12 - Dynamics of Policymaking in the EU–Turkey Agreement

Summary

In this chapter, we show how an EU policymaking episode, the EU–Turkey agreement, the most important episode in our study, is domesticated in national policymaking and how this works out differently depending on the member state. We compare the debates in four member states and then zoom in on the debates in the two most concerned member states. For Germany, this episode was instrumental in solving a domestic conflict between the chancellor and the governing parties, including her own party. Once the agreement was sealed, the German debate did not entirely subside, but it lost intensity and eventually faded out. The Greek debate, by contrast, picked up shortly before the conclusion of the agreement and then stayed intense during the implementation phase. Several years after the agreement was concluded, it gave rise to new domestic episodes in Greece, since the problems it created for Greece continued to remain unsolved.

Introduction

In the previous chapter, we analyzed the different ways national episodes are linked to the transnational and supranational levels. In this chapter, we shall analyze the different ways one and the same EU-level episode spills over to national-level decision-making. For this purpose, we have a closer look at the most important episode at the EU level – the EU–Turkey agreement, for which we coded the policymaking process not only based on international sources but also based on the national press in four of our eight member states. We selected the two countries most concerned by this agreement – Germany (as the open destination state that received the largest number of refugees) and Greece (as the frontline state where the largest number of refugees arrived during the peak of the crisis). In addition, we chose one transit state (Hungary) and one closed destination state (the UK). While Hungary was also directly concerned, since large numbers of refugees had crossed its territory before it closed it off by building fences at its southern borders, the UK as a nonmember of the Schengen area was least concerned by this episode. In comparing the national debates, we expect the episode to have been particularly salient in the media of the two most concerned members, and this is, indeed, the case. Of the 1,574 actions we coded based on the two types of press sources on this episode, roughly a third (34.6 percent) come from the Greek media, a sixth (17.8 percent) from the German media, an eighth (13.3 percent) from the Hungarian, and a sixteenth (6.4 percent) from British sources. The remainder (27.8 percent) were reported in the international press.

In the literature, the question of the Europeanization of the public debate in the member states has been prominent (e.g., Koopmans and Statham 2010; Risse-Kappen 2015). In the present chapter, we start by reversing the perspective. 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. The first section of the chapter is devoted to these questions. In addition, we attempt to show that the very same episode has very different implications for domestic policymaking. For this purpose, we zoom in on the politicization of the agreement in Germany and Greece in particular. In the two countries most concerned by the agreement, it gave rise to bottom-up attempts to solicit support from EU agencies and fellow member states. In the case of Germany, support for the EU–Turkey agreement was vital for the political survival of the chancellor: It allowed her to escape from the trap of her open-doors policy. If she was the driving force in negotiating this agreement with Turkey, she could clinch it only with the support of the EU authorities and all the other member states. Once the agreement had been concluded, the episode faded from the attention of the German public. In Greece, by contrast, support from the EU and the other member states was needed once the agreement had been concluded. For Greece, the agreement had an ambivalent character: While it successfully stopped the inflow of refugees, it left a large number of them stranded within Greek borders, and Greece could provide for them only with support from the EU and the other member states. In fact, the consequences of the agreement in Greece lingered for several years and led to two new domestic episodes at the very end of our observation period.

The Actors Involved in the Debate on the EU–Turkey Agreement

In Chapter 7, we have seen that at the EU-level, member state governments and EU actors play a dominant role in the policymaking process and that international conflicts prevail. The member state governments provide the pivotal link between the domestic and the international levels of EU policymaking. Accordingly, we expect that the domestic debate on EU policymaking processes in a given member state places greater emphasis than the international debate does on the contribution of domestic actors from the state in question to the EU-level policymaking process. First, we expect that the national debate pays particular attention to the role of the member state’s own government in EU-level policymaking. From the domestic point of view, it is the national executive that is the main representative of the national interest in the EU policymaking process. In addition, we expect other domestic actors to be more prominent in the national debate as well. EU policymaking is likely to be contested at the national level, that is, the scope of conflict expands to some national actors who do not become visible in the international debate but who have a role to play in the determination of the government’s position in EU policymaking. In the domain of asylum policy, these national actors are not expected to primarily include interest associations, as is posited by intergovernmentalists, but rather political parties and civil society organizations, as posited by postfunctionalists. Third, EU actors are key interlocutors of the national government in each member state, which implies that the greater focus on domestic actors is unlikely to be at the expense of EU actors. Instead, we expect the greater focus on domestic actors to reduce the focus on national actors from other member states. In this regard, the German government is likely to be a special case, given its key role in the management of the refugee crisis in general and in particular in the negotiation processes of the EU–Turkey agreement. In other words, we expect the German government to be more present than any other foreign government in the other member states as well.

Table 12.1 provides a first assessment of these expectations. The first part of the table shows that member state governments are generally even more prominent in the four national debates than in the debate at the EU level. But, as expected, this is not at the expense of a lesser representation of EU actors. Only in the UK debate are EU actors less present than in the international debate. This part of the table also confirms that, in addition to national governments, other domestic actors also get more attention in the national debates than in the debate at the EU level. The increased presence of national actors is particularly striking in the German and the UK debates. The national government is also very much present in the Greek debate, but other domestic actors participate comparatively rarely in Greece. Except for Hungary, the increased presence of domestic actors is above all at the expense of the third country, Turkey. The more detailed data in the second part of the table indicate that, as expected, the increased presence of domestic actors is also at the expense of actors from other member states (both governments and other actors), which are much more present in the EU-level debate than in the national debates. As expected, the German government is an exception in this respect, since it is, indeed, quite present not only at the EU level but also in the debates of the other member states. This is additional evidence for the exceptional role played by the German government in this EU episode.

Table 12.1 The distribution of actor types in the EU–Turkey episode, by level and country

Actors	Country
	EU	Germany	Greece	Hungary	UK	Total
(a) Broad categories
EU	24.3	24.7	28.4	23.8	15.0	25.2
Member state governments	29.8	38.9	37.4	31.4	38.0	34.8
Other domestic actors	20.4	26.5	18.7	23.8	36.0	22.4
Turkey	20.2	7.8	8.4	19.1	7.0	12.9
Supranational	5.3	2.1	7.0	1.9	4.0	4.8
Total	100.0	100.0	100.0	100.0	100.0	100.0
n	436	283	545	210	100	1,574
(b) Detailed
EU	24.3	24.7	28.4	23.8	15.0	25.2
German government	9.9	21.2	5.5	9.1	8.0	10.2
Greek government	4.4	4.2	23.5	1.0	1.0	10.3
Hungarian government	1.4	1.8	0.6	12.9	3.0	2.8
UK government	0.5	0.0	0.0	0.5	17.0	1.3
Other government	13.8	11.7	7.9	8.1	9.0	10.3
Other Germany	3.4	18.4	0.6	3.3	1.0	5.0
Other Greece	1.8	0.7	6.4	2.4	0.0	3.2
Other Hungary	0.0	0.4	0.0	8.1	0.0	1.1
Other UK	0.0	0.0	0.0	0.5	26.0	1.7
Other member states	15.1	7.1	11.7	9.5	9.0	11.4
Turkey	20.2	7.8	8.4	19.1	7.0	12.9
Other supranational	5.3	2.1	7.0	1.9	4.0	4.8
Total	100.0	100.0	100.0	100.0	100.0	100.0
n	436	283	545	210	100	1,574

Table 12.2 presents the target actors of the EU–Turkey debate at the different levels. Three types of actors predominate as targets – EU actors, Turkey (except in the UK debate), and the national government (except in the Hungarian debate). Other domestic actors are essentially irrelevant as target actors. A more detailed analysis shows that, with the exception of Hungary, these other domestic actors mainly target the national government, which underlines the key role of the national government in linking the national debate to EU-level policymaking. Note that the Hungarian government seems to fulfill this linkage role to a lesser extent than the governments of the other member states do. Germany, in turn, is exceptional to the extent that Turkey constitutes by far the most important target in the German debate, which once again reflects the fact that it was German actors, above all the German chancellor, who directly negotiated with Turkey. The relative absence of Turkey as a target actor in the UK, by contrast, points to the relative lack of importance of the episode for the UK.

Table 12.2 The distribution of target actor types in the EU–Turkey episode, by level and country

Target actors	Country
	EU	Germany	Greece	Hungary	UK	Total
EU	38.5	21.1	32.7	39.2	36.4	31.9
German government	5.2	20.2	0.4	5.4	0.0	7.7
Greek government	2.3	6.4	20.8	3.1	0.0	9.6
Hungarian government	0.0	0.9	0.4	2.3	0.0	0.8
UK government	0.0	0.0	0.0	0.0	36.4	0.5
Other government	6.3	2.8	3.1	4.6	9.1	4.0
Other Germany	0.6	2.8	0.0	3.9	0.0	1.5
Other Greece	1.2	0.0	1.2	2.3	0.0	1.0
Other Hungary	0.0	0.0	0.0	0.8	0.0	0.1
Other member states	9.8	2.8	3.5	7.7	0.0	5.3
Turkey	36.2	43.1	33.1	30.8	18.2	35.9
Supranational	0.0	0.0	5.0	0.0	0.0	1.6
Total	100.0	100.0	100.0	100.0	100.0	100.0
n	174	218	260	130	11	793

The key role of Germany in this episode also becomes apparent if we consider the role of top leaders in the decision-making process for this episode. As Table 12.3 shows, national top leaders dominate the national debates in every country except Hungary, where the two top leaders from Turkey – President Erdoğan and Prime Minister Davutoğlu – are even more present than the Hungarian prime minister, Orbán. Angela Merkel, the German chancellor, not only dominates in Germany, together with EU top leaders – Juncker, the Commission president, and Tusk, the president of the European Council – she also dominates at the EU level, together with the two Turkish top leaders. She accounts for no less than 6.7 percent of the actions reported at the EU-level (compared to her accounting for 4.6 percent of the actions in all the EU-level episodes taken together; see Table 7.3) and also has a strong presence in the national press of the other member states. In the German debate, she is responsible for 10.6 percent of the actions. This confirms Merkel’s key role in this episode. In the other countries, the prime ministers also dominate – Alexis Tsipras in Greece (8.8 percent of the Greek actions), Victor Orbán in Hungary (5.7 percent of the Hungarian actions), and David Cameron in the UK (8.0 percent of the UK actions).

Table 12.3 Executive decision-making in the EU–Turkey agreement by level and country, share of top leaders^a

Top leaders from …	Country
	EU	Germany	Greece	Hungary	UK	Total
EU	3.7	5.3	4.0	5.2	5.0	4.4
Germany	6.7	10.6	3.9	4.8	7.0	6.2
Greece	1.8	0.7	8.8	0.0	1.0	3.8
Hungary	0.9	0.0	0.6	5.7	3.0	1.4
UK	0.0	0.0	0.0	0.0	8.0	0.5
Turkey	8.9	2.8	4.0	9.5	6.0	6.0
Other individuals	51.4	60.1	53.0	51.9	63.0	54.3
No names	26.6	20.5	25.7	22.9	7.0	23.4
Total	100.0	100.0	100.0	100.0	100.0	100.0
n	436	283	545	210	100	1,574

^a Major actors in bold, secondary actors in italic

The exceptional role of German actors in this episode is also confirmed once we consider the role of the various actors in the different phases of the policymaking process. To be sure, EU actors dominate all the stages of this process, as can be seen from Table 12.4. But German actors were responsible for no less than one fourth of the actions in the negotiation phase, most of which were accounted for by the German top leader, and, together with EU and Turkish actors, German actors also dominated the claims making. By contrast, actors from Greece were responsible for the bulk of the actions in the implementation phase. This contrast between the engagement of German and Greek actors indicates the different significance of the episode for the two countries most concerned. For Germany, the episode became less relevant once the agreement had been concluded, while it took on its greatest significance for Greece in the implementation phase.

Table 12.4 Role of actors from different countries by policy stage, percentages^a

Actor country	Policy_stage
	Claims (+nonstate)	Proposal	Negotiation	Adoption	Implementation	Total
EU	21.4	51.4	42.0	56.6	26.1	25.2
Greece	12.7	2.9	8.7	10.5	39.2	14.7
Germany	16.3	11.4	24.6	6.6	6.5	15.1
Other member states	18.9	14.3	1.5	7.9	3.9	16.1
Turkey–Libya	15.9	14.3	14.5	14.5	14.4	15.6
other	14.8	5.7	8.7	4.0	9.8	13.3
Total	100.0	100.0	100.0	100.0	100.0	100.0
n	1,241	35	69	76	153	1,574

^a Major actors in bold, secondary actors in italic and bold

Finally, we compare the conflict configurations at the EU level with the configurations that we observe based on the national debates. At the EU level, we found previously that the EU–Turkey episode was characterized by the conflict between the EU/its member states and Turkey (see Chapter 7). As is shown in Table 12.5, the same conflict structure emerges from the German and the Hungarian debates. In the case of the Greek debate, this conflict is still the most pronounced, but it appears to be much weaker than in the EU or in the German and Hungarian debates.¹ This is quite surprising, given the fact that the Greek debate of the EU–Turkey episode was by far the most salient one. As it turns out, however, the Greek debate was far less conflictive than the debates in the other countries. In terms of polarization, too, it was the least polarized of all the debates compared. The Greeks covered this episode a lot, but overwhelmingly in positive or neutral terms. There was less critique of the agreement in the Greek debate than in the other countries. This may not be so surprising after all, given that Greece was the main beneficiary of the agreement.

Table 12.5 Conflict scores for the dominant conflict lines, by episode^a

	EU member state	Trans-national	EU/member state–third country	EU/member state– international organization	EU/member state–civil society	Intra-EU
EU	0.04	0.02	0.28	0.02	0.08	0.01
Germany	0.05	0.11	0.28	0.02	0.04	0.00
Greece	0.08	0.02	0.13	0.05	0.04	0.02
Hungary	0.08	0.04	0.30	0.02	0.01	0.02

^a Major conflict lines are in in bold, and minor conflict lines are in italic-bold.

The Politicization of the EU–Turkey Agreement in Germany

Compared to the German national episodes, the German debate of the EU–Turkey episode was much more politicized. This is shown by Figure 12.1, which presents the politicization of the three purely domestic episodes in Germany during the first phase – the suspension of the Dublin regulation (the case of Border Control), the revision of the Asylum Law, and the introduction of the Integration Law – in relation to the politicization of the EU–Turkey agreement in Germany.

Figure 12.1 Politicization of German episodes, phase 1

The politicization of crisis policymaking in Germany starts with the suspension of the Dublin regulation and the revision of the asylum law, which run largely in parallel. After an early peak in fall 2015, the intensity of the politicization of these two episodes subsequently declines and reaches very low levels as they end. The episode of the Integration Law is generally little politicized. By contrast, the EU–Turkey agreement has already been more politicized than the domestic episodes in late fall 2015 and, at its peak in spring 2016, reached a level that was far beyond German domestic episodes and comparable to some of the highest levels of politicization of domestic episodes in the frontline states.

Table 12.6 indicates the salience of the different types of actors in the four episodes in phase 1. It distinguishes between international actors – including the EU, Turkey and other governments, and domestic actors – the chancellor, the national government (including other national institutions such as regional governments), and the government coalition partners – the CDU-CSU (senior coalition partner) and the SPD (junior coalition partner) as well as the combined opposition and civil society organizations (CSOs). As the table shows, the relevant actors vary considerably by type of episode. In the case of the EU–Turkey agreement, the public debate was dominated by international actors: Roughly half of the actions reported on the policymaking process about this agreement were accounted for by international actors – EU actors (24.7 percent), other member state governments or supranational actors (19.1 percent; most prominent among them being the governments of Greece, Austria, and Hungary, and other supranational actors), and Turkey (7.8 percent). The other half of the actions in this episode is roughly equally divided between the chancellor, government actors, governing parties, opposition parties, and CSOs. This cast of actors differs sharply from the domestic episodes, where the international actors are only marginally present (Border Control) or entirely absent (Asylum Rules and Integration Law). In the domestic episodes, the governing parties prevail, together with the government in the cases of Asylum Rules and Integration Law, with the opposition and CSOs taking the secondary role.

Table 12.6 The salience of the different types of actors in the four episodes of phase 1: percentages

Actors	Episode
	Asylum Rules	Integration Law	Border Control	EU–Turkey	Total
International actors	0.0	0.0	12.7	51.6	23.9
Merkel	8.1	2.3	15.6	10.6	10.2
National government+	40.4	27.9	16.8	14.1	22.5
Government coalition partners	37.3	47.7	39.3	9.5	27.9
Opposition–civil society	14.3	22.1	15.6	14.1	15.5
Total	100.0	100.0	100.0	100.0	100.0
n	86	173	161	283	703

It is noteworthy that Chancellor Merkel played an outsize role in the EU–Turkey agreement and in the episode on border control, where she accounts for roughly the same share of actions as the rest of her government. As the driving force behind the suspension of the Dublin regulation and the EU–Turkey agreement, she is most conspicuously present in these two episodes, where she provided the linchpin between the two levels of the decision-making process. Note, however, that she was not omnipresent in all episodes of crisis decision-making, as is illustrated by the integration law, where the specialists of the policy subsystem remained in charge and she played only a minor role. This is to suggest that even under crisis conditions, crisis policymaking does not always shift to the top executive. In the case of the German integration law – a legislative novelty for Germany that the SPD, the junior coalition partner, had demanded and that had been in the making for a long time – the crisis actually provided the window of opportunity to finally get it done.

As already mentioned, the EU–Turkey agreement was the German chancellor’s plan B for alleviating the German burden of hosting asylum seekers once the relocation mechanism had failed. Germany, and the German chancellor in particular, were heavily involved in the decision-making process for this agreement, as is documented in the previous tables. As outlined in Chapter 5, Chancellor Angela Merkel had made the unprecedented decision to keep the borders open for refugees during the night of September 4, 2015. Her decision meant that Germany suspended the Dublin regulation for Syrian refugees. Germany did reintroduce identity checks for refugees at the border on September 14, but no one who applied for asylum was refused entry. Subsequently, in spite of massive internal critique, Chancellor Merkel kept insisting on her open-doors policy.

Merkel’s decision to suspend the Dublin regulation was immediately criticized by representatives of foreign governments, members of her own party, and members of the opposition, which led Merkel to defend her decision in repeated public statements. Thus, members of the Austrian and Hungarian governments accused Germany of attracting the floods of Syrian refugees by keeping its doors open. Prime Minister Orbán declared that refugees were “Germany’s not Europe’s problem.” Critique also came from the EU: In December, Donald Tusk, the president of the EC, called for a reversal of the chancellor’s refugee policies. He demanded that the Dublin rules be respected and called on European states to limit the influx of refugees coming to Europe. “We can’t run away from our commitments. Not even Germany,” he declared. Domestic critique came above all from Merkel’s own party, especially from Horst Seehofer, the leader of the CSU and prime minister of Bavaria. Seehofer went as far as threatening to file a complaint of unconstitutionality against the chancellor’s decision to open German borders for refugees (see the case study in Chapter 6).

Domestically, the chancellor defended herself by describing the refugee crisis as a great national duty, comparing it to the challenges posed by the reunification of Germany and drawing parallels between the refugee crisis and the Eurozone crisis. She reiterated her optimistic stance: “I will say it again and again. We can and we will do it.” She also appealed to the German public by appearing in the famous TV talk show Anne Will one Sunday night in early October. Internationally, she originally (in mid-September 2015) appealed to the other member states for help and asked for a joint EU summit, pointing out that there was a need to discuss border controls with Greece and Turkey and to address the conflicts in the countries of origin. She promised that Germany would lead by example, that is, by taking in more refugees than the quota requirement stipulated. In return, she expected that other member states would follow with their more limited means. In a speech before in the European Parliament in early October, she appealed to European values and called for more support for refugees.

It was only after the failure of the relocation scheme in late September that support from Turkey became the crucial plan B for Merkel. On October 18, she traveled to Ankara to meet President Erdoğan to negotiate what was then still called the joint action plan, which had been elaborated by Frans Timmermans, the EU Commission’s vice president. From this point on, she systematically pursued an agreement with Turkey. In November, she intensified her efforts at the G20 meeting in Turkey, where she discussed a “quota solution” with Turkish prime minister Davutoğlu. While the German coalition partners continued to battle over the asylum packages, Merkel called for a concerted action at the European level, pointing out that without the help of Turkey, the number of refugees coming to Germany would not be reduced. On November 24, at yet another European summit on the refugee crisis, European heads of state met with President Erdoğan, and Merkel declared that Turkey would be a key partner in finding a solution to the crisis but failed to find an agreement. In the new year, Merkel pursued the negotiations with Turkey. Thus, she and Turkish prime minister Davutoğlu met with their cabinets in Berlin for the first German–Turkish government consultations on January 22, a meeting that ended without any new resolutions. While the V4 countries, together with Austria and Bulgaria, opted for closing the Balkan route, Merkel continued to single-mindedly bet on a deal with Turkey. Thus, at the EU summit on February 18, she demanded that negotiations with Turkey be continued. Eventually, the EU–Turkey summit in early March was the turning point, and in the final rounds of negotiations in the first half of March, Merkel played a crucial role.

After its adoption, the agreement was criticized by the domestic opposition from the left and by CSOs as well as international NGOs such as Amnesty International. In response to such critique, Merkel again traveled to Turkey. In April, she went to visit a refugee camp on the Turkish border with Syria; in May, she went to meet President Erdoğan. She wanted to provide evidence that the agreement was sensible and working as planned, to reassure the Turkish president of Germany’s commitment to the agreement, and to voluntarily accept additional contingents of refugees. In May 2016, she continued to defend the agreement before the German public on TV, invoking the humanitarian responsibility of the EU.

The Politicization of the EU–Turkey Agreement in Greece

Greece is the other member state where the EU–Turkey agreement has been heavily politicized. Figure 12.2 presents the politicization of the Greek episodes in phase 1. However, even if the agreement was heavily politicized in Greece, its politicization did not reach the level of the politicization of the hotspot episode to which it was closely linked. As we have already observed above, the EU–Turkey agreement episode in Greece was not very conflict intensive and comparatively little polarized. This is not to say that there was no opposition to the agreement: Civil society organizations; the radical left opposition; and even parts of Syriza, the governing party, criticized the implications of the deal for refugees in Greece. But the EU proved to be generally highly supportive, Turkey proceeded to implement its part of the deal, and Greece also received support from other international actors. Domestically, the Greek prime minister defended the agreement, as did the government and the mainstream opposition.

Figure 12.2 Politicization of Greek episodes, phase 1

Table 12.7 presents the salience of the various types of actors in the three Greek episodes during phase 1. The dominance of international actors is striking not only in the EU–Turkey episode, where international actors account for almost two thirds of the actions but also in the two domestic episodes that we have already discussed in the previous chapter. Compared to Germany, the national government and especially the governing parties generally play a more limited role, which again confirms the extent to which Greece was the object of top-down interventions in this first phase of the crisis. As for the Greek prime minister, he is conspicuously present in all the three episodes, although his position is somewhat less prominent than the position of the German chancellor in the two episodes where she was most important.

Table 12.7 The salience of the different types of actors in the three episodes of phase 1: percentages

Actors	Episode
	Summer 2015	Hotspots	EU–Turkey	Total
International	43.1	42.7	64.6	54.5
Tsipras	7.2	8.1	8.8	8.3
National government+	28.8	24.6	13.4	19.2
Government parties	4.6	0.3	0.6	1.1
Opposition–Civil society	16.3	24.3	12.7	16.9
Total	100.0	100.0	100.0	100.0
n	153	321	545	1,019

Phase 1: The Management of the Refugees Trapped in Greece

The Greek debate before the adoption of the agreement was closely intertwined with the creation of the hotspots and of an EU border and coast guard capable of controlling the EU borders between Greece and Turkey. Greece was fighting on two fronts: On the one hand, it was struggling with Turkey, accusing it of supporting people smugglers, with Turkey replying that it was doing what was possible and claiming that it was stopping 500 persons every day. Repeatedly, the Turkish President turned to threatening not only Greece but the European leaders as well that he would flood the EU with refugees if the EU did not offer Turkey a better deal for its support in managing the refugee crisis. On the other hand, Greece was struggling with the other European member states, which reminded it of its responsibilities as a frontline state. When the agreement was eventually reached, it was perceived to be a diplomatic success of Greece (and Cyprus) by Greek prime minister Alexis Tsipras. He claimed that Greece had achieved the best available deal with regard to the refugee issue, but he also warned that the agreement would be difficult to implement and that a key condition for its success would be a reduction in refugee flows. The main opposition party, ND, agreed, calling the agreement a “positive step.” But it added that the agreement had to be implemented effectively, and it had some doubts about the government’s capacity to do so.

Greece’s prime minister warned that the number of refugees crossing the border to Greece could not be limited unless the smugglers on the Turkish side of the border were stopped. EU migration commissioner Avramopoulos asked the EU to increase pressure on Turkey to crack down on smugglers. Turkish and Greek officials serving as liaison officers were installed on both sides to monitor the deal. And Turkey did, indeed, abide by the agreement. As a matter of fact, the number of arrivals dropped sharply after the agreement was signed: While the average number of arrivals was around 2,000 per day in January and February 2016, it fell to 130 in April 2016. If Greece had counted more than 860,000 arrivals in 2015, the number of arrivals dropped to 36,000 in the year after the deal was signed, before climbing again to nearly 75,000 in 2019. In addition, the number of dead and missing migrants in the Aegean Sea decreased from 1,175 cases in the 20 months before the agreement to 310 in the period after its adoption in March 2016 until March 2019.² In spite of a lot of frustration on the Turkish side, this centerpiece of the agreement held.

On the Greek side, after the adoption of the agreement, the debate focused on its implementation, which put great pressure on the country. Economically battered by the Euro area crisis, Greece did not have the capacity to deal with the large number of refugees who were now trapped in the country as a result of the agreement. The hotspots on the islands were no longer open facilities where refugees passed through on their way to northern Europe; rather, the refugees were now confined to these camps. As a result of the agreement, roughly 60,000 refugees were stranded in Greece – in the camps on the islands, in the port of Piraeus, and at the Greek northern border in Idomeni. Overcrowding in substandard living conditions and destitution became an integral part of the asylum process in Greece, contrary to reception obligations and human rights standards of the member states. This situation was heavily criticized by NGOs. Thus, a few days after the conclusion of the agreement, Doctors Without Borders (Médecins Sans Frontières), one of the key nongovernmental organizations helping refugees and migrants arriving in Greece, announced that it would stop all activities linked to the hotspots on the Greek islands of Lesvos and Samos. The NGO said its decision was prompted by its objections to the agreement, which it described as a “cynical mechanism” that jeopardized asylum and showed “contempt” for humanitarian needs. The NGO had also temporarily withdrawn from the refugee camp in Idomeni, this time citing concerns about the safety of its staff. One month later, it was the turn of Oxfam to denounce the European Union for its failure to deliver a fair and safe system for receiving refugees in Greece: “Europe has created this mess and it needs to fix it in a way that respects people’s rights and dignity.” Oxfam highlighted problems at the overcrowded hotspots on Lesvos (Moria), where riots had occurred at that time. In addition, the UNHCR expressed its disapproval and suspended cooperation in harsh terms: “UNHCR has till now been supporting the authorities in the so-called hotspots on the Greek islands, where refugees and migrants were received, assisted, and registered. Under the new provisions, these sites have now become detention facilities. Accordingly, and in line with our policy on opposing mandatory detention, we have suspended some of our activities at all closed centres on the islands. This includes provision of transport to and from these sites. However, UNHCR will maintain a presence to carry out protection monitoring, to ensure that refugee and human rights standards are upheld, and to provide information on the rights and procedures to seek asylum.”³

In order to implement the EU–Turkey agreement for the return of refugees from Greece to Turkey and to speed up the procedures pertaining to asylum requests, Greece’s parliament, under high time pressure, adopted an asylum amendment bill on April 1 that adapted the Greek legislation to the EU directive on asylum procedures. It also introduced provisions for registering refugees, allowing them to find work and to qualify for international protection. In addition, immediately after the adoption of the agreement, Greece had appealed to its European partners for logistic help to implement the deal. In response to such calls for help, the Commission had immediately started coordinating the implementation of the agreement, and the EU border agency Frontex called on the EU member states to provide 1,500 police and 50 readmissions experts. On April 1, approximately 350 Frontex officers from Germany, France, Portugal, Estonia, Latvia, and Lithuania arrived on Lesvos to assist in the readmission process for the refugees and migrants. In the end, 397 police and 47 readmission experts were actually provided.

However, the situation on the islands hardly improved. By the end of May, the mayor of Lesvos, one of the islands most affected by refugee flows, urged the government to speed up the asylum application procedure, as the extended stay of refugees and migrants on the island was causing stress and frictions for applicants. “The delay of the asylum procedures requires de facto that refugees stay in Lesvos for a long time. This creates frustration and friction between our guests, some of which have already turned to delinquency, given the lack of money and fear of their possible readmission to Turkey,” the mayor wrote in a letter to the government. The mayor of another island, Chios, accused the government of ignoring the gravity of the situation, especially after the NGOs had left and nobody knew how to deal with the situation.

While the overall responsibility for managing migration flows in Greece rested with the Greek authorities, the Commission and EU member states continued to provide support to the Greek authorities in the implementation of the EU–Turkey agreement to improve migration management and reception conditions in Greece. EU actions focused in particular on helping to alleviate the situation on the Greek islands. By 2019, over 2.07 billion euro in EU funding had been allocated to Greece to support migration management since the start of 2015, including 816 million euro in emergency assistance and over 643 million euro for projects under the EU Emergency Support Instrument. The European Border and Coast Guard Agency and the European Asylum Support Office deployed staff on the ground in Greece to support the Greek authorities. The Commission also deployed a team in Athens and ensured a permanent presence on the hotspot islands. Since 2016, a permanent Commission representative has been stationed on both Samos and Lesvos to support the Greek and international partners on the ground.⁴

Phase 2: International Protection Bill and Reception Centers

Nevertheless, the situation for refugees in Greece remained tense. Two of the three Greek episodes that occurred within a short time span at the very end of the period covered by our analysis in late 2019/early 2020 and that we introduced in the previous chapter (see Figure 11.1) – the International Protection Bill and the reception centers, once again concerned the management of the refugees in Greece. By the time these episodes took place, the situation in the camps on the Greek islands had hardly improved at all. As pointed out in the previous chapter, these two episodes were dominated by bottom-up cross-level politicization. Together with the Turkey Border Conflict, which we discussed in the previous chapter, they took place against the background of mounting problem pressure, that is, increasing arrivals of refugees, overcrowded refugee camps on the Greek islands, and increasing tensions between Greece and Turkey.

The first of the two interlinked episodes concerns the International Protection Bill (IPB), the first act related to immigration policy adopted by the recently elected New Democracy government. The bill was not directly concerned with border control; rather, it was designed to streamline domestic Greek asylum rules, once again attempting to improve and accelerate the asylum and return processes. Among other things, the bill was intended to relieve the pressure on the islands and to construct new “closed” centers for rejected asylum seekers, who would be confined to these new centers. The domestic debate of the bill was dominated by civil society organizations claiming that it contravened international law and would not work anyway, but to no avail – the bill passed without much ado in parliament. However, resistance to some of the bill’s provisions continued among the islanders in Lesvos, Chios, and other afflicted islands, who were wary of the prospect of getting new “closed” centers in their neighborhoods. Regional and municipal authorities demanded that after five years of shouldering the problem, the easternmost islands should be unburdened from refugee reception. The domestic debate was accompanied by an intense international debate: At the same time as it introduced the bill, the government was trying to entice and contain Turkey and to get support from the other European member states. On the one hand, it accused Turkey of gradually allowing more migrants to slip through its borders to get more concessions from Europe, and it appealed to Turkey for support of a commonly beneficial solution to the problem. On the other hand, it multiplied meetings with representatives of fellow member states in order to raise their awareness of the imminent threat at the border with Turkey and to induce them to share the burden. The other member states responded by providing assurances or by pointing out that the key was to return nonrecognized asylum seekers to Turkey as envisaged under the EU–Turkey agreement.

The domestic conflict with the islanders intensified in the second episode, which was focused on the detention centers on the islands. The regional authorities of the Northern Aegean, where most centers were to be built, adopted a collision course with the national government, engaging in protest participation as well as judicial challenges to the government’s plans. They feared that once built, the centers would sprawl like Moria on Lesvos and consolidate the image of the islands as “migrant barracks.” It did not help that the government decided to expropriate real estate on the islands to build the new centers. The standoff between the government and the islanders culminated in a confrontation of far right and far left groups, each opposing the hotspots for their own reasons, with riot police that had been sent to supervise and protect the start of the construction process. Faced with a sort of low-key guerilla warfare, the government eventually retreated, asking the riot police to return to Athens and promising to delay the construction of the centers and to “consult” with local authorities.

At the same time, the international conflict continued, with the Greek government continuing to fight an international bottom-up battle on two fronts. On the one hand, Greece continued to blame Turkey for using the refugees to blackmail the European Union. The Turkish government responded by criticizing Greece for manipulating the data concerning the refugee crisis, for its inhumane treatment of the migrants, as well as for pushing illegal migrants back to the Turkish borders. On the latter points, Turkey was joined by the UNHCR, which warned that the conditions in the Greek reception centers were awful, asked Greece to make sure that the new asylum procedures were in line with international law, and pointed out that the UN was generally opposed to detention centers for asylum seekers. In addition, in April 2020, Amnesty International and many others documented how Greece systematically used “pushbacks” and other human rights abuses to prevent refugees from entering the EU.⁵ On the other hand, the Greek government both criticized its European partners for their lack of solidarity (e.g., for their lack of willingness to accommodate 3,000 unaccompanied migrant minors) and asked for a reform of the Dublin regulation, as well as for EU support for decongesting the islands, for financing the new accommodation and predeparture units, for border controls by boosting Frontex, and for returning rejected asylum seekers to Afghanistan. In the European Council’s debates on the Multiannual Financial Framework (MFF), which were going on at the time, the Greek government fought for increasing the funds for migration/refugees.

Conclusion

In this chapter, we have shown how an EU policymaking episode is domesticated in national policymaking, and how this works out differently depending on the member state. We have compared the debates in four member states and then zoomed in on the debates in the two member states most concerned. For Germany, this episode was instrumental in solving a domestic conflict between the chancellor and the governing parties, including her own party. Once the agreement was sealed, the German debate did not entirely subside, but its intensity lessened and eventually faded out. The Greek debate, by contrast, picked up shortly before the conclusion of the agreement and then stayed intense during the implementation phase. Several years after the agreement had been concluded, it gave rise to new domestic episodes in Greece, since the problems it created for Greece continued to remain unsolved.

While the EU–Turkey Deal stopped the inflow of refugees, it did not work out as expected in other respects, with important implications for Greece. According to the deal, all refugees who would enter Greece after March 20, and those whose asylum applications were not accepted, would be returned to Turkey on chartered ships. However, the return of refugees to Turkey developed only sluggishly and despite the rapid adaptation of Greek asylum law to the new situation, only 2,441 migrants had been returned three years after the signing of the agreement.⁶ Also, the promise of one-to-one resettlements did not work out as expected: From March 2016 to March 2021, only slightly more than 28,000 Syrian refugees were resettled in the European Union from Turkey, far short of the maximum 72,000 outlined in the deal. Discussions of bringing Turkey into the European Union and easing visa processes for Turks meanwhile mostly stalled, as President Erdoğan’s government increasingly turned authoritarian after the coup in summer 2016. The agreement did not usher in a period of harmonization of EU–Turkish relationships. As we have seen in the previous chapter, in early 2020, Turkey’s president moved to reopen the border for refugees, using Turkey’s geopolitical position as a buffer between Syria and Europe to put renewed pressure on the EU and on Greece in particular.

But the deal succeeded in externalizing a significant part of the management of the EU’s refugee crisis to Turkey. In exchange for stopping the flow of refugees to the EU, it provided Turkey with 6 billion euros to arrange for the refugees if not with the other goods it originally promised. The exchange of funds for the management of refugees, the part of the EU–Turkey Deal that worked, provided a blueprint for other externalization agreements – with Libya and Morocco. Moreover, the New Pact on Migration and Asylum presented by the European Commission on September 23, 2020, assigned a prominent place to cooperation with third countries of origin and transit of migrations flows.⁷ The German presidency progress report on key elements of a European migration and asylum policy stated in 2020 that “action on promoting and advancing tailor-made partnerships with key third countries needs to be taken without further delay and with the aim to show tangible results.”⁸ In the eyes of some critics, however, the new pact only proposed “more of the same,” which they did not consider to be enough to improve the EU’s management of its external borders.⁹

On its fifth anniversary in spring 2021, leaders in both Turkey and Europe suggested that the agreement would endure in some form or another. Commission president von der Leyen and European Council president Charles Michel met with President Erdoğan in Ankara on April 6 and signaled that additional funding for Turkey was forthcoming, as long as the country continued upholding its end of the agreement.¹⁰ In June 2021, a new 3.5 billion euro package for Turkey was on the table of the Commission, to be disbursed until 2024.¹¹ The proposal included an additional 2.2 billion euro for Jordan, Lebanon, and Syria. The commission wanted to “gradually move from humanitarian priorities to socio-economic support and development,” according to the draft text. This would include “funding for migration management and border control,” precisely the areas rights activists and parliament have flagged as being of serious concern. Meanwhile, the reception camps for asylum seekers on the Greek islands of Leros and Kos were almost empty, and on Samos and Chios only a few hundred migrants remained. Only on Lesvos did 5,000 migrants continue to live in a provisional tent camp with a capacity of 8,000 people.¹²