Assessing the role of the European Council and the European Commission during the migration and COVID-19 crises

Abstract

Over the past decades, ‘emergency politics’ has become a quasi-permanent feature of the European Union (EU). According to some, this has reinforced the trend towards a greater role for the European Council (EUCO) in EU agenda-setting, to the detriment of the European Commission (Commission). In this article, this claim is critically assessed by analysing two major crises: the 2015-2016 migration crisis and the COVID-19 pandemic. By systematically tracing the various agenda-setting roles played by EU actors in these crises, two claims are made. First, using a more fine-grained typology of agenda-setting roles, the relationship between EUCO and the Commission is shown to be more nuanced than is often suggested. Second, EUCO and the Commission cannot be considered monolithic players. Instead, actors within these institutions operate outside of formal channels to purse their own policy goals. This puts in doubt the usefulness of focussing on the EUCO-Commission relationship in a purely inter-institutional sense.

Keywords:

• agenda-setting
• emergency politics
• European Commission
• European Council
• 2015–2016 migration crisis
• COVID-19 pandemic

Several observers have noted the increasing role of the European Council (EUCO) in EU politics over the past decades. Whereas initially EUCO was an informal group that met a few times a year to discuss overall developments, it has since the 1990s developed into a central governing institution of the EU (Van Middelaar and Puetter 2022). This is borne out by the fact that many key decisions on EU policies are made in EUCO. In addition, as Carammia et al. (2016) have shown, the EUCO agenda has undergone a change from selective targeting to routine monitoring of issues. As a result, EUCO is increasingly operating as the de facto ‘government’ of the EU where legislation can be pre-negotiated, as in the case when the European Commission’s (Commission) Six-Pack proposals were pre-decided in EUCO’s 2010 Task Force on strengthening economic governance (Moloney and Whitaker 2023).

This new form of intergovernmentalism and its impact on agenda-setting in the EU is increasingly debated within the literature. Some scholars have argued that while the Commission’s agenda-setting powers have been affected by the creation of new positions within ‘the EU’s leadership landscape’, the overall impact is marginal (Nugent and Rhinard 2016: 1211). Others have argued that EUCO is setting the agenda as it is the ‘control room’, with the Commission being part of the ‘machine room’ that implements EUCO’s decisions (Smeets et al. 2019).

The relationship between EUCO and the Commission has been subject to specific scrutiny in the context of crisis situations. Over the past decades, EU politics has been dominated by a series of crises, which tended to absorb a large amount of political attention and led to breakthroughs in European integration. This has increasingly led EU institutions to engage in what Kreuder-Sonnen and White (2022) call ‘emergency politics’: a mode of politics that deviates from established patterns in response to threats that are seen as exceptional and urgent.

Scholars of EU politics have offered different accounts of the impact this pattern of emergency politics has on the agenda-setting roles of EUCO and the Commission. For instance, Dinan et al. (2017) argue that crises have ‘strengthened a tendency towards more intergovernmentalism in the conduct of EU affairs at the expense of supranationalism.’ As a result, this ‘has ensured the elevation of the European Council’, while ‘the political influence of the Commission […] has somewhat receded’ (Dinan et al. 2017: 10–11). In contrast, Dehousse and Magnette (2022) have noted the resilience of the EU’s institutional system and point out that crises have often also resulted in a strengthening of the EU’s supranational institutions (including the Commission).

In a more general argument about the process of European integration, the ‘new intergovernmentalism’ approach has argued that since the Maastricht Treaty European integration has mainly been driven by member states, particularly within EUCO (see Puetter 2013). However, in a critique of the new intergovernmentalism approach, Schimmelfennig (2015: 725–726) argued that throughout the history of EU policy making, new policy areas have often been characterised by intergovernmental policy making in the early stages, only to revert back to more regular patterns, with greater roles for supranational institutions, later on.

What these debates in the literature have in common is that they pitch EUCO and the Commission against each other and look at shifts between these two institutions in terms of more intergovernmental or more supranational patterns of policy making. Particular attention has been paid in this regard to the roles of the two institutions in agenda-setting processes (see Bocquillon and Dobbels 2014). Agenda-setting is a key element of policy making (in the EU and elsewhere), as it determines what issues are taken up and from which angle (cf. Princen 2009: 3ff.). Moreover, both EUCO and the Commission can be seen as prime agenda-setters within the EU: EUCO because, while it lacks formal decision-making powers in most areas, it can set the informal agenda via its ‘conclusions’, the Commission because it has the exclusive right of initiative in the vast majority of EU legislative processes and plays an initiating role in EU policy making more generally. For those reasons, the question of how the two institutions relate to each other in setting the EU’s agenda is particularly relevant especially when both EUCO and the Commission seek to set the agenda.

Thus, this article seeks to offer nuance and a different reading of the ‘Commission vs EUCO’ debate in EU agenda-setting by analysing the overall roles of the two institutions, and actors within them, in shaping the Union’s response to two crises that hit the Union in recent years: the 2015–2016 migration crisis and the COVID-19 pandemic of 2020–2021. In doing so, it makes two central claims. First, it argues that a more refined understanding of each institution’s agenda-setting role can be obtained by distinguishing between different activities within the overall concept of ‘agenda-setting’. By doing so, our analysis extends and further refines other typologies of agenda-setting roles that have been proposed in the literature before (Kreppel and Oztas 2017; Smeets et al. 2019). Second, our analysis problematises the notion of ‘the’ Commission and ‘the’ EUCO as two distinct, monolithic players. Rather, dynamics in the migration and COVID-19 cases show that actors within these institutions can (and will) act on their own, thereby both bypassing these formal institutions and channels and presenting them with faits accomplis that the two institutions later only ratify. In such cases, speaking of ‘the Commission’ or ‘the European Council’ setting the agenda is misleading as to what actually drove the agenda-setting process.

Both points suggest that framing the relationship between EUCO and the Commission in terms of ‘intergovernmentalism vs supranationalism’ offers only a limited understanding of actual agenda-setting dynamics in the EU and their development over time. Thus, we seek to move away from the existing body of literature that has generally focussed on ‘EUCO v Commission’ within the broader debate on ‘supranationalism versus intergovernmentalism’ and present a more complex dynamic of the relationship between the two institutions, and the roles of the actors within them.

The remainder of this article is structured as follows. In the next section we introduce our theoretical framework. We then outline our research design. This is followed by the case analyses of the two crises. We end with a discussion on the nuanced relationship between EUCO and the Commission in responding to the two crises and the role of different actors in pursuing their own policy goals outside of the formal channels within these institutions.

Theoretical framework

The literature on EUCO-Commission relationships during crises has offered different assessments. The prevailing view in the post-Lisbon crisis literature has been one of decline for the Commission while EUCO has moved to the ‘centre of the decision-making stage on crisis related issues’ (Nugent 2017: 169). Much of the scholarly debate has focussed on whether deliberative intergovernmentalism (Puetter 2012) or new intergovernmentalism (Smeets and Beach 2020) can explain the current policy-making dynamics between EUCO and the Commission. Puetter (2012) advances a framework of deliberative intergovernmentalism where policy co-ordination is shifted to intergovernmental forms and away from the supranational sphere. Thus, the focus of deliberative intergovernmentalism is on EUCO and the Council of Ministers (Council) rather than on the Commission.

Smeets and Beach (2020) argue that EUCO rather than the Commission is the agenda-setter in the new post-Lisbon governance framework. Here the heads of state or government (the Heads) in ‘the control room’ set the agenda through placing issues on the agenda and provide the necessary political momentum to drive issues forward (Smeets and Beach 2020: 1139). This shift in roles has also been noted by other scholars. In setting the agenda, EUCO can signal to the Commission the preferences of the Heads and ‘task’ the Commission (Deters and Falkner 2021: 16). A by-product of this expansion of EUCO’s agenda-setting power is the limited room of manoeuvre that the Commission now has if the institution seeks to play a political role in policy making that challenges the consensus within EUCO (Fabbrini and Puetter 2016: 491).

These assessments fit into what Bocquillon and Dobbels (2014) call a ‘principal-agent’ model for understanding agenda-setting relations between the two institutions. To this principal-agent model, they juxtapose a ‘joint agenda-setting’ model, in which the two institutions are much more dependent on each other. Whereas the principal-agent model is premised on a hierarchical relationship, in which one institution is in command and the other takes its orders, the joint agenda setting model stresses the collaborative element in agenda-setting by the Commission and EUCO. Such a pattern of mutual dependence has been found by Smeets and Beach (2022) in the reform of the EU’s migration and asylum framework. Studies on the response to the COVID-19 pandemic have so far primarily focussed on the EU’s economic and health response to the crisis. They highlight the roles of both the Commission and EUCO in the process (e.g. Vanhercke and Verdun 2022; Kassim 2023).

The literature offers support for both the principal-agent and the joint agenda-setting model. It is likely, however, that the relationship between EUCO and the Commission is not fixed but varies across issues and situations. Although overall institutional and political developments form an important background to the relationship between the two institutions, agenda-setting processes are also crucially shaped by events and specific political constellations (Princen 2011). It is therefore important to look for the conditions under which a specific relationship between the two institutions occurs.

As a starting point, it is important to acknowledge that the general notion of ‘agenda-setting’ includes a range of different activities. Building on the agenda-setting literature, we can distinguish between four types of activities that all fall under the heading of ‘agenda-setting.’ For a nuanced assessment of the agenda-setting roles by the Commission and EUCO, one needs to specify which of these roles is played by whom at what point in time. Based on that, it can be determined how the agenda-setting activities of the two institutions related to each other around a given issue.

The first agenda-setting activity is raising attention to issues. Attention is the most basic currency of agenda-setting. In fact, ‘agendas’ are generally defined in terms of attention, as an agenda consists of issues that receive (serious) attention (Cobb and Elder 1972: 86; Kingdon 2003 [1984]: 3; Princen 2009: 19). In this context, an ‘issue’ is a situation that is seen as a problem – something that needs to be addressed by government action. This first activity therefore relates to attempts by actors to draw attention to a certain problem.

The second agenda-setting activity that we discern is defining problems. The point here is the way in which an issue is defined as problematic. This is a process in its own right, as has been shown by studies on the politics of problem definition (Rochefort and Cobb 1994; see also Kingdon 2003 [1984]: 109ff.; Stone 1989). For instance, within the EU the issue of genetically modified food crops has been defined alternatively in terms of an environmental problem (genetically modified crops may reduce biodiversity), a health problem (genetically modified foods may pose risks to human health) and an economic problem (how to capitalise on the technology’s potential to increase the competitiveness of European agriculture). One and the same issue (genetically modified crops) can therefore be defined in different ways, with different implications for policy making (Princen 2021: 172).

In many cases, raising attention to an issue goes hand in hand with defining that issue, as actors usually seek to raise attention to an issue in certain terms, e.g. as an environmental or rather an economic problem. Nevertheless, the two are not identical. In some cases, debate over the most appropriate definition of a problem occurs after the problem has been identified. This is particularly likely to be the case in crisis situations, when policy-makers are confronted with a problem they do not immediately fully understand. Hence the importance of ‘sense making’ during crisis situations (Boin et al. 2005). Moreover, even if an actor raises attention to an issue in certain terms, the problem definition may shift afterwards. Being able to raise attention to an issue does therefore not imply being able to control the problem definition that eventually prevails in the political debate.

While the two agenda-setting activities we discussed above relate to problems, the third and fourth agenda-setting activities relate to the definition of solutions. The third agenda-setting activity is defining overall policy approaches. This concerns the choice between broad lines of action. An actor that is successful in this regard is able to define the broad outlines of the policy approach to be taken and (thereby) to rule out certain policy options. For instance, in the sovereign debt crisis the definition of broad policy approaches included whether or not to bail out member state governments, whether or not to issue Eurozone bonds, whether or not to move towards a banking union and whether or not to tighten EU-level oversight of national budgets.

The fourth and final agenda-setting activity consists of formulating specific policy proposals. This activity is about coming up with concrete proposals that are ready for adoption. In Kingdon’s (2003 [1984]) multiple streams approach, formulating these types of proposals is the key activity in the ‘stream of solutions’ or ‘policy stream.’ This is to be distinguished from the definition of overall approaches, as the two activities may well be carried out by different actors and institutions. Whereas some institutions operate at the level of ‘macropolitics’ (Baumgartner and Jones 1993: 21), which is concerned with choices in the overall political system, other institutions are part of ‘policy subsystems’ (Ibid.) which deal with specific policy issues. Broad choices may be made by macropolitical institutions, leaving the detailed specification of policy proposals to policy subsystems, although in some cases one and the same institution may be active on both levels. In the analysis of our cases, we assess whether the Commission and/or EUCO played these various agenda-setting roles.

Research design

The migration crisis and the COVID-19 pandemic are two of the main crises the EU faced during the past ten years. Both were unprecedented and required swift political responses. As a result, they offer good material to study the way emergency politics affects the agenda-setting roles of EUCO and the Commission. At the same time, the two crises differed in relevant respects. To begin with, issues of migration and asylum had become EU competencies in the 1990s (Zaun and Ripoll Servent 2021), whereas health largely belonged to the remit of the member states (Princen and Rhinard 2006). Therefore, both the authority and the capacity of EU institutions to deal with these issues was higher during the migration crisis than in the midst of the COVID-19 pandemic. As Princen (2011) has argued, this affects the agenda-setting challenges that actors face in the EU. Moreover, the substantive issues at stake were, obviously, different in the two crises. These differences allow us to assess whether agenda-setting dynamics in the EU are similar for both crises or (partly) specific to a particular crisis. When the analysis of the COVID-19 pandemic is added to the migration crisis, this leads to what Yin (1994: 46) called a ‘theoretical replication’ logic behind the selection of cases: by studying multiple cases that belong to the same universe of cases but also show contrasts, one can study whether theoretical assumptions hold in different contexts.

The cases themselves are studied through the method of structured, focussed comparison put forward by George and Bennett (2005). It is structured because the questions we ask of each case are the same for both the migration and COVID-19 crises. These questions follow from our typology of agenda-setting activities, which guided and structured the case analyses. It is focussed because with our theoretical framework we zoom in on a specific aspect of the two crises, i.e. the agenda-setting roles of EU actors in them. Within each case, we start by identifying the key issues that were at stake. We then map for each agenda-setting activity what actors played a role and what or whose approach prevailed in the end (George and Bennett 2005: 67).

For our case studies we apply process tracing, a research method that allows for the tracing of ‘causal mechanisms using detailed, within-case empirical analysis of how a causal mechanism operated in real-world cases’ (Beach and Pedersen 2019: 1). Process tracing also has analytical value in so far as causal inferences can be made about how processes work in for example the policy response to a crisis (Ibid). Thus, process tracing provides us with the appropriate frame in which to present a more nuanced understanding of the agenda-setting dynamics between EUCO and the Commission through our detailed reconstruction of the two crises, and to further our understanding of the actors within the two institutions that shaped the agendas of EUCO and the Commission. An added advantage of process tracing is that it allows this study to assess whether the same actors – Commission Directorate-Generals (DGs) and various policy units of the General Secretariat of the Council, member states, Council Presidencies, senior officials and leadership figures within EUCO and the Commission – played the same roles during the same stages in the four agenda-setting activities of our framework across two case studies. In this way, we can identify the relative importance of the Commission and EUCO in each of these activities.

The substantive issues that we focus on in the two crises are informed by novel data collected from key-informants during interviews. In total, 22 elite in-depth semi-structured interviews were conducted with key-informants from the Commission and EUCO in the policy areas directly affected by both crises. They were interviewed using a variety of video conferencing platforms e.g. Webex between the 16 February and the 29 September 2021. The interviews lasted between 45 and 60 min. Of the 22 in-depth interviews with key-informants, 11 were with senior bureaucrats on the shaping of different aspects of the Union’s response to the migration crisis e.g. relocation and the 2016 EU-Turkey statement, while the remaining 11 interviews were with high-ranking officials on how the EU addressed the challenges of the COVID-19 pandemic in the areas of health, transport, borders and the economy. The key-informants were recruited for this research on the basis of their direct involvement in formulating the policy response to one of the two crises. Only senior bureaucrats who were directly involved in formulating the policy response to one of the crises were interviewed. As such, owing to the fact that the officials interviewed for this research continue to work in the EU institutions and due to the sensitivity of the background information provided, the anonymity of all of the key-informants was guaranteed from the outset of the data collection stage. Thus, the ‘Chatham House Rule’ was in force during the interviews. To ensure anonymity each interview with the key-informants is therefore referred to as a code in text e.g. EC1. The code corresponds to the number of the interview in chronological order in the text for each crisis, the institution of the official (sans specific DG), the date of the interview and the video conferencing platform. A list of interviews is provided in the Online Appendix.

We opted for in-depth interviews for this study as the key-informants were able to provide details for the four stages of the agenda-setting activities of EUCO and the Commission, which are not available when conducting an analysis of primary documents, such as EUCO conclusions. However, aware of the pitfalls that come with been over-dependent on interview data only (see Yin 2018: 119), we have used primary documents and secondary sources where possible and available. An overview of these documents and sources is included at the end of this manuscript.

The 2015–2016 migration crisis¹

Those seeking refuge and a new life from the turmoil unleashed by the ‘Arab Spring’ (Webber 2019: 135), began to cross the eastern and central Mediterranean in large numbers in early 2015, and by October a peak of 221,1721 persons arrived in front-line member states (United Nations High Commissioner 2022²). By the end of 2015 the number of arrivals began to decrease sharply before being cut even further with the implementation of the EU-Turkey Statement in March 2016 (Khan and Fleming 2020). At the onset, the migration crisis was defined by a humanitarian response in the form of the relocation mechanism, and then later a security approach as laid out in the Valletta summit and the EU-Turkey Statement. Driving these responses were EUCO, the Commission, two Presidencies of the Council, and various member states, DGs and the European External Action Service (EEAS).

Raising attention to issues

The increasing numbers of people crossing the central Mediterranean route was noted for the first time by EU leaders in March 2015 (EUCO 2015a). It was however the sinking of a migrant boat with the loss of 700 people on board, off Lampedusa on the 19 April, which brought the issue of migration to the attention of EUCO. At the extraordinary summit of EUCO on the 23 April, the Heads committed to several orientations of which two would become the most salient as the crisis unfolded: (1) preventing illegal migration flows through greater co-operation with African countries to combat smuggling, and with Turkey on the migratory flows from Syria and Iraq; and (2) considering emergency and voluntary options of relocation of migrants from frontline member states. These orientations would be closely monitored by EUCO. To provide a more systematic approach to the challenge of migration, the Commission had been tasked by EUCO to table a ‘European Agenda for Migration’, with the Commission having to report back to the Heads by June 2015 (EUCO 2015b). At the 23 April and the 25-26 June summits, the text of the EUCO summit conclusions had avoided the use of the term ‘crisis’ (EUCO 2015b; EUCO 2015c), however at the 23 September summit the migration wave of April - August 2015 was referred by the 28 Heads as a ‘refugee crisis’ (EUCO 2015d). This was echoed by President of EUCO Tusk (Crisp 2015) and First Vice-President of the Commission Frans Timmermans (Gotev 2015a). Only in October would President of the Commission Juncker eventually concede that the migration wave had developed into ‘an unfolding emergency’ (Euractiv 2015a).

Defining problems

The increasing number of arrivals into the EU was defined as a problem which required either a humanitarian approach or a security approach. From the onset of the emerging crisis, the Commission at the political level in the Berlaymont and on the ground in the front-line member states took a humanitarian approach. At the political level, the College supported the humanitarian approach, allowing Dimitris Avramopoulos, Commissioner for Migration, Home Affairs and Citizenship, to take this line when defining the policy approach through the European Agenda on Migration (EC1; Commission 2015a). However, there was a split between the political level and the technical level within the DG for Migration and Home Affairs (DG HOME) with Avramopoulos supporting the humanitarian approach and officials in favour of a security approach (EC1; EC2). The split within DG HOME was also evident in EUCO during the April - August 2015 period. Sweden and other likeminded member states were initially resistant to using the term ‘illegal migration’, while EUCO, with the exception of Hungary, supported a humanitarian approach through strengthening the Triton and Poseidon search and rescue operations in the Mediterranean in the April summit (EUCO 2015b; EuroCouncil1). Conversely, Hungary, after Lampedusa, was going so far as to define the developing crisis as one of border protection (EuroCouncil1). Of the two approaches, the security approach prevailed as the one that defined the EU’s response to the crisis. A combination of an unsustainable number of people arriving after Germany suspended Dublin III on the 5 September (Alkousaa et al. 2016), and the refusal by Juncker to change the Commission’s humanitarian Agenda on Migration (Gotev 2015b), created the perception among member states that the College and the political level of DG HOME were too focussed or even obsessed with the centre piece of the Agenda on Migration – relocation, while failing to give the returning of illegal arrivals or border security the necessary political attention, – and united EUCO around a security approach (EC3; EuroCouncil1).

Defining overall policy approaches

Before the shift towards a security approach, relocation was the main policy response within the humanitarian approach. In line with the 23 April EUCO orientations (EUCO 2015b), the Commission proposed its Agenda on Migration in May 2015 (Commission 2015a). While the Agenda on Migration contained several elements, implementing the relocation of arrivals from front-line member states to other member states based on a fixed number increasingly became the policy response to the crisis for the Commission at the political level (EC4; CM1; Gotev 2015c). Relocation as a humanitarian approach prevailed due to two factors. First, EUCO was not unified on the approach to take on the crisis (EC4; EuroCouncil1). Hungary, and later, Poland, the Czech Republic and Slovakia (Gotev 2015c; Gotev 2015d) were concerned about relocation as they considered it a pull factor, however there was wide support for the proposal among front line member states, with Greece, Italy, Malta and Cyprus pushing the policy, as these member states could not cope with the high number of arrivals and were not in a position to return these migrants (EC5; CM2; Euractiv 2015b). While Germany wanted to move people who were travelling through the Western Balkan route in a more organised fashion via relocation (CM3; EC3; Euractiv 2015b). Within EUCO there was a lack of solidarity for the front-line member states, thus the Commission had to show support by pushing hard for relocation (EC4; CM1; CM3), with Juncker stating that ‘No country should be left alone to address huge migratory pressures’ (Reuters 2015).

Within EUCO the nature of the conversation had sharply shifted to a security approach from September 2015. Member states that had supported the humanitarian approach such as Sweden were submerged with applications for asylum (Jacobsen 2015) and there was now a realisation in EUCO that this was ‘not just a migration crisis, but a crisis that threatened the very existence and purpose of the EU’ (EuroCouncil1) and the Schengen area (Tusk 2015a). Within this context, a number of the Heads floated the idea that there was a need to talk to African countries on irregular migration. Such a move had been one of the orientations agreed by EUCO on the 23 April (EUCO 2015b), however as the number of arrivals had increased from September so too had support among member states to stem the flows of migrants and increase the rate of returns in the EUCO summits on the crisis in September, October and November (EUCO 2015d, 2015e, 2015f). EUCO therefore agreed to hold a summit in Valletta in November 2015 (Gotev 2015e), with President Tusk’s cabinet tasked with drawing up the agenda for it in conjunction with the member states and the EEAS (EuroCouncil2; International Centre for Migration Policy Development 2022). Thus, by the sixth EUCO summit on migration held on the 17–18 December 2015, EUCO was increasing focussed on a security approach through making the commitments made at the Valletta summit operational on returns and the EU-Turkey agreement operational (EUCO 2015g).

The Valletta summit provided long-term solutions to a crisis that required immediate action. For the security approach to have an impact there was a realisation within EUCO that there had to be deals with third countries to stop the flows. EUCO had on the 23 April agreed that co-operation was needed with Turkey on the migratory flows from Syria and Iraq (EuroCouncil1; EUCO 2015b). Austria, France, Germany, and the Netherlands had been pressuring the Commission to engage with Turkey on closing the route down as the crisis escalated (EuroCouncil1; EC5). Germany in particular had emerged as the main destination for migrants, and with the Dutch Presidency of the Council, began to conduct secret negotiations with Ankara on a more ambitious deal than the one Tusk was finalising with Turkey (Barker and Spiegel 2016). The large sums of EU funding required to secure a deal with Turkey on stopping the flow across the Eastern Mediterranean under the German-Dutch proposal, meant that not only was the Commission present in the negotiations, but Juncker was also updated every evening by phone (CM3; EC6). However, notwithstanding the presence of the Commission and Juncker’s involvement, the main drivers during the negotiations were the Dutch Presidency and Germany (EC6; CM1; Barker and Spiegel 2016). When the deal was announced a few hours before the EUCO summit scheduled for the 18-19 February 2016 (EUCO 2016a), it was presented as a fait accompli to the majority of EUCO (CM3; EuroCouncil1). Despite the surprise, EUCO, and especially Tusk, very quickly took ownership of the proposed deal when the Heads rejected the German-Dutch proposal (Stupp and Gotev 2016) in an informal EUCO summit on the 7 March, after the EUCO-Turkey summit (Tusk 2016). In doing so, the proposed deal would not only have a stronger impact, but Tusk and EUCO ensured that a common EU approach to stemming the flows across the Eastern Mediterranean, rather than a Dutch and German one prevailed at the EUCO summit on the 17-18 March (CM2; EUCO 2016b).

Formulating specific policy proposals

The Commission as early as the 13 May 2015 signalled its intent to ‘table a legislative proposal by the end of 2015 to provide for a mandatory and automatically-triggered relocation system’ in its Agenda for Migration (Commission 2015a). Following on from this commitment, the Commission in its proposal for a Council decision on the 27 May to relocate 40,000 persons from Italy and Greece, and its subsequent recommendation on the 8 June on a European resettlement scheme for a further 20,000 arrivals had stated that all³ member states should participate (Commission 2015b, 2015c). EUCO on the 25-26 June, while agreeing that all member states would participate, stated that the distribution of the 60,000 persons would need to be done by consensus. As more technical work was needed on the proposals, EUCO agreed that the proposals should be moved to the JHA Council configuration (EUCO 2015c). Here decisions could be taken by qualified majority voting (QMV) rather than by unanimity, allowing for Hungarian and Polish opposition to the relocation mechanism to be bypassed (EuroCouncil1; EC4). Aware that the Commission’s humanitarian approach to the crisis, along with the interests of front-line member states and member states that were the main destinations for migrants could be addressed all at once, Juncker’s head of cabinet and chief of staff, Martin Selmayr ignored the advice of the technical experts in DG HOME and increased the proposed number of those who were to be relocated by threefold to 120,000 in the 22 September proposal (Ibid). The Luxembourgish Presidency under pressure from the Commission adopted the September proposal by QMV on the 22 of that month (Ibid; Council of the EU 2015).

Unlike the relocation proposals, a broader spectrum of actors was involved in formulating a key aspect of the 11-12 November Valletta Summit on Migration, the returns policy. In drawing up the agenda for the Valletta summit, it quickly became clear within Tusk’s cabinet that returning irregular migrants was key for a number of member states who wanted a strong deterrence in order to stem the flows of people entering the EU. However, the EEAS and the DG for International Co-operation and Development (DEVCO) argued for a package approach as the EU could not go into the summit demanding returns without supporting re-integration programmes. Supporting DEVCO and the EEAS were Ireland, Luxembourg, and Spain who already had pre-existing links with third countries through development contributions (EuroCouncil2; CM1). Tusk’s envoy for Valletta, Pierre Vimont (Tusk 2015b), developed the five-basket approach which incorporated both sides of EUCO, and the arguments of DEVCO and the EEAS in the first domain ‘Return, readmission and reintegration’ of the Joint Valletta Action Plan (International Centre for Migration Policy Development 2022).

Germany’s efforts to formulate a policy on stemming the flow across the Eastern Mediterranean route was making a prospective EU-Turkey deal unacceptable (EuroCouncil1; Stupp and Gotev 2016). There were two problems with the proposed deal. First Germany, which had experienced a huge influx after it had adopted an open border policy in September 2015, was willing to concede to all of Turkey’s proposals, including in areas that were extremely sensitive to Cyprus, in order to stem the flow (Euractiv 2016). Second, the failure of Germany and the Dutch Presidency to consult the rest of the member states, especially Cyprus, caused deep displeasure among a number of the Heads at the informal EUCO summit on the 7 March (Baczynska 2016; Stupp and Gotev 2016). To ensure that all the member states could agree to the proposed deal, and that Cypriot interests were protected, Tusk brought the proposed deal into the framework of the 28 (Barker and Spiegel 2016; Crisp 2016). Thus, Tusk ensured that an EUCO approach to securing the borders through stemming the flows, rather than a Dutch and German one prevailed when the EU heads adopted a common position at the second summit on the deal on the 17-18 March (EuroCouncil1a; CM1; EUCO 2016b).

Evaluation of EUCO and Commission influence during the migration crisis

When evaluating EUCO and Commission influence we observe several notable findings within the four agenda-setting activities of our framework. First, on raising attention to issues we observe that the increased arrivals into the EU were taken up at the EUCO level with EUCO then tasking the Commission to operationalise the orientations agreed by the Heads at the 23 April summit. Thus, at this stage of the agenda-setting process, the relationship between EUCO and the Commission is hierarchical. Second, on defining problems, the Commission – the political level of DG HOME, and Juncker, considered the crisis a humanitarian one, while EUCO viewed the crisis as a security issue post-September after a surge of arrivals, following on from the suspension of Dublin III by Germany, unified the member states. Only in this period does the EUCO-Commission relationship revert back to a hierarchical relationship from a more balanced approach between April-August 2015. Likewise, on the third agenda-setting activity, defining overall policy approaches, the relationship between the political level of DG HOME, and Juncker, and EUCO, before September, was balanced due to the divisions within EUCO as frontline member states and countries of destination supported the Commission’s relocation decisions. However, when Tusk and his cabinet stepped in to define the summit goals of Valletta, the EUCO-Commission relationship became more hierarchical. Similarly with the EU-Turkey Statement and Action Plan, Tusk would play a crucial role in defining the overall approach of the eventual deal, after the initial attempt by the Dutch and the Germans to define such an agreement was rejected by the Heads. This balance in the relationship between EUCO and the Commission continued into the final stage of the agenda-setting process, formulating specific policy proposals. Here the political level of the Commission – DG HOME, Juncker, and later Selmayr, shaped the content and eventual relocation decisions. This was later pushed through with the assistance of the Luxembourg Presidency. On the formulation of proposals at the Valletta summit there was a range of actors involved including Tusk’s envoy Vimont, DEVCO, the EEAS, and member states such as Ireland, Luxembourg, and Spain shaping the ultimate policy, while the Netherlands through holding the Presidency, Germany, and later Cyprus were important actors in formulating the initial, and final agreement with Turkey. Tusk also played a key role in ensuring that the interests of EUCO, rather than the Netherlands and Germany, were reflected in the eventual deal. Thus, our findings suggest that the EUCO-Commission relationship was hierarchical at the initial stage of the crisis – raising attention to issues and evolved into a balanced relationship between April-August 2015 in the remaining three agenda-setting activities – defining problems, defining overall policy approaches and formulating specific policy proposals. However, from September onwards the EUCO-Commission relationship became more hierarchical in the three stages – defining problems, defining overall policy approaches and formulating specific policy proposals as the crisis worsened.

The COVID-19 pandemic

The outbreak of the COVID-19 pandemic in Italy on the 31 January 2020 presented EUCO and the Commission with a health crisis that Europe had not faced in over hundred years (Amante 2020). As member states began to close their borders in March in order to control the spread of the virus, the EU was faced with both a health crisis, and potentially another economic crisis. In response, the EU sought to keep the Single Market open through implementing the Green Lanes initiative, save the summer holiday season through the EU Digital COVID Certificate (DCC), offset the economic fallout by approving the Recovery and Resilience Facility (RRF), and re-open the Union via a vaccine strategy. Leading the response on these issues was the Commission and several DGs. However, as health (Princen and Rhinard 2006) and border control are national competences (Zaun and Ripoll Servent 2021), member states exercised significant influence in formulating the DCC and the vaccine strategy, while the Franco-German axis played a crucial role in establishing the RRF.

Raising attention to issues

The Commission was the first institution to raise awareness of COVID-19 when the DG for Health and Safety (DG SANTE) opened an alert notification on the Early Warning and Response System on the 9 January 2020 followed by convening the Health Security Committee on the spread of the virus in China on the 17 of that month (Commission 2022). Before the first of a series of EUCO March summits, the new President of the Commission Ursula von der Leyen established a ‘coronavirus response team’ at the political level to co-ordinate the EU’s response to the developing crisis in the areas of health, transport, economy and borders on the 2 March (Commission 2020a). COVID-19 first appeared on EUCO’s agenda on the 10 March 2020 when the Heads by video conference agreed on a number of priorities of which three would remain salient throughout the crisis: (1) limiting the spread of the virus; (2) promoting research, including research into a vaccine (later evolving into the distribution of the vaccine); and (3) tackling socio-economic consequences. EUCO also stressed the need for close co-operation with the Commission in fighting the pandemic (EUCO 2020a).

Defining problems

The COVID-19 pandemic from the start was defined as both a health crisis and an economic crisis by the Commission and EUCO (Commission 2022; EUCO 2020a). The first two problems facing the EU – freedom of movement and transport, contained elements of both, while the inoculation drive and the Union’s economic recovery were dealt with as sperate issues. The first problem in the initial phase of the crisis was how to stop the spread of the virus. The Commission sought to curtail the entry of nationals from third countries by proposing on the 16 March a ‘temporary restriction on non-essential travel to the EU for a period of 30 days’ (Valero and Michalopoulos 2020; EUCO 2020b), which increasingly became a problem for mainly southern member states with substantial tourist sectors, who sought an approach that would restore freedom of movement and save their upcoming tourist season (CM1a; EC1a).

Member states followed up the Commission’s 16 March guidance by closing their borders in order to halt the spread of COVID-19. In doing so the member states had curtailed the freedom of movement of road freight, resulting in a knock-on effect across the Union’s economies (Neslen et al. 2020). To address this second problem, the Commission issued guidance on how to operate its Green Lanes initiative, the guidelines of which were issued on the 23 March (Commission 2020b; EUCO 2020c). The aim of the initiative was to keep road freight moving as freely as possible (EC2a; CM2a).

The third problem facing the Union was the socio-economic consequences of the crisis. In order to mitigate these consequences of the crisis, a mechanism was needed to allow the Commission to raise funds to alleviate those challenges (Macron 2020).

The fourth problem facing the EU was the slowness of the inoculation drive in the initial phase of the vaccine rollout from January - April 2021 within a number of member states, which AstraZeneca was increasingly been blamed for by EUCO and the Commission (Blenkinsop and Strupczewski 2021; Peel et al. 2021). The decision by AstraZeneca to de facto sell their vaccine twice, first to the EU and then to the UK, and then prioritise shipments to the UK hampered the vaccine rollout in Austria, Bulgaria, Czechia, Croatia, Estonia, Latvia, Slovakia and Slovenia (Euractiv 2021a; Khan 2021). These member states had placed more orders for the Oxford/AstraZeneca vaccine than the BioNTech/Pfizer one, which became a problem when shipments of Oxford/AstraZeneca to the EU were delayed (EuroCouncil1a).

Defining overall policy approaches

In order to preserve the balance between the economic interests of member states and fighting the pandemic, Greek Prime Minister Kyriakos Mitsotakis on the 12 January 2021 floated a Green Pass (Mitsotakis 2021; Michalopoulos 2021), which was later taken up by the Commission’s Directorate-General for Justice (DG JUST) on the 17 March (Eccles et al. 2021) after EUCO tasked the executive to work on the issue on the 26 February. The Green Pass was considered a solution to preserving the balance between the economic interests of member states and fighting the pandemic (EUCO 2021a).

Whereas the Green Pass originated in the Council, Green Lanes was an initiative launched by the Commission on the 23 March 2020 to remind member states that the EU still had free movement despite the pandemic. Von der Leyen took a strong and personal lead with the Green Lanes initiative by focussing on all freight rather than essential freight such as food and personal protection equipment (CM2a; EC2a). EUCO endorsed the Commission’s Green Lane initiative on the 26 March (EUCO 2020c), with the caveat that they were only recommendations (CM2a; EC2a).

The recovery element of the EU’s response to the crisis was initially defined by France and Germany in their proposal for an EU recovery fund of €500 billion worth of grants on the 18 May (Macron 2020; Goßner and Lawton 2020). Ahead of the Franco-German announcement, EUCO on the 23 April tasked the Commission to ‘urgently come up with a proposal’ for a recovery fund (EUCO 2020d). The Commission tabled its proposal to establish the RRF on the 28 May (Commission 2020d). The Commission since the Financial crisis had sought to create a mechanism like the RRF with a centralised fiscal capacity for quite some time (EC3a; CM3a). The Frugal Four were also involved in defining the core elements of the RRF – in particular the balance between loans and grants in the fund, after having lost the fight on the loan only option (CM3a; Grüll 2020).

The Commission initially defined the response to the vaccine shortfall through proposing an export control mechanism for vaccines – a mechanism to clarify the movement of vaccines leaving the EU on the 29 January 2021 (Vela 2021; Commission 2021a). When the Commission dropped this proposed measure on the 29 January (Herszenhorn and Vela 2021), EUCO then dealt with vaccine shortfall on the 25 March and called for COREPER to factor in the ‘the spirit of solidarity’ when allocating 10 million BioNTech/Pfizer doses (EUCO 2021b). COREPER on the 1 April agreed to a ‘solidarity’ mechanism that was to provide 2.85 million additional doses of the BioNTech/Pfizer to member states in need (COREPER 2021).

Formulating specific policy proposals

The Commission was reluctant to wade into the Council discussions on the Green Pass, having had faced difficulties with member states such as Greece flouting the Council recommendations on the temporary restrictions on non-essential travel into the EU from Israel (CM1a; Michalopoulos 2021). Having waited for EUCO to adopt a common position on Mitsotakis’ proposal for a Green Pass at the 21 January summit, DG JUST then tabled a proposal for the DCC on the 17 March (Commission 2021b). DG JUST recognised that the DCC could only bring about greater coherence on border checks, as the Commission was aware that it was the sole right of the member states to introduce such checks (EC1a; Commission 2021b). During the negotiations on the DCC in the JHA Council between March and May 2021, there was great reluctance among Belgium, France, Germany, Sweden, Denmark, the Netherlands and Romania, which stated that it would be impossible to explain to their citizens that only those who were vaccinated were eligible for the DCC (CM1a; Eccles et al. 2021; Pollet et al. 2021; Brzozowski 2021). A compromise was reached on the 26 May in the JHA Council to secure an agreement between the member states which extended the DCC to citizens who had recovered from COVID, had been vaccinated or who had tested negative (CM1a; CM4a; Commission 2021b).

While the overall policy approach on Green Lanes was driven by Von der Leyen, the policy was formulated by Adina-Ioana Vălean, Commissioner for Transport and the DG for Mobility and Transport (DG MOVE) (EC2a; Valean 2020). The aim of the Green Lanes initiative was to reduce the time for checks or health screenings on road hauliers to 15 minutes to reduce delays at the borders of the member states (Commission 2020c). DG MOVE was effectively forced to include testing in the initiative after France, and later Germany, started implementing testing regimes for truck drivers when both member states closed their borders on the 16 March (France 24 2020a; 2020b), which the DG did not have the tools to stop (EC2a; CM2a). DG MOVE tried to push the initiative further by bringing a legislative aura to the concept, however it remained a Commission communication as member states were somewhat sceptical of Green Lanes and did not want to have their hands tied in fighting the pandemic (CM2a; EC2a).

The Commission’s intervention in the negotiations on the RRF were relatively straightforward compared to the other two policy areas after France and Germany had proposed a €500 billion EU Recovery Fund in May (Macron 2020). In doing so, the two member states provided the opening for the Commission’s Secretariat-General to table its proposal for the RRF on the 28 May (CM3a; Commission 2020d). The substance of the negotiations on the Commission’s proposal were conducted in COREPER II rather than in the Economic and Financial Affairs configuration (CM3a; Council of the EU 2021). The Commission’s initial proposal for the RRF had placed more weight on grants over loans (Commission 2020d: 40), with the Secretariat-General arguing that grants would provide more fiscal space for the member states that would be hit the hardest from the crisis (EC4a; Valero 2020). Initially, the Frugal Four opposed any grants, which was rejected by the non-Frugal member states at the EUCO summit on the 19 June 2020 (EUCO 2020e; Herszenhorn et al. 2020a). In an effort to reach an agreement on the RRF, EUCO President Charles Michel called for an extraordinary summit for the 17 and 18 July (Baume 2020). At the summit, the Frugal Four changed tactics moving to oppose the RRF outright despite Michel’s proposal for a smaller 2021-2027 Multiannual Financial Framework (Bayer 2020), with the Austrians and the Dutch particularly strong in their opposition (CM2a; Herszenhorn et al. 2020b). Unable to maintain their position due to the need to take action, and with Germany and France keen to secure an agreement, the Frugals agreed to the RRF on the fourth day of the summit − 21 July in exchange for lowering the ratio between grants and loans, in favour of the loan aspect (CM5a; Rios and Morgan 2020; Herszenhorn et al. 2020c).

Conversely it was the Commission and EUCO that ultimately formulated the specific policy to address the slow delivery of the Oxford/AstraZeneca vaccine. The initial attempts by the Commission at curbing exports of vaccines were not successful after the Von der Leyen cabinet on the 29 January 2021, without consulting the member states, pushed the control mechanism through by written procedure first via the College and then later through Comitology (EuroCouncil1a; Valero 2021b). In doing so the Commission triggered Article 16 of the Withdrawal Agreement (EC5a; Valero 2021). Despite this early misstep, DG SANTE and DG TRADE’s ‘COVID-19 vaccines export transparency and authorisation mechanism’ was triggered by the Draghi government on the 4 March in order to prevent the export of 250,700 doses of the Oxford/AstraZeneca vaccine to Australia (Ministry of Foreign Affairs and International Cooperation 2021). The Commission’s export control mechanism therefore proved to be an important component in addressing the shortfall of vaccines over the longer term.

However, in the short term the slow deliveries of the Oxford/AstraZeneca vaccine were causing domestic problems for an Austrian coalition government that was fighting for its survival. Unable to fall back on the formal contract positions, Vienna decided to fall back on the solidarity argument to see whether it would generate the extra doses that it needed (CM5a; EuroCouncil11a; Euractiv 2021b). In order to strengthen its position, Austria recruited Bulgaria, Croatia, Czechia, Latvia and Slovenia heading into the EUCO summit on the 25 March (Khan 2021; Euractiv 2021b; Kurz et al. 2021). Here EUCO confirmed the ‘pro-rata population key for the allocation of vaccines’, while inviting COREPER ‘to address the issue of the speed of deliveries of vaccines’ (EUCO 2021). COREPER was tasked by EUCO to address the shortfall after the Commission had failed to do so through its proposed control mechanism (EuroCouncil1a; EUCO 2021c). Due to the salience of the issue, the negotiations were dealt with in COREPER II, despite health falling under the responsibility of COREPER I (CM5a; EuroCouncil1a). COREPER II was brought into the process by EUCO after member states complained about a lack of transparency in the Commission’s vaccine steering board, which was in charge of managing the contracts and negotiations with vaccine developers (EuroCouncil11a; Herszenhorn et al. 2021). COREPER II on the 1 April agreed to a ‘solidarity’ mechanism that was to provide 2.85 million additional doses of the BioNTech/Pfizer to Bulgaria, Czechia, Croatia, Estonia, Latvia, and Slovakia. Austria, Czechia and Slovenia gained no extra allocations of vaccines under the solidarity mechanism (COREPER 2021).

Evaluation on EUCO and Commission influence during the COVID-19 pandemic

Using the four agenda-setting activities of our framework when assessing the influence of EUCO and the Commission in this case study we discern the following conclusions. First, on raising attention to issues, we find a more balanced relationship between EUCO and the Commission, with the Heads not only stressing the need for a joint European response, but explicitly calling for close co-operation with the executive. This co-operative relationship continues in the second stage of agenda-setting – defining problems, where both EUCO and the Commission consider the pandemic to be both a health crisis and an economic crisis. Both EUCO and the Commission held AstraZeneca responsible for the slow rollout of the EU’s vaccine strategy during the initial phase of the campaign while the executive had carte blanche in defining the problem of closing borders on the freedom of movement of goods in the Single Market. Conversely, on the impact of restricting the entry of third country nationals into the EU, southern member states with large tourism sectors became increasingly concerned for their 2021 summer season, while France and Germany argued that there had to be an EU response to the economic shock of the pandemic. The relationship between EUCO and the Commission began to shift increasingly towards a hierarchical one in the third agenda-setting activity, defining overall policy approaches. On saving the 2021 tourist season EUCO, based on a Greek initiative, tasked the Commission to develop a Green Pass. Likewise, EUCO tasked the Commission in April 2020 to urgently develop a proposal for a recovery fund. While the Commission briefly defined the initial response to the shortage of AstraZeneca vaccines, it would be EUCO that shaped the eventual policy approaches. Only with the Green Lanes initiative, did the Commission maintain the freedom to shape the overall policy approach to the backlog of heavy goods vehicles on the borders of member states. In the fourth and final stage of agenda-setting – formulating specific policy proposals, the EUCO-Commission relationship maintained its balance within the four policy areas. Of those four policy areas, the Commission – DG MOVE and Vălean, only had sole influence over one, the Green Lanes initiative. On formulating the DCC and the RRF the Commission – DG JUST and the Secretariat-General respectively shared responsibilities with the JHA Council, COREPER II, Michael, and the Frugal Four on shaping the eventual policy. Likewise, on the eventual policy to address the shortage of vaccines, the Commission – DG SANTE and DG TRADE through its ‘COVID-19 vaccines export transparency and authorisation mechanism’ and EUCO, and later COREPER II via the ‘solidarity’ mechanism dealt with the vaccine shortfall. Our findings therefore indicate that in first two stage of our typology – raising attention to issues and defining problems, the EUCO-Commission had a co-operative relationship which developed into a hierarchical one in defining overall policy approaches and formulating specific policy proposals.

Discussion and overall conclusions

Based on the case analyses, a number of conclusions can be drawn. Some of these relate to the relationship between EUCO and the Commission as institutional actors in EU agenda-setting. In addition, the two cases show dynamics that go beyond this strictly inter-institutional dynamic but are important for understanding the EUCO-Commission relationship.

To start with the former, in the two cases different patterns can be discerned in the relationship between EUCO and the Commission, which also vary over time in each case. Depending on the issue and the political constellation surrounding it, the relationship oscillated between more co-operative (with EUCO and the Commission working in tandem) and more hierarchical modes (in which EUCO set the overall agenda, which the Commission subsequently developed further).

In terms of the conceptual framework, that we introduced above, EUCO was typically involved in raising attention to issues, defining problems and (sometimes) defining overall policy approaches. The Commission, by contrast, played the greatest role in defining problems, defining overall policy approaches and formulating specific policy proposals. The area of overlap therefore lies in the definition of problems and of overall policy approaches. Here, the balance between the two institutions varied, depending on who took the initiative. Specifically, during the migration crisis, both EUCO and the Commission took the initiative during different stages. In the initial phase of the crisis, while EUCO raised attention to the issue, the Commission defined the problem, the overall policy approaches and the specific policy proposals within a humanitarian frame. As the crisis deteriorated, initiatives were based on EUCO’s priorities of increased security. Conversely, during the COVID-19 pandemic both EUCO and the Commission raised attention to the problems of the pandemic while both EUCO and the Commission took the initiative when defining problem. In the latter two agenda-setting activities – defining overall policy approaches and formulating specific policy proposals – the Commission’s ability to take the initiative was increasingly curbed by EUCO.

In a number of ways, agenda-setting dynamics showed features that are familiar from the literature. The rough division of labour between EUCO and the Commission, whereby the former establishes the broad lines and the latter drafts specific proposals, tended to be clearer during the migration crisis than amid the COVID-19 pandemic, even though both ‘sides’ were also engaged in the other type of activity. Moreover, the ‘competitive co-operation’ that Bocquillon and Dobbels (2014) observed in the relationship between EUCO and certain member states on the one hand and the Commission on the other was stronger during the migration crisis than the COVID-19 crisis. Likewise, differences between DGs are often important in understanding the Commission’s role in agenda-setting. At the same time, this pattern did not hold in all instances and was reversed in some.

This puts into question the usefulness of framing the debate on the institutional balance in terms of ‘EUCO’ versus ‘Commission’ as distinct, monolithic institutions. In many cases these institutions do not act as uniform bodies but instead exhibit marked differences between their component units (DGs, individual Commissioners and their close advisers in the Commission, and Council presidencies, and member states in EUCO). The decisive agenda-setting push may then come from one of these component units rather than the institution as a whole. This confronted other actors in those institutions with faits accomplis and agenda dynamics that were not of their own making and can therefore not usefully be seen as the result of activities by ‘EUCO’ or ‘the Commission’ as a whole.

Our analysis therefore seeks to offer a more nuanced contribution to the debate on EUCO-Commission relations. On the one hand, it argues for a more fine-grained conceptualisation of ‘agenda-setting’ by looking at specific agenda-setting activities by these two institutions. This shows that the two institutions are engaged in partly different and partly overlapping activities, which moreover shift over time. On the other hand, it offers an antidote to an exclusive focus on the EUCO-Commission relationship in purely inter-institutional terms (e.g. Smeets and Beach 2022). In both cases, both the Commission and EUCO in the end were confronted with, and had to acquiesce to, an agenda-setting dynamic that were set in motion by some of their component parts and that therefore escape a conceptualisation of EUCO and the Commission as single, monolithic bodies.

Acknowledgements

The authors would first like to thank the key-informants for participating in this study. The authors would also like to thank the anonymous reviewers for their very helpful comments during the review process. Their comments undoubtedly strengthened the article. Finally, the authors would like to thank the editors and the editorial staff of the journal for their guidance during the review process. David Moloney would especially like to thank Professor Daniele Caramani from the European University Institute for his support during the data collection stage.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

David Moloney

David Moloney is a Post Doctoral Research Associate in the Economic and Social Research Council-funded project entitled ‘Strategic development of the UK-EU relationship: Can neighbours become good friends?’ at Open University. His research focuses on EU decision-making in times of crises. [[email protected]]

Sebastiaan Princen

Sebastiaan Princen is Associate Professor at Utrecht University’s School of Governance. His research focuses on policy making in the European Union. His books include Agenda-setting in the European Union (2009, Palgrave) and The Politics of the European Union (with Herman Lelieveldt, 3rd ed., 2023, CUP). [[email protected]]

Notes

1 A timeline with the key events in both cases can be found in the Online Appendix.

2 A detailed list of non-academic sources can be found at the end of this manuscript.

3 With the exception of Denmark, Ireland and the United Kingdom.

4 List of EU primary documents and EU websites listed here in order of how they are cited in the main text.