Introduction

In this chapter, we present the actors and conflicts at the EU level. The study of these aspects of the crisis management includes the analysis of the actors and conflict configurations in the different episodes as well as the politicization of the episodes. We begin by introducing expectations about the actors and conflict structures at the EU level, which reiterate some considerations we have already introduced in Chapters 1 and 2. Next, we proceed to presenting the actor distributions and conflict structures in the six EU-level episodes. In a third step, we show how the various episodes have been politicized by the different actors and adversarial camps that we identified previously, overall and in the two key phases of the refugee crisis – the peak phase preceding the conclusion of the EU–Turkey agreement and the phase following the adoption of this agreement.

As we have argued in Chapters 1 and 2, in the multilevel polity of the EU, the supranational level is not just another level at which international agreements are negotiated to be subsequently implemented nationally. Polity membership creates a foundational interdependence that stems from the original choice to become a member of a compulsory association. Market integration and the extensive pooling of core state powers have increased this interdependence over time. Still, the EU is not a full-fledged federal system, and the degree of interdependence varies by policy domain. As we have observed in Chapter 4, in the domain of asylum policy, responsibility is shared between the EU and its 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: On the one hand, the interdependence restricts the possible policy responses of national policymakers, and on the other hand, the independence that national policymakers still enjoy constrains the decision-making at the EU level. The limited competence of the EU in the asylum domain poses a great challenge for joint EU policymaking in times of crisis.

In terms of relevant actors, the grand theories of European integration locate the power alternatively in the supranational agencies – the Commission (neofunctionalism) or the European Council (new intergovernmentalism) – or in the member states (liberal intergovernmentalism, postfunctionalism). Given the low capacity and lack of policy resources of supranational institutions in the asylum policy domain, we expect supranational entrepreneurship to be highly constrained (Moravcsik Reference Moravcsik2005: 362–363). Under such conditions, the success of the policy proposals by supranational actors depends on the support by the member states. In the case of the refugee crisis, opposition to joint solutions and conflicts between the member states have been reinforced by two conditions: First, the member states were asymmetrically affected by the crisis and unequally prepared to deal with it. While the frontline and open destination states, the states directly hit by the crisis, favored joint solutions, the bystander and to some extent also the transit and closed destination states were less affected by the crisis and therefore were less ready to share the burden (Noll Reference Odmalm2003; Bauböck Reference Bauböck2018). Second, joint action was constrained, and conflicts between member states were reinforced by the politicization of national identities produced by the uneven distribution of crisis pressures within the EU polity. Consistent with the predictions of postfunctionalism, 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 pointed out by Ferrara and Kriesi (Reference Ferrara and Kriesi2021), this decision-making scenario is consistent with the postfunctionalist notion of “constraining dissensus.”

It is the territorial channel of representation in the EU that provides the most important (although not the exclusive) conduit for the politicization of the reciprocal constraints and related conflicts. Accordingly, intergovernmental coordination has become the key decision-making mode in the EU in general, and particularly in crisis situations. In this mode of decision-making, the heads of member state governments (in the European Council) and responsible ministers (in the Council of Ministers) assume a decisive role. They provide the critical link between the two levels of the EU polity. As a result of their 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 – the 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 crucial role not only in domestic policymaking in the refugee crisis but also in policymaking at the EU level.

Under crisis conditions, the role of key executives of both the EU and member states is likely to become even more prominent. Under such conditions, which combine high political pressure in the sense of conflict-laden salience with high time pressure (urgency), executive decision-making is expected to become the preferred mode of decision-making both at the supranational and the national level. In a crisis, policymaking is no longer confined to the policy-specific subsystem (asylum policy in our case); rather, it becomes the object of macro-politics or “Chefsache,” to be taken over by the political leaders who focus on the issue in question. The decision-making mode of intergovernmental coordination corresponds to the EU-specific version of executive decision-making.

Foremost among the expected conflict lines are the vertical and transnational conflicts involving member states and the EU. In Chapter 2, we have formulated some expectations about these conflict lines. At this point, we reiterate the general expectations formulated in Chapter 2. In the short run, that is, in the early phases of the crisis, we expect open destination and transit states to share a common interest in stopping the flow of arrivals and in sharing the burden of accommodating refugees, which aligns them with the frontline states but opposes them to the restrictive destination states and the bystander states. While at first the transit states’ interests are clearly in line with those of the open destination and frontline states, the position of transit states is likely to get more ambiguous as the crisis progresses, since they clearly benefit from the secondary movements of the refugees within the EU. Moreover, the frontline and destination states are also divided with regard to the reform of the CEAS: Together with the other member states, open destination states are in favor of restoring the Dublin regulation, while the frontline states demand reform of the CEAS to share the responsibility for accommodating the flood of new arrivals.

The configuration of member states’ interests is further complicated by country-specific conditions. Thus, as a nonmember of the Schengen area, the UK largely stands outside of conflicts involving burden sharing. The ambiguous crisis situation of transit states provides room for mobilization by political entrepreneurs, as has been the case of Prime Minister Orbán in Hungary and of Foreign Minister Kurz in Austria. Similarly, the ambiguous situation of frontline states, which have to deal with incoming arrivals but have an incentive to close their eyes to secondary movements, also provides opportunities for political entrepreneurs to exploit the crisis, as we have also discussed in previous chapters. Moreover, the directly concerned states that are interested in joint solutions do not necessarily all sit in the same boat. In general, their support for joint solutions depends on the specific conditions attached to them: If the EU intervention comes with strings attached and is perceived to impinge upon the state’s sovereignty, it may not be accepted even if it were to bring direct relief from the crisis pressure. Thus, external border control, demanded by open destination states, may involve the direct intervention of the EU in the national sovereignty of frontline states, as was the case in two EU episodes – the episodes of the hotspots and the EBCG. In the hotspot episode, the frontline states were expected to take back all the responsibilities they shoulder under current EU legislation, an expectation to which, as we have seen in Chapter 5, they responded with foot-dragging and other forms of informal resistance. In the EBCG episode, Greece was reluctant to subscribe to the plan to deploy the transformed EBCG without the consent of the directly concerned member state. Such resistance may be overcome by external pressure, as in the case of the hotspots, where the border closures at Greece’s northern border with Northern Macedonia put an end to Greek resistance, or by compromise solutions, as in the case of the EBCG, which implied that the EBCG could not be deployed without the consent of the directly concerned member state, which, in the case it refused to give its consent, risked a suspension of its membership in the Schengen area.

In addition to vertical and transnational conflicts involving member states, there are two other types of international conflicts involved in the policymaking at the EU level. One of them results from the EU’s strategy to externalize the burden of border control during the refugee crisis. As we have seen (Chapter 5), two of the six episodes at the EU level involved this kind of response to the crisis – the EU–Turkey agreement and the EU–Libya arrangement. In such instances, we expect the EU to present a more united front, since the externalization of the border control provides the EU member states with a public good from which they all benefit. Instead, the main conflict is expected to involve the EU and/or its member states on the one hand and the third country to which the burden is intended to be externalized on the other hand. In the case of Turkey, it was above all the EU that confronted the third country, while in the case of Libya, it was Italy, the member state most concerned by refugee arrivals from Libya. The other type of international conflict refers to other international organizations, which may get involved in the management of the crisis. Thus, White (Reference Wihtol de Wenden2020: 81f) points to the involvement of NATO in the management of border control with Turkey. Arguably, however, it was not NATO but UN organizations such as the UNHCR that played a considerable role in the management of the refugee crisis at the Turkish border with Europe. The UNHCR not only supported the reception efforts in the frontline states but also was a vocal critic of the situation in the hotspots and in the Mediterranean.

At the EU level, the conflict structure is expected to be dominated by these four types of international conflicts: vertical conflicts between the EU and the member states, transnational conflicts between member states, externalization conflicts with third countries, and conflicts with other international organizations. As we have seen in the previous chapter, at the national level, partisan, intragovernmental, and societal conflicts prevail, in addition to international conflicts. At the EU level, however, partisan conflicts are likely to be negligible, given the weakness of the European parties, while conflicts with civil society organizations are likely to play an important role, given the large number of humanitarian NGOs active in the migration policy domain (e.g., NGO ships in the Mediterranean rescuing migrants or NGOs supporting migrants in the camps). In addition to humanitarian NGOs defending the migrants, civil society actors also include migrant organizations, think tanks and individual experts making proposals for joint solutions (e.g., Gerald Knaus, head of the European Stability Initiative, the think tank that first floated the idea of the EU–Turkey Deal), or the media (e.g., by exposing shipwrecks or inhumane conditions in the camps). Finally, there is a possibility of intra-EU conflicts between different EU authorities. Conflicts between the Commission and the Council involve conflicts between the EU and the member states and are, therefore, already covered by the vertical conflicts introduced above. However, the crisis management may also pit other EU authorities against each other – for example, the Commission/Council against the European Parliament, all three institutions against specialized agencies like the ECB or Frontex, or different factions within one and the same institution (e.g., different Directorates-General [DGs] of the Commission).Footnote ¹ We do not expect to find a lot of such internal conflicts, not only because conflicts within agencies are more difficult to pinpoint by our approach, which relies on public sources, but also because we assume that in the refugee crisis, the conflicts mainly involved member states, with respect to which the EU authorities took a rather homogenous position.

The Actors

For the actor distribution, we first show the distribution over three summary categories – member states, EU actors, and other actors. At the EU level, member states are virtually exclusively represented by national governments. Thus, the category of the member states is almost exclusively composed of national governments and their agencies and includes only a few actions attributable to local governments (1 percent) and to governing parties (2 percent). The category of EU actors is dominated by the Commission, which accounts for roughly half (49 percent) of the actions attributable to EU actors, the other half being almost equally divided between the Councils (European Council and Councils of Ministers) (24 percent) and other EU actors (European Parliament, parties, and specific agencies) (27 percent). The category of others consists of third countries (Turkey and Libya) (36 percent), supranational organizations (roughly 24 percent), and civil society organizations (roughly 40 percent). Table 7.1a presents the distribution of the actions in the six EU-level episodes over these three actor categories. As we can see, the member state governments and EU actors jointly dominate decision-making in four out of the six episodes. The other actors are very important only in the two episodes that aim at the externalization of border control. Obviously, in these two cases, the third country that is directly concerned plays a key role, as can be seen in Table 7.1b. Civil society is also important in these two episodes (as well as in the relocation episode). It includes above all NGOs (43.2 percent) but also experts and media (17.6 percent), migrants and their organizations (16.0 percent), and opposition parties (18.4 percent).

Table 7.1 confirms above all the central role of member state governments. At the EU level, they are even more important than at the national level (see Chapter 6): In all episodes except for the EU–Turkey agreement, they are the most salient actors and account for almost half of the actions overall, compared to a third at the national level (see Table 7.1). Table 7.1b demonstrates that, except for the closed destination states, which are least present at the EU level, all types of member states are roughly equally represented in EU episodes. However, each type is not equally represented by its component members. Thus, Germany accounts for no less than three-quarters (76.8 percent) of the actions of the open destination states, while Sweden is virtually absent at the EU level (with a share of only 3.5 percent of EU-level actions of open destination states). Even Luxembourg and the Netherlands (which assumed the EU presidency in the second half of 2015 and in the first half of 2016, respectively) have a greater presence among the open destination states, with 7 to 8 percent of the latter’s actions each. The interventions of the frontline states are mainly attributable to Italy (60 percent) and Greece (30 percent); those of the transit states mainly to Hungary (47 percent), Austria (34 percent), and Bulgaria (12 percent). The actions of the bystander states are more evenly distributed among a larger number of states, but the Czech Republic and Slovakia (22.6 percent each) as well as Poland (16.5 percent), which together with Hungary formed the V4 group, are the ones most present. Finally, the closed destination states are above all represented by France (with a share of 68 percent of the corresponding actions). The German government is the most salient member state government at the EU level, closely followed by the Italian government, and, at a greater distance, by the governments of Hungary, Austria, and Greece.

The presence of the different types of member states varies, however, from one episode to another. Thus, the open destination states (and above all Germany) were most involved in the EU–Turkey agreement. The frontline states dominated in the EU–Libya episode (Italy) and in the hotspot episode (Greece), and they were also heavily present in the Dublin Reform episode, where they are the key promoters of reform. The transit and bystander states, in turn, predominated in the relocation episode, where they were the main adversaries of a joint solution. The closed destination states, finally, were a minor force in all episodes, which reflects the fact that they were hardly affected by the crisis.

We also present the target actors in Table 7.2. While national governments are the preferred targets at the national level (see Chapter 6), at the EU level, it is the EU institutions that are the most important targets – overall and in four out of the six episodes. Only in the externalization episodes is the third country targeted even more frequently. This already foreshadows that the conflict lines run between the member states (the most important actors) on the one hand and the EU institutions and third countries (the most important targets) on the other hand. In terms of member states, the frontline states are the most frequent targets. Especially in the hotspot episode, Italy and above all Greece were the privileged targets of the interventions by the EU and other member states. In the EU–Libya episode, it was Italy that played the key role as both actor and target.

Turning to the individual actors, the question is whether the top executives played the expected role in the policymaking processes at the two levels. In this respect, we distinguish between four types of actors: top executives at the EU and the national level; other individual actors who have been mentioned by name in the media reports; and institutional actors, who are responsible for actions that have not been explicitly attributed to any individual. The top executives at the EU level include two leaders – Commission president Jean-Claude Juncker and Council president Donald Tusk. At the national level, they include the prime ministers and presidents (where they are not merely symbolic figures) representing our eight countries: Alexis Tsipras for Greece; Matteo Renzi, Paolo Gentiloni, and Giuseppe Conte for Italy; Viktor Orbán for Hungary; Werner Faymann, Christian Kern, and Sebastian Kurz for Austria; Angela Merkel for Germany; Stefan Löfven for Sweden; François Hollande, Emmanuel Macron, and Manuel Valls for France; David Cameron for the UK; and Ahmet Davutoğlu and Tayyip Erdogan for Turkey.

Table 7.3 compares the role of top executives in the decision-making processes at the EU level with their role at the national level. As is immediately apparent, top executives play a more important role in EU decision-making than in decision-making at the national level. Moreover, national top executives are more prominent policymakers in these crisis episodes at the EU level than EU top executives are. This confirms the expected pivotal role of government leaders of the member states in the two-level EU decision-making. They account for no less than one sixth of the actions (17.4 percent) in the policymaking processes at the EU level. The most prominent individual actor at the EU level is the prime minister from the most important member state, German chancellor Angela Merkel, who, on her own, accounts for 4.6 percent of all actions. She is followed by Commission president Jean-Claude Juncker and the Hungarian prime minister Victor Orbán. By contrast, EU top executives are virtually absent from the policymaking process at the national level. In national policymaking, power is not only more divided among a larger number of participants but also more focused on national policymakers.

The influence of top executives varies by stage in the policymaking process. As is shown in Table 7.4, EU top executives are most important in the proposal and negotiation phases of this process, while top executives of member states have a most important role to play in the negotiation and adoption phases of this process – mirroring the respective institutional roles of the EU Commission and the EU Council. Thus, Commission president Juncker is the individual leader most present in the proposal stage, accounting for 7.4 percent of the corresponding actions, while German chancellor Angela Merkel is responsible for no less than 10.8 percent of the actions in the negotiation phase and for 4.3 percent in the adoption phase. Only Jean-Claude Juncker comes close to her in the negotiation phase, accounting for 5.2 percent of the actions. No one else is as prominent as Chancellor Merkel in the adoption phase. These shares are all the more remarkable if we keep in mind that at the various decision-making stages, institutional actors predominate in the public sphere. As Table 7.4 shows, in the public, the actions in the decision-making stages are above all attributed to institutional actors. By contrast, it is the public claims-making that is attributed above all to individual actors and, as the table shows, it is in this respect that the top executives of the member states are also highly present. They constitute the public face of the decision-making process at the EU level during the crisis, which implies that they are also the actors who take public responsibility for these decisions and who are most likely to be blamed for them by the public.

Conflict Lines

The actors involved in the policy debate, either as initiators or as targets, reveal only one aspect of the conflict. To understand the nature of the conflict in a given policy episode, we need to make refinements both conceptually and in operational terms. Our indicator for conflict intensity, which we use here, captures both the directionality of actors’ action vis-à-vis their targets (positive, negative, or neutral) and the type of actions they undertake (see Chapter 6). It suffices to reiterate at this point that conflict intensity is a composite indicator of the actor direction vis-à-vis the target and the type of policy action that the actor undertakes.

Table 7.5 presents the average intensity of six conflict types in the six EU episodes. The most important conflicts per episode are printed in bold, the second most important conflicts are in italic. As expected, the vertical and transnational conflicts constitute the two most important conflict lines in four out of the six episodes – the two episodes involving the Asylum Rules (Relocation and Dublin Reform) and two of the Border Control episodes (Hotspots and EBCG). By contrast, the two Externalization episodes (EU–Turkey and EU–Libya) above all gave rise to conflicts between the EU and the respective third countries, with the EU–Turkey agreement the most conflictual of all the episodes. Compared to these three types of conflicts, the other conflict types were at best secondary. Conflicts with international organizations and intra-EU conflicts were generally of low intensity. The exception concerns the EBCG episode, where the transformation of Frontex into the new EBCG created some intra-EU conflicts. Conflicts between the EU/its member states and civil society have been of some importance in two episodes – the EU–Turkey agreement and the relocation quotas. In the EU–Turkey case, NGOs and opposition parties heavily criticized the deal because they did not consider Turkey a safe third country, given the human rights abuses in Turkey. They also criticized the implications of the deal for the refugee camps on the Greek islands, which suddenly became closed centers where the refugees were stuck and had to count on being returned to Turkey. In the relocation episode, NGOs like Amnesty International and left-wing opposition parties pleaded for a relocation of refugees across Europe, while right-wing opposition parties like Jobbik in Hungary, UKIP in the UK, and RN in France refused to accept additional quotas of refugees.

To represent the resulting conflict structure between the nine types of actors we have distinguished in Tables 7.1 and 7.2, we calculated two types of dissimilarities for all the actor pairs involved (i.e., thirty-six pairs): the average distance between their positions on the six episodes at the EU level and the average conflict intensity between them (as actors and targets) across all six episodes. We then multiplied the two types of dissimilarity for each pair, which amounts to weighting the distance between the two actors’ positions with the conflict intensity between them. Finally, we analyzed the resulting matrix of dissimilarities with a multidimensional scaling (MDS) procedure. Such a procedure allows us to represent the overall actor configuration in a low-dimensional space, in our case in a two-dimensional space. Actors who took similar positions in the six episodes and who did not get involved in conflicts with each other are placed closer to each other in the resulting space, while actors who opposed each other in substantive terms and fought against each other to impose their own position against the position of their adversaries are located at some distance from each other. Figure 7.1 presents the resulting summary actor configuration.

Figure 7.1 Overall configuration of conflict structure at the EU level: MDS result

We can distinguish three camps in this actor configuration: the EU, which forms the core of the policymaking space, and two adversarial camps – the noncooperative camp of the transit and bystander states, and the humanitarian camp of civil society, which also includes the supranational institutions. The core camp of the EU is joined by the frontline states (Greece and Italy above all) and the closed destination states (represented above all by France), which share similar positions. The open destination states (mainly Germany) are located at the mid-point between the EU camp and the civil society camp, which indicates that their position is closer to the humanitarian position of the civil society and the UNHCR, the most important supranational actor. The third countries, Turkey and Libya, are located between the noncooperative camp and the EU, which indicates that their position is more in line with the EU than that of the noncooperative camp but that nevertheless they are to some extent adversaries of the EU.

The noncooperation by the bystander and transit states became most obvious in the two episodes concerning the asylum rules. Three bystander states (Poland, the Czech Republic, and Slovakia) together with one of the transit states (Hungary) formed the Visegrad four group (V4), which blocked the implementation of the relocation scheme and prevented any reform of the Dublin regulation. In addition, under the leadership of yet another transit state, Austria, they embarked on the elaboration of a unilateral solution to the external border control, the closing down of the Balkan route, which they implemented by the end of February 2016. In the short run, this solution isolated Greece, but it ended up being instrumental in getting the cooperation of Turkey in the external border closure in the Aegean Sea, which allowed keeping Greece in the Schengen area.

Given the triangular configuration of the conflict space, two dimensions are needed to accommodate the relationships between the key actors at the EU level. The vertical dimension might be called the humanitarian dimension, separating most clearly the pro-humanitarian civil society actors from the noncooperative bystander and transit states. The horizontal dimension distinguishes most clearly between both the humanitarian and the noncooperative camps on the one hand and the EU camp on the other hand. We propose calling this the pragmatism dimension. The EU, together with the frontline and the destination states, tried to find a pragmatic solution to the crisis, which was opposed by the principled opposition from two sides – the civil society actors who opposed the pragmatic “realism” of the EU in the name of humanitarian principles and the V4 actors who opposed it in the name of the principles of national self-determination.

Politicization

The politicization indicators allow for yet another summary presentation of the conflicts that characterized the refugee crisis. We have already presented the thematic focus of the politicization at the EU level in Chapter 5. We would now like to focus on the contribution of the various actor types and actor camps to the politicization of the policymaking process at the EU level during the refugee crisis. Figure 7.2a presents these contributions as well as their two components – salience and polarization – for the more detailed actor types. In this figure, the overall politicization and its components have each been standardized to the 0 to 1 range. As the figure shows, the EU actors dominate the politicization at the EU level, as well as its components: They are not only the most salient actors at this level, but they also contribute most to the overall polarization. Together with the destination and the frontline states, they are most supportive of the policy proposals at this level, but together with their allies, they also face strong opposition from the two adversarial camps, and they constitute the most frequent targets of this opposition – as we have seen in Table 7.2. In other words, EU actors constitute the most conspicuous actors on the supportive side of the policy proposals, which makes them at the same time the most conspicuous adversaries of the opponents of these proposals.

(a) Actor types;

(b) actor camps

Figure 7.2 Politicization and its components by actor types: standardized averages.

The contribution to the politicization by all other types of actors is more limited, since they are both less salient (they account for at most a third of the actions of the EU actors) and less polarizing (they are at best roughly half as polarizing as the EU actors). The closed destination states are the least politicizing actors of all, which confirms the limited stakes they had in the refugee crisis.

In the previous section, we have seen that the EU actors are allied to the frontline and destination states, which both count on joint solutions at the EU level, and opposed by two camps – the civil society camp and the camp of the transition–bystander states, with the third countries being caught somewhere in between. If we combine the actors into these opposing camps, we get a better sense of the politicization by the opposing forces. As we can see from Figure 7.2b, once we combine the actors of the camps, the EU actors no longer stick out. Taken together, the member states allied to the EU actors are contributing just as much to the politicization of the policy response in the refugee crisis as the EU actors themselves are, in terms of both salience and polarization. The two opponent camps contribute to the overall politicization to a lesser degree, since they are less present, although they are still highly polarizing. Compared to these three camps, the third countries are hardly contributing to the politicization at all. While they are contributing to the conflict intensity, as we have seen previously, they are much less visible in the European public sphere, and their opposition to the EU decision-makers is also less pronounced than the opposition from some member states and from civil society.

If we finally break down the actors’ contributions to the politicization of the policymaking process by episodes, we find that the politicization of the relocation quotas by the adversaries from bystander and transit states dwarfs all other contributions to the politicization of EU episodes (see Figure 7.3). Overall, the EU–Turkey episode has been more politicized than the relocation episode because it has been politicized by a broader set of actors, which notably does not include the bystander and transit states (see Table 4.2). However, the single most important contribution to the politicization of the refugee crisis at the EU level has been made by the opponents to the relocation quotas. This goes a long way to explain why this kind of proposal had no chance for success in subsequent debates and why later attempts to reform the Dublin regulation, which always contained some related policy ideas, have repeatedly failed.

Figure 7.3 Politicization by broad actor camps and episodes: standardized averages

Phases of the Policymaking Process at the EU Level

At the EU level, we can clearly distinguish between two phases in the policymaking process – the phase preceding the conclusion of the EU–Turkey agreement and the phase following it. About half of the actions at the EU level fall into the short first phase that lasts for less than one year, from the adoption of the European Agenda for Migration in May 2015 to the adoption of the EU–Turkey agreement in March 2016. The second half of the actions is drawn out over a four-year period that ends in February 2020. The first phase, which corresponds to the peak period of the crisis, was dominated by the two most politicized episodes – the EU–Turkey agreement and the relocation quotas. It also included most of the hotspot episode. As is shown in Table 7.6, the two most important episodes also extended into the second phase, but during this period, they no longer dominated to the same extent. The Dublin Reform became as important as the two more specific issues, and the attention shifted to the external borders in the center of the Mediterranean. This is also illustrated by Figure 7.4, which displays the relative politicization of the episodes in the two phases.

Figure 7.4 Politicization by episode and phase, average index value

Distinguishing between the two phases, we can also detect the development of the overall conflict structure at the EU level over the course of the crisis. The conflict structure was not yet as clear-cut in phase 1 and really became consolidated only during the long, drawn out phase 2. This is illustrated in Figure 7.5, which displays the conflict structures in each one of the two phases. In the first phase, at the peak of the crisis, the open destination and the frontline states constituted a cluster of their own, in the middle of the space. Their joint interest in stopping the flows and sharing the burden between all member states brought them together and placed them in opposition to some extent to all the other major actors. During the second phase, their interest became more aligned with those of the EU actors, and they also found a close ally in the closed destination states (above all, France). In the configuration of the second phase, the tripolar structure presented in Figure 7.1 emerged as a consequence of the management of the crisis. The contrast between the bystander–transit states alliance (V4) on the one hand and the frontline–destination states–EU alliance on the other hand becomes quite clear. The civil society actors and the other supranational actors (mainly the UNHCR) constitute the third (humanitarian) pole, which is opposed to both the intransigent defenders of exclusively national solutions (the V4) and the pragmatic defenders of burden sharing and of a reform of the Dublin regulation, who can rely on the support of the Commission and of the most important member states – Germany, France, and Italy. The third countries, which have become key partners of the EU in this policy domain, are located in between the three poles, ready to play off one against the other.

Figure 7.5 The conflict structures at the EU level, by phase: MDS results

Conclusion

At the EU level, international conflicts prevailed. These were mainly of three types – 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. The episodes that did not involve third countries were characterized by the first two types, while the externalization episodes obviously involved third countries. Other types of conflicts were secondary. The emerging conflict structure, which was consolidated only in the long period after the conclusion of the EU–Turkey agreement, is characterized by the antagonistic relationship between three camps – the EU core coalition (including destination and frontline states in addition to EU actors); the coalition of transit and bystander states; and the coalition of civil society actors, international organizations, and domestic opposition parties. The two-dimensional conflict space is structured by a dimension that opposes the pragmatic, “realist” EU and its allies to its principled adversaries, and a dimension that distinguishes its humanitarian from its nationalist adversaries. At the EU level, the sovereignty camp is composed of member states that have been largely spared by the refugee crisis and that refuse to share the burden of refugees with the hard-hit destination and frontline states. The latter in turn seek the help of the EU actors in their quest for burden sharing with the member states that have been largely spared by the crisis.

The actor configuration confirms the expectation that member states and their key executives play a crucial role in the two-level game of EU crisis management. In a policy domain where the EU shares its competences with the member states, it is unable to impose its policy proposals without the cooperation of the member states. As we have seen, Germany, the “hobbled hegemon” (Webber Reference Weinberg2019), and its chancellor Angela Merkel played a key role in policymaking at the EU level. 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 the case of Germany because of its size and influence, which enabled it to take the lead in common initiatives. As is suggested by the public goods literature, Germany as the largest member state and the recipient of the largest number of refugees was most engaged in the search for joint policy solutions, since it had potentially more to lose (in absolute terms) from the nonprovision of the public good in terms of stability and security, and since it also was the member state that was best able to unilaterally make a significant contribution to the provision of the public good (Thielemann Reference Thies2018: 69).