter, immigration issues may be rendered salient by the operation and effects of politics and the wider socioeconomic context within which they are embedded (Hadj-Abdou, Bale, and Geddes Reference Hagelund2022), and party strategies play an important role in this context (Abou-Chadi, Cohen, and Wagner Reference Abou-Chadi, Cohen and Wagner2022). As the emergency politics literature reminds us, there can be strategies of “crisisification” (Rhinard Reference Riker2019). According to this strategy, action is often delayed until a foreseeable policy problem escalates into a crisis, and the ensuing crisis is then “exploited” to increase support for public office-holders or their policy agendas (Boin, ’t Hart, and McConnell Reference Boin, Hart and McConnell2009; Rauh Reference Rauh2022). There is, however, also an alternative strategy of political entrepreneurs that consists of them creating a crisis where there is hardly a policy problem at all. Several episodes among our selection correspond to the latter pattern.

Thus, the low association between problem/political pressure and politicization in Hungary is explained by the fact that three of the five episodes occurred after the crisis peaked and problem pressure ceased to exist. These episodes all refer to measures that the Fidesz government under Victor Orbán introduced in its attempt to outbid its radical right competitor as a defender of the national cause – the Legal Border Barrier Amendment further tightened the already very tough border control regime, and the other two episodes served to attack NGOs’ supportive of refugees. In the frontline states, too, only two of the five episodes in Greece and only one of the Italian episodes were launched at the time or just preceding the time when the crisis peaked. The two Italian episodes that occurred in the aftermath of the crisis as well as two of the three Greek episodes that occurred late in the day responded more to endogenous political dynamics triggered by a political entrepreneur than to external pressure, and their high public salience is more likely the result of political dynamics than their cause. In Italy, the two episodes were related to port closures in fall 2018, which were a direct consequence of the policy of the new minister of the interior and leader of the Lega, Salvini, who attempted to exploit the refugee issue for his own political purposes. In Greece, the political entrepreneur in question was Turkish president Erdogan, whose policy to incite refugees to cross the border into Greece in order to put pressure on the EU led to two belated Greek episodes: One of them was the domestic conflict on the islands created by the increasing number of arrivals, and the other was the direct Greek reaction at the land border to Turkey. The last German episode, finally, was in many ways similar to Salvini’s port closures. It was instigated by the new minister of the interior, Seehofer, who aimed to toughen the German border controls for his own political purposes in June 2018.

The legislative cycle not only played a role in the strategies of the new ministers of the interior, Salvini and Seehofer, but it also helped to initiate one of the three late episodes in Greece. The so-called International Protection Bill was the first act related to immigration from the newly elected New Democracy government, which aimed at streamlining the asylum process, making it faster and stricter.

Triggering events launched at least one of the German and French episodes. In both instances, the events were terrorist attacks. Thus, after the terrorist attack by a Tunisian refugee on a Christmas market in Berlin on December 19, 2016, the issues of return and deportation of rejected asylum seekers became particularly salient in the public debate in Germany, which triggered the introduction of a new act on deportation (return) in January 2017 and its adoption in July 2017. In France, border controls became a highly salient issue after the November 2015 terrorist attacks (Bataclan, Stade de France) in Paris. Following these attacks and ahead of UN climate talks in Paris, France introduced border checks on all of its borders. Subsequently, citing the persistent threat of terrorism, France renewed the border checks every six months up to the end of the period covered.

The issue of return also provides an example of the implications of early decisions to open the door to a large number of refugees. The issue became pressing in the aftermath of the peak of the crisis as large numbers of asylum seekers who did not qualify for asylum in the destination countries were required to return to their country of origin. Not only in Germany, but also in Sweden, one of the episodes deals with this issue. Finally, the last Swedish episode was a direct sequel to an earlier episode that had introduced temporary residence permits for asylum seekers for a limited period of time, after which the measure had to be amended again.

Zooming in on EU Policymaking

We have already introduced the basic distinction between external border control measures and internal measures concerning asylum rules. At the EU level, external border control actions have been somewhat more frequent, with 57 percent of all actions in the six policymaking processes, but asylum rules have been important, too, accounting for 43 percent of all the actions. Figure 5.3 presents the development of the politicization of decision-making processes across the period covered, with a focus on these two types of episodes at the EU level. The left-hand graph illustrates the predominance of Border Control episodes during the peak phase. The right-hand graph provides the details for the four episodes that focused on border control. As we can see, the EU–Turkey agreement dominated the peak phase completely: The externalization of refugee protection to Turkey was the single most politicizing policy decision taken during the crisis, not only at the EU level but overall. It was more salient than any other episode, but in terms of polarization, it was only slightly above the rather high average. The other Border Control episodes – the much more limited deal with Libya, the hotspot approach, and the reinforcement of the European Border and Coast Guard – were much less politicized. As a matter of fact, on average, the two episodes referring to asylum rules at the EU level – the relocation quotas and the Dublin regulation – were more politicized than the Border Control episodes at the EU level and even more politicized than both types of episodes at the domestic level. This is illustrated by Figure 5.4, which shows the average level of politicization and of its components – salience and polarization – by episode type and level of polity. As this figure clarifies, polarization is high for both types of episodes at both levels of the polity. However, in terms of the average salience, the episodes concerning asylum rules at the EU level stick out, which makes them most politicized overall. At the national level, border control episodes are somewhat more polarized and salient than episodes concerned with asylum rules, but not by much.

Figure 5.3 Thematic focus of policymaking at the EU level: development of issue-specific politicization over time

Figure 5.4 Politicization by episode type and level of polity

At the EU level, the relocation of refugees was the Commission’s first attempt to come to terms with the crisis. But, as we have already seen, the Commission had been blocked in its attempt to introduce a relocation mechanism to provide for burden-sharing between member states in summer 2015. But it did not give in. In his speech on the 2015 state of the union, which was held on September 9, at the very moment when the crisis blew up, Commission president Juncker announced a proposal for a second mandatory emergency mechanism that aimed to relocate a further 120,000 persons seeking international protection from Greece, Italy, and Hungary. Under the pressure of the crisis situation, the response was immediate: On September 14, an extraordinary meeting of the Council of Ministers of the Interior took place in order to adopt this plan. While the European Parliament endorsed the emergency mechanism on September 17, the plan met with great resistance from eastern European member states. Nevertheless, under German pressure, at another extraordinary meeting of the Council of Ministers on September 22 that was arranged by the Germans, the relocation mechanism was adopted by qualified majority voting: Twisting the arms of several reluctant member states (including Poland), the Germans obtained the required majority.

As van Middelaar (Reference van Middelaar2017: 110) observes, this “revolutionary decision,” pushed through by the Germans, who did not want to be left alone with the task of receiving and integrating refugees, turned into a fiasco. From the seeming German victory, the European refugee policy would not recover. The Czech Republic, Slovakia, Hungary, and Romania had voted against the relocation mechanism; Finland had abstained. Hungary, which had originally been proposed as a beneficiary of the emergency relocation mechanism, rejected the offer. In Poland, the liberal center right government soon was to be replaced by a conservative right government, which joined the eastern European resistance against the mechanism in the so-called Visegrad group (V4). Two of the countries that had voted against the mechanism – Hungary and Slovakia – appealed to the ECJ against the decision, and Hungary later organized a referendum over the relocation quota (see below). Eventually, the ECJ upheld the decision in September 2017, and the Hungarian referendum held on October 3, 2016, after the largest ever advertising campaign, failed to reach the quorum due to opposition boycott. Nevertheless, in central and eastern Europe, the fight for public opinion was lost for a long time. From this point on in this part of Europe, the acceptance of refugees was viewed not as a humanitarian act but as submission to Berlin. As a result, the implementation of the decision fell far short of the expected numbers. Van Middelaar (Reference van Middelaar2017: 110–112) suggests that the crucial mistake was the attempt to keep the European Council, where qualified majority decisions are not possible, out of the loop.

At the same time as Germany tried to alleviate its burden with internal burden sharing, it also sought the help of Turkey to stop the arrival of refugees on the Greek islands. The contacts were already established in late summer 2015. Only with controlled external borders could Germany maintain its welcome culture. Between October 2015 and May 2016, Angela Merkel traveled no less than five times to see President Erdogan in Turkey, bowing to him in an unusual bout of European “realism.” A first joint action plan of the EU with Turkey was agreed at the EU Council meeting on October 15–16. On November 29, the EU Council decided to implement this plan, but in mid-December, eleven member state governments rejected the implementation plan (Webber Reference Weinberg2019: 167). Arrivals remained high, and the negotiations between Turkey and the EU continued, driven by the German chancellor, who, backed by the European Commission, fought for her political survival. Slominski and Trauner (Reference Slothuus and Bisgaard2018: 109) point out that the deal was negotiated in a format that shielded the EU member states from the other EU supranational institutions, notably the EP and the ECJ. After a dramatic finish during the early days of March, negotiations eventually succeeded: The EU–Turkey agreement that finalized the deal between the EU and Turkey was adopted on March 18, 2016. In the aftermath of the agreement, the implementation of the deal gave rise to protracted additional negotiations, which we followed until September 2016, at which point the episode breaks off in our data.

The deal stipulated that as of March 20, 2016, new irregular migrants entering Greece from Turkey had to be returned to Turkey. For every Syrian being returned to Turkey from the Greek islands, another Syrian was to be resettled in the EU. The maximum number of people to be returned according to this mechanism was 72,000. As part of the agreement, Turkey promised to take necessary measures to prevent new sea or land routes from Turkey to the EU. In return, the EU promised to pay Turkey up to 6 billion euro to contribute to its expenses with Syrian refugees by the end of 2018. It also promised to upgrade the customs union, accelerate visa liberalization for Turks in the EU, and relaunch the accession process. As a result of the deal, arrivals on the Greek islands dropped sharply, as did registered deaths and missing persons in the Aegean Sea.

Since the adoption of the EU–Turkey statement more or less coincided with the closing of the west Balkan route (see below), the question is which of the two measures was responsible for the effective closure of the EU’s borders. As van Middelaar (Reference van Middelaar2017: 118) argues, both measures contributed to this result. The arrivals started to decline once the west Balkan route was closed, but the decline was accentuated after the adoption of the EU–Turkey Deal. He suggests that Turkey agreed to the deal only once it realized that the EU was ready and able to close the border without its cooperation. There is an important difference between the two measures, however: While closing the west Balkan route abandoned Greece, an EU member state, the EU–Turkey Deal allowed Greece to stay in the Schengen area.

The EU–Turkey agreement, the most important measure at the EU level, was one of the first examples of EU realist foreign policy, and it has been criticized by those who do not consider Turkey a place where asylum protection is in accordance with international standards (Niemann and Zaun Reference Niemann and Zaun2018: 8). Legal considerations in this respect have been partly removed by declaring Turkey a safe third country. Other critiques argued that this deal exposed the EU to blackmail by a leader with clear authoritarian leanings. Moreover, the agreement did not deal coherently with the situation in Greece: It did not cover the more than 42,000 refugees who had entered Greece before March 20 and who remained in Greece after the agreement. And finally, the deal did not work out as planned. While the number of arrivals dropped by 97 percent three years on, the number of returns remained very limited (only 2,441 migrants had been returned since March 2016), and the number of resettlements of Syrians from Turkey to EU member states remained rather limited as well (roughly 20,000 in total).Footnote ³ The threat of being returned to Turkey and the closing off of Greek borders to the north seem to have been sufficient to dissuade most refugees from making the crossing to the Greek islands. Eventually, after the summer 2016 coup in Turkey, negotiations on the implementation of the deal went sour and, except for its financial contribution, the EU did not deliver on its promises.

The hotspot approach, another border control measure adopted by the European Council during the peak period, was part of the European Agenda on Migration. The European Asylum Support Office, Frontex, and Europol were to work on the ground with frontline member states, in particular Greece and Italy, to swiftly identify, register, and fingerprint incoming migrants. On the whole, notwithstanding the “assistance” rhetoric, hotspots were designed to shift back to frontline states all the responsibilities they (theoretically) have to shoulder under current EU legislation: to identify migrants, provide first reception, identify and return those who do not claim protection, and channel those who do so toward asylum procedures in the responsible state – in most cases, none other than the frontline state itself. The implementation of the approach in Greece and Italy has been slow, due in part to the need to build the procedures from scratch and with shortcomings in infrastructure, staffing, and coordination but also due to foot-dragging on the part of the two frontline states (see below).Footnote ⁴ At the end of 2016, the reception facilities in the two countries were still inadequate, particularly in terms of accommodation and international standards for unaccompanied minors.

The creation of the European Border and Coast Guard (EBCG), the third border control measure implemented swiftly in late 2015, involved extension of the already existing border control agency Frontex, which had been created in 2004 on the eve of the “Big Bang enlargement.” The proposal for the creation of the EBCG was drawn up in record time by the Commission in the midst of the crisis situation, between September and December 2015 (Niemann and Speyer Reference Niemann and Zaun2018: 32f). Frontex’s mission was to coordinate operational cooperation; assist member states in training, technical equipment, and joint return operations; follow up on technical innovation; and conduct risk analyses (Niemann and Speyer Reference Niemann and Zaun2018: 26f). The former Frontex had been underfunded and lacked administrative staff, a deficiency that was addressed by creating a standing 1,500-member-strong rapid reaction pool of border guards and technical equipment, to which the member states committed explicit contributions that could not be withheld. The new EBCG would have funding worth 322 million euros by 2020, up from the 114 million euros that had been originally budgeted for 2020.

The critical question in the creation of the EBCG was whether it had the right to intervene even if the member state on the territory in which it wanted to intervene did not agree – a critical question for the constitutional set-up of the EU, as van Middelaar (Reference van Middelaar2017: 123) points out. In this case, and contrary to the relocation issue, the European Council agreed to a compromise solution: If a member state did not cooperate within thirty days with an emergency plan designed by the EBCG on behalf of the Council, the Commission could start the procedure to suspend the country’s membership in the Schengen area. In other words, the EU could not control the external border against the explicit will of a member state, but it could exclude the country from access to the area of free movement if it did not cooperate. This provision allowed for the closure of a possible gap in the external border without forcing a joint solution on a resisting member. The new EBCG soon proved to be too limited, however. In his state of the union speech in 2018, Commission president Juncker confirmed that it should have an additional 10,000 border guards by 2020, and he provided a blueprint for the future of the EBCG (Angelescu and Trauner Reference Angelescu and Trauner2018).

With the closure of the eastern Mediterranean, the focus of the refugee streams shifted back to the central Mediterranean and to the sea crossing between Libya and Italy. Following up on an Italian deal with Libya, in February 2017, the European Council also turned its attention to the support of Libya in controlling the central Mediterranean route. The Malta Declaration of February 3, 2017, outlined a number of measures as part of a comprehensive strategy to strengthen the EU’s intervention along this route. The declaration pledged 200 million euros to the North of Africa window of the EU Trust Fund for Africa, with a priority to be given to Libya for 2017.Footnote ⁵ A series of measures followed, all of which were designed to actively support Libyan authorities in contributing to efforts to disrupt organized criminal networks involved in smuggling migrants, human trafficking, and terrorism.

The EU episodes concerning asylum rules refer to the relocation of refugees in particular and to the reform of the Dublin regulation in general. Having failed in the short term with its relocation measures, the Commission repeatedly proposed a reform of the dysfunctional Dublin regulation as a long-term response to the crisis. This crucial internal solidarity measure was, however, repeatedly shelved – a blatant case of non-decision-making in the face of a major crisis. The new Commission, which took over after the EP elections in 2019, rapidly proposed a new plan for the reform of the CEAS – the so-called Pact on Migration and Asylum, which has met with the same lack of success as the attempts of the previous Commission. As the Covid-19 crisis hit the EU, asylum policy more or less disappeared from the agenda of EU decision-makers, and further reform steps have been shelved once again.

Zooming in on Policymaking at the National Level

At the national level, the thematic focus of policymaking varies heavily across the type of member state, as is shown in Figure 5.5, which presents the share of border control actions by member state type and crisis period. Border controls include measures to secure the external borders of the EU as well as border closures between EU member states. While border control was more in the focus in all member states during the peak phase of the crisis, it was the exclusive issue in frontline states. Thus, in all three phases, major policymaking episodes in Italy were exclusively devoted to border control issues, as were all episodes except one in Greece. Border control issues were also dominant in transit states during the first two periods but lost much of their importance in these states during the third period. By contrast, in both types of destination states, asylum rules prevailed in the prepeak period as well as in the postpeak aftermath of the crisis.

Figure 5.5 Share of border control actions by member state type and crisis period: percentages

Conclusion

Table 5.2 provides an overview of the episodes ordered by level and type of member state. It indicates the thematic focus, the start and the end dates as well as the duration of each episode, and the extent of its overall politicization. In terms of timing, we have observed that the politicization of the responses adopted by government during the crisis was most intense during the peak period, both at the European and the national level. This is in line with the expectation that the combined problem and political pressure during the peak period would incite the authorities to rapidly initiate and adopt policy responses to come to terms with the massive inflow of refugees. However, the association between politicization and pressure, both problem and political pressure, proved to be rather variable across member states and looser than expected. We have tried to account for this finding by taking a closer look at the endogenous political dynamics during the crisis. Policy responses at the national level were not only required by the failure of the CEAS and by the inability of the leaders to adopt joint solutions at the EU level, but these policies were also the result of a series of endogenous factors at the national level, which operated independently of problem pressure and, in part at least, created the political pressure in the first place. The strategies of political entrepreneurs – Orbán, Salvini, Seehofer, and Erdogan – most clearly fitted this bill, but anticipation of crisis situations to come, legislative cycles, conspicuous events like terrorist attacks, and sequels of policy decisions made earlier in the crisis all contributed to these endogenous dynamics.

In terms of thematic focus, we distinguished between border controls and internal retrenchments of asylum rules. At the EU level, four of the six episodes concerned the control of external borders, with the EU–Turkey agreement dominating all other episodes, which turned out to be the single most politicized episode during the refugee crisis because of the episode’s very high saliency at the peak of the crisis. In terms of polarization, however, it does not stick out, since all types of episodes, whether dealing with border controls or with the retrenchment of asylum rules, were typically highly polarized. At the national level, the mix of measures depended on the type of member state: In frontline states, border controls prevailed, while in the UK, asylum rules prevailed. In transit states, open destination states, and France, both types of measures were important for coming to terms with the crisis situation.

With respect to the substantive content of the policy responses, continuity prevailed, with the possible exception of integration laws in Germany and Austria, which, however, also only adopted what other countries (e.g., the Netherlands) had already implemented before (see Joppke Reference Juhász2017). The crisis did not prove to be an opportunity for reforming the existing system. Instead, failure to reform at the EU level and retrenchment at the national level were the predominant responses. The internal rebordering between member states constitutes a persistant threat to the internal freedom of movement policy, the retrenchment of asylum rules contradicts Europe’s humanitarian values, and the externalization of the border control to Turkey makes the EU vulnerable to the whims of the Turkish president. The outcome is a form of stagnation or inertia that reproduces the policies in the asylum domain without providing the output the polity is meant to produce.

In the subsequent parts of this volume, we shall analyze in detail the actor configurations, conflict structures, and political dynamics of policymaking during the crisis to show how this state of affairs came about.