Introduction: EU external migration policy and EU migration governance: introduction

ABSTRACT

While the EU’s external migration policy arguably constitutes the most dynamic strand of EU migration policy, especially since the 2015/2016 asylum crisis in Europe, and while there is also a growing body of research on this dimension, more comprehensive, conceptually informed social science publications analysing this policy area in a more encompassing fashion are scarce. In an attempt to make progress on this front, this special issue focusses on four sets of issues: (1) conceptualising EU external migration policy; (2) identifying the drivers and conditioning factors of EU policy; (3) analysing interdependencies and interactions between different policy areas and instruments; and (4) tracing the influence and reactions of third country actors and non-European host countries of migration. After describing key developments and aspects of EU external migration policies over time and after reviewing the state of the art in research on EU external migration policy, the key contributions of the individual papers are identified and contextualised.

KEYWORDS:

• European Union
• external action
• migration policy
• migration governance
• remote control

Introduction

This special issue seeks to take stock of EU migration governance, with a particular focus on the European Union’s (EU) external migration policy. While there is yet no unified definition of what the EU’s external migration policy is, we refer to it in this special issue as any policy that aims at managing migration outside the territory of EU member states. This includes various policies of extraterritorial migration management such as the externalisation of European borders to the Sahel region, but also restrictive border practices, return and readmission agreements and Mobility Partnerships with third countries as well as multilateral aid programmes which are either provided in exchange for migration control or designed as remedies for the ‘root causes of migration’. While the internal dimension of EU migration policy was originally more pronounced, the external aspect has gradually evolved into a central pillar. Since the asylum crisis in 2015/2016, it has arguably become the most dynamic strand within EU migration policy. The impact of the war in Ukraine, setting off the fastest mass migration in Europe since World War II, also has an important external dimension, of course, and may reinforce the trend of integrating external migration into comprehensive EU foreign policy action (across various policy nexuses). EU migration governance, on the other hand, includes EU migration policies but exceeds them and encompasses also the unintended effects of European migration policies. These more informal means of EU influence may for example follow from other countries reacting to EU policies by either backing or opposing them. As opposed to formal ways of EU influence through polices, informal influence through migration governance may not unescapably have an impact on migratory movements to the EU (cf. section 4 for more detail).

Despite the growing body of research on both the internal and external dimensions of EU migration policies over the years, the more comprehensive volumes dealing with external aspects of this policy tend to primarily take stock of the existing policies and lean towards legal analysis (e.g. Carrera et al. 2018; Kassoti and Idriz 2022). The Social (especially Political) Science literature on the external dimension – in terms more encompassing publications such as edited volumes or special issues – does not seem to have adequately followed suit developments at the political level with regard to analysis and conceptualisation of this EU policy.

This special issue seeks to fill this gap by addressing the following broader research questions: (1) How may (certain aspects of) EU external migration policy and politics be conceptualised (Bergmann and Müller 2023; Brumat and Freier 2023; Czaika, Bivand Erdal, and Talleraas 2023; Tittel-Mosser 2023; Cusumano and Riddervold 2023; Vaagland 2023; Zaun and Nantermoz 2023; Stutz 2023; Cardwell and Dickinson 2023)? (2) Which considerations and preferences drive EU external migration policies pre- and post-2015? What conditions these policies and what is their nature, i.e. are they open or rather restrictive (Bergmann and Müller 2023; Cardwell and Dickson 2023; Cusumano and Riddervold 2023; Stutz 2023; Zaun and Nantermoz 2023)? (3) To what extent and how have EU objectives in terms of migration control affected other policy areas and fostered the integration of different policy areas (Bergmann and Müller 2023; Zaun and Nantermoz 2023)? (4) Going beyond more Eurocentric perspectives: How may third country actors influence policy processes and outcomes (Vaagland 2023; Tittel Mosser 2023) and how do other regions of the world and migrant host countries outside Europe perceive EU external migration policy (Brumat and Freier 2023)?

We proceed as follows: In section 2 we specify the evolution of EU external migration policies over time and its intensification in the context of the 2015/16 crisis. Section 3 reviews the literature on EU external migration policy, both in terms of the main research questions asked and with regard to the most important conceptual approaches used. Finally, we discuss the broader conceptual, empirical and methodological contributions of this special issue to research on EU external migration policies.

The evolution of EU external migration policies over time and its intensification in the context of the 2015/2016 crisis

Both internal and external EU migration policies have evolved since the Tampere European Council of 1999. The EU’s external action in the area of migration is largely based on the 2005 Global Approach to Migration and Mobility, the basic framework for EU migration policy relations with third countries, and the Commission’s (2015) European Agenda on Migration, which updates several external relations aspects in this area. The perception of an asylum or refugee crisis accelerated the externalisation of migration policies and the implementation of new, or the reinforcement of old, measures.

Since the beginning of its cooperation on external migration policy, the EU has developed a host of bilateral cooperation frameworks (such as readmission agreements, the Mobility Partnerships, the Common Agendas for Migration and Mobility, the EU-Turkey deal, etc.), regional cooperation frameworks (such as the Rabat Process and the Khartoum Process), and an intercontinental framework (the Africa-EU Partnership on Migration, Mobility and Employment). Furthermore, the dramatic increase of deaths at sea in the context of increased arrivals of Syrian refugees in 2015 entailed the adoption of several policy instruments, e.g. the EU Emergency Trust Fund for Africa (EUTF Africa), the Lebanon and Jordan Compacts, EU-Libyan cooperation on border control and CSDP missions in the Mediterranean for migration control purposes as well as an expansion of Frontex’s role and resources. Especially in view of the latter points, the crisis may also be understood as a policy window that allowed the approval of policy measures that could have otherwise been proven too controversial for adoption.¹

Against this background, we will now give an overview of key EU policy responses to the 2015/2016 crisis, including both those with older roots and those that were adopted from scratch.

Externalising migration control and linking it to the EU’s Common Security and Defence Policy

As a response to the crisis, migration control increasingly became streamlined in other areas of EU foreign policy. The Council (2015a) agreed to extend the scope of the EU Capacity Building Mission (EUCAP) Sahel Niger to support Nigerien authorities in the prevention of irregular migration. EUCAP Sahel Niger is a mission under the EU’s Common Security and Defence Policy (CSDP) to train national police and military forces for the fight against organised crime and terrorism. After this change of the mission’s mandate, EUCAP Sahel Niger introduced an outpost in Agadez, a major transit city for migrants from Western Africa on their way to Libya and hence potentially to Europe.

This new focus on preventing migration through Agadez ultimately resulted in a Nigerien law criminalising any kind of migration, including travel of Nigerien citizens or citizens of other member states of the Economic Community of West African States (ECOWAS) who normally have these rights (Tubiana, Warin, and Saeneen Mohammud 2018, 22–23). This approach has led to major criticism because it does not take into consideration nomadic migration patterns in the region and has diminished the possibilities of many of those previously generating income from transportation in Agadez. This is expected to contribute to even more instability in an already fragile region, hence undermining a key goal of EUCAP Sahel Niger.

In June 2015, the Council (2015b) also launched the EU naval operation against human smugglers and traffickers in the Mediterranean (EUNAVFOR Med) – later renamed ‘Operation Sophia’ – which focused on the surveillance, assessment and dismantling of human trafficking networks by searching and seizing suspicious vessels. This operation was largely modelled after the Italian search and rescue operations Mare Nostrum and the CSDP mission Atalanta (Boşilcă, Stenberg, and Riddervold 2020). Critics rightly highlight that the EU’s focus on smugglers and especially traffickers had been mainly the result of increased criticism of its member states’ push back actions on the high seas and its own criminalisation of migrants crossing the Mediterranean. Moreover, the increased engagement in search and rescue cannot disguise the fact that migrants who are rescued at sea have no systematic access to protection and human rights (Franko Aas and Gundhus 2015).

The use of CSDP missions for migration prevention purposes is part of a trend that dates to pre-crisis times: In May 2013, for example, the EU and Libya agreed on an EU Integrated Border Assistance Mission in Libya (EUBAM Libya). The main activities under this mission include the training and support of Libyan border guards, police and customs (Council of the EU 2013). As a result of that mission, Libyan border guards regularly intercept migrant boats that are leaving Libyan shores for Europe. The close cooperation between the EU and Libya in this area has received strong criticism because of reports of severe human rights violations by Libyan authorities, including open slave trading (Baker 2019; Human Rights Watch 2019).

Reinforced border protection and search and rescue

We have already highlighted the problematic role of the EU in the area of search and rescue when addressing naval CSDP missions. But this topic has also been repeatedly discussed in the context of border protection under Frontex. Frontex, the European Border Agency, has been introduced in 2005 (Council 2005). At the beginning, the agency had a purely supportive role and a budget of roughly six million Euros. This budget expanded dramatically over time and had reached almost 544 million Euros in 2021 (Frontex 2021). At the same time, Frontex’ role was considerably expanded. While it initially assisted member states in border protection at sea and on land, in search and rescue at sea and in the return of irregular migrants, e.g. by organising joint return flights, the adoption of the European Coast and Border Guard (ECBG) Regulation in 2016 (Council of the EU 2016) and its reform in 2019 (Council of the EU 2019) significantly enhanced its role. Nowadays, Frontex is a fully-fledged agency with its own mandate that can enforce EU borders even against the will of member states (Niemann and Speyer 2018). While Frontex previously mainly relied on border guards from the member states, it will count 10,000 EU border and coast guard officers by 2027. Léonard and Kaunert (2022) argue that the 2015–16 crisis has led to an intensification of Frontex’s security practices, which increasingly include taking part in the collection of intelligence information and cooperating with security organisations like Europol and NATO.

Return and readmission agreements

The EU has been trying to conclude return and readmission agreements with various third countries since the early 2000s. So far, it has concluded such agreements with Hong Kong (2004), Macao (2004), Sri Lanka (2004), Albania (2006), Russia (2007), Ukraine (2008), North Macedonia (2008), Bosnia and Herzegovina (2008), Serbia (2008), Moldova (2008), Pakistan (2010), Georgia (2011), Armenia (2014), Azerbaijan (2014), Turkey (2014), Cap Verde (2014), and Belarus (2020). Additionally, non-binding arrangements have been made with Afghanistan, Guinea, Bangladesh, Ethiopia, Gambia, and Ivory Coast (European Commission 2021). Moreover, EU member states have struck their own ‘deals’ that aim at the return and readmission of migrants. Cases in point include bilateral deals that Italy has concluded, among others, with Libya (Tasch 2015), or Spain’s readmission agreement with Morocco (Cassarino 2007). A case of a multilateral deal is the EU-Turkey Statement (2016), according to which migrants arriving in the EU from Turkey were to be returned in exchange for a funding of six billion Euros to support migrants in Turkey and the same number of refugees from Turkey being resettled in Europe. Although this deal is often referred to as the EU-Turkey Statement, it had no legal quality under EU law and was merely an intergovernmental agreement between EU member states and Turkey with no oversight of the European Parliament or the European Court of Justice. Scholars have therefore argued that the EU-Turkey deal is a case of strategic venue-shopping in which member states avoid EU channels of policymaking to circumvent liberal veto players (Slominksi and Trauner 2018).

Readmission agreements are often adopted in exchange for financial or other support from the EU. Many readmission agreements, for example, have been concluded as part of Mobility Partnerships which facilitate other, non-irregular forms of migration. While the EU has always – and particularly after 2015 – been keen to agree on return and readmission agreements with countries in Northern and Western Africa, these attempts have remained unsuccessful so far. Such agreements often go against the interests of these countries, which have a strong concern for non-interference in sensitive areas of national sovereignty (Wolff 2014; Zardo and Loschi 2020). In addition, many of them are migrant-sending countries and benefit from remittances from their citizens living in Europe, which explains why support for such agreements is low among their publics (Reslow and Vink 2014). Likewise, especially in Northern Africa, there is little appetite for taking back Sub-Saharan migrants who have simply transited through these countries (Wolf 2014).

Development aid and trade cooperation

Linking migration policies and development aid is not new in the EU but dates back at least to 2005. At the time, Spain experienced increased immigration by boat on the Canary Islands, but also in the Spanish exclaves of Ceuta and Melilla. Especially in the latter case, Spain would forcibly prevent the entry of immigrants, which led to international criticism. EU member states at the time suggested that migration control should become more streamlined with development aid, especially as it was often considered the most attractive ‘carrot’ that the EU could offer to sending and transit countries (Lavenex and Kunz 2008). Hence, cooperation on migration was another precondition for receiving such aid that the EU introduced. This led to frustration among recipients and low levels of compliance (Betts and Milner 2006).

In the context of the 2015/2016 crisis, this approach was levelled up. One result of this is the adoption of the EU Emergency Trust Fund (EUTF) Africa in November 2015. The EUTF Africa aims at addressing the root causes of instability, forced displacement and irregular migration and at improving migration management in Africa. 251 projects worth 4.9 billion Euros have been adopted under the EUTF. These projects aim at contributing to four main pillars and objectives: greater economic and employment opportunities; strengthening the resilience of communities; improved migration management in countries of origin, transit and destination; and improved governance and conflict prevention and reduction of forced displacement and irregular migration. These programmes are implemented in 26 African countries and across three regional windows (Sahel and Lake Chad, Horn of Africa, and North of Africa) (European Commission 2022a). While previous policies have mainly streamlined migration control cooperation as a pre-condition for development aid, the EUTF Africa aims at preventing migration through targeted development support that seeks to create jobs and opportunities for migrants. However, it is unlikely that this approach will work, as there might be other drivers of migration (Zaun and Nantermoz 2021; Nyberg Sørense, Van Hear, and Engberg-Pedersen 2003).

Western countries have often adopted an implicit responsibility-sharing approach by advocating the protection of refugees in neighbouring countries in exchange for financial support. The EU-Jordan and EU-Lebanon Compacts are recent examples of this approach. Through the EU-Jordan Compact, for instance, the EU has provided loans, grants and preferential trade agreements in exchange for access to public education and legal employment for Syrian refugees (Vaagland 2021, 52; European Commission 2016). While the Compact has been praised for being innovative, it has also been criticised. For instance, the Jordanian government often issued work permits in areas that were not attractive to Syrian refugees and jobs in the informal market were often much better paid. Thus, many of the 200,000 work permits that the Jordanian government had promised where never taken up (Howden, Alfred, and Patchett 2017).

Legal migration channels

While EU policymakers have often highlighted the need to create more legal migration channels to destroy the business model of human smugglers and traffickers, the EU has not adopted any meaningful legal migration channels for other than either highly qualified migrants or seasonal workers to date. The exception are Mobility Partnerships. These partnerships are usually based on political declarations and provide the framework for dialogue and practical cooperation on migration and mobility issues. Participation in them is strictly voluntary (European Commission 2022b). Access to legal migration channels in Mobility Partnerships is usually provided in exchange for return and readmission programmes. The legal channels provided consist mainly in access to circular migration programmes which hire temporary workers in understaffed sectors, such as care or hospitality. Yet, as all legal migration in these programmes tends to be circular, i.e. requiring a return to the country of origin, this does not constitute a true channel to permanent migration. Overall, scholars have therefore suggested a ‘conceptual overstretch’ of Mobility Partnerships (Parkes 2009).

The literature on EU external migration policy: the state of the art

This section serves a two-fold purpose. First, we sketch the (development of the) main research questions raised in EU external migration policy, along with key contributions on those issues. Second, we provide an overview of the most important theoretical debates and conceptually informed studies in this field.

Main research questions and issues

The early phase of EU migration policy research has mainly focused on the institutional evolution and gradual communitarisation of EU immigration and asylum policy (Den Boer and Wallace 1994; Monar 1998) as well as the dynamics explaining this development, for example in terms of ‘venue-shopping’ (Guiraudon 2000) or ‘spillover’ (Niemann 2006, 2008). Such research employing EU integration theories has been revitalised in the context of the 2015/2016 crisis (Niemann and Speyer 2018; Slominski and Trauner 2018; Zaun 2018).

A core, if not the core concern of this early phase was whether EU migration policy is security-driven and restrictive as opposed to human rights-oriented and liberal (e.g. Lavenex 2001). Especially in the external dimension of the EU’s migration policy, scholars agree that policies continue to be rather restrictive and security-focused, thus undermining the liberalisation achieved in the internal dimension in recent years (Helbling and Kalkum 2017; Slominski and Trauner 2018).

Since the early 2000s, when externalisation efforts increased, research on the EU’s externalisation of migration control has mushroomed. It is closely investigating the EU’s attempts to export classical migration control instruments to third countries and to facilitate the return of illegal migrants and rejected asylum applicants to third countries (Boswell 2003, 622; Lavenex 2006; Mc Namara 2013). Related to this, research has analysed the impact that the EU has had on third countries’ migration policies (Freyburg 2012; Nizhnikau 2017). In more conceptual terms, this has been termed ‘Europeanisation beyond the EU’ (see below).

Less frequent but growing in recent years are analyses that focus on ‘prevention’ or the ‘root causes’ of migration to the EU (e.g. Boswell 2003; Wunderlich 2013a). Yet again related are accounts which examine the different policy nexuses linked to migration and how this affects the EU and third countries, such as the migration-security nexus (Huysmans 2006), the migration-development nexus (Chou 2006; Lavenex and Kunz 2008), the migration-trade nexus (Panizzon 2017), as well as the nexus between migration, trade and development (Tapinos 2000).

In view of the limited ‘hard’ law regulating EU external migration policy, legal scholars have analysed the growing body of ‘soft’ law, i.e. non-binding (‘new’) governance measures used in this area. In this context, it has been noted that many instruments fall outside the scope of legislative processes and thus democratic scrutiny (Cardwell 2018) and that there may be accountability gaps, e.g. with regard to Frontex operations (Rijpma 2019). Not least given its nature as soft law, scholarship has devoted increasing attention to the (lack of) implementation of EU external migration policy (Wunderlich 2013b; Tittel-Mosser 2020).

Established debates also concern the (deficient) coherence or cohesion of the EU’s (external) migration policy (Parkes 2009; Wunderlich 2013b). Taking this one step further, several authors have analysed the lacking success (Reslow 2017), functioning (Seeberg 2017), or effectiveness in terms of goal-attainment (Slominski and Trauner 2018) of EU action in this policy field. Closely related are studies that uncover the unintended consequences of EU external migration policy, such as the high death toll resulting from the extremely risky passage migrants face on their way to Europe, due to the almost complete absence of legal and safe migration routes (Burlyuk 2017; Reslow 2019).

Despite a continued focus on the EU, some scholarship in this area has started to move beyond a strictly Eurocentric perspective. For example, studies have analysed the position of third countries in dealing with the EU (Betts and Milner 2006) or scrutinised why third countries choose to (not) take part in EU initiatives (Reslow 2012), emphasising that third country actors are not merely policy-takers but policy agents that may substantially affect outcomes (Wunderlich 2013a; El Quadim 2014; Stock, Üstübici, and Schultz 2019; Fakhoury 2021). Analyses have also increasingly looked at how and to which degree factors largely exogenous to the European integration project have impacted on EU external migration policy, including the Arab Spring (Richey 2013), the Covid-19 pandemic (Turner 2021) or the rise of right-wing parties (Bergmann, Hackenesch, and Stockemer 2021).

In addition, while the term ‘European refugee crisis’ in the sense it is used in media and public discourse to describe the events of 2015/2016 mainly suggests a crisis driven by the large inflow of immigrants, scholarly literature considers the crisis to be at least to a considerable extent an endogenous crisis of the Common European Asylum System (CEAS) as well as a political crisis (Guiraudon 2017; Niemann and Zaun 2018). Analyses of the EU’s responses to the crisis are manifold (Lavenex 2018; Scipioni 2018).

One aspect that has received only moderate attention so far, is the relationship between the EU and international organisations in migration policymaking. The EU’s ties, for example with the International Organization for Migration (IOM) or the UN High Commissioner for Refugees (UNHCR), have mostly been analysed in terms of implementing EU policies and characterising the relationship and cooperation (Geiger and Pécoud 2010; Lavenex 2015; Den Hertog 2017). Similarly, little scholarship has focused on trans-regional governance, i.e. ‘sets of formal and informal institutions that cut across and connect different geographical regions’ (Betts 2011, 25). Betts (2011) has done some work on EU relations with the East African Community (EAC), the Southern African Development Community (SADAC) and the Intergovernmental Authority on Development (IGAD), while Bisong (2018) has analysed the EU’s cooperation with the Economic Community of West African States (ECOWAS). Also quite rare are studies comparing the EU’s migration policy with that of other entities, such as the United States or Mercosur (Koff 2017; Acosta Arcarazo and Geddes 2014; Brumat and Freier this volume).

Most important conceptual approaches and theoretical debates

Widely employed for conceptualising EU external (migration) policy are theories of securitisation. In the literature, two schools are broadly distinguished. The Copenhagen School of security studies (e.g. Buzan 1991) builds on constructivist thinking in that migration is not considered a security issue per se, but one that is rather discursively constructed (Buonfino 2006; Karyotis 2007). The Paris School, in turn, moves beyond discourse and concentrates on security practices, such as institutional procedures, new (surveillance) technologies and administrative routines (Bigo 2002). Securitising practices that have enhanced migration control in EU external migration policies, such as the European Border Surveillance System (EUROSUR) or the operation of Frontex, are used to substantiate and illustrate this reasoning (Huysmans 2006; Balzacq 2008; Leonard 2010). Some studies caution against too much faith in the general line of argument in the securitisation literature, for example by finding little evidence of such discourse or practices in EU migration policy in the aftermath of 9/11 (Boswell 2007). Other authors suggest that phenomena in EU external migration may rather fit a somewhat different logic, namely the logic of risk (e.g. Beck 1986). Following this logic, threats are more ambiguous and response measures rather long-term, institutionalised, preventative, and tending towards the normalisation of hitherto exceptional practices (Schmidthäusler and Niemann 2014). Neal (2009:, 348) uses the concept to describe the ‘institutionalisation of normalisation in the form of European Union technologies and practices’, while Van Munster (2009, 144) identifies an ‘institutionalisation of a permanent exception’ in his analysis of EU migration policy.

While initial research on EU migration policies has often drawn on EU integration theories, scholars focusing on its external dimension since the early 2000s have done so to a much smaller extent. However, in the context of the 2015/2016 crisis and the host of policies adopted, integration theories have experienced a renaissance in the field. For instance, Niemann and Speyer (2018) make use of neofunctionalism to account for the (functional spillover-induced) reform of the European Border and Coast Guard in response to the crisis. A somewhat complementary argument has been made by Meissner (2007), who draws on historical institutionalism and explains Frontex’s extended role mainly in terms of path-dependent agency-empowerment.² Liberal intergovernmentalist approaches can mainly be found in studies using concepts such as ‘venue-shopping’ or ‘multi-level games’. Guiraudon (2000) has argued that national home affairs ministries started to cooperate on migration policymaking at the EU level in order to sidestep the liberal constraints they encountered in the domestic context, for example through the need to compromise with other ministries, the scrutiny of courts and the activity of pro-migrant interest groups. As the supranationalisation of EU migration policy progressed in subsequent years, Lavenex (2006) suggested that ministries of the interior sought to maximise the advantages of this trend by moving to a new venue, i.e. shifting migration policymaking to the foreign policy arena. This trend seems to have intensified with the 2015/2016 crisis (Slominski and Trauner 2018). In addition, Reslow and Vink (2014) show how the strategies of actors at three different levels (national, EU, international) can either enable or undermine the adoption of agreements with external partners.

A framework that emerged in the EU external policy literature is that of EU ‘actorness’, often defined as the ability to function ‘actively and deliberately in relation to other actors in the international system’ (Sjöstedt 1977, 16). Bretherton and Vogler (2006) suggest that actorness consists of three sub-concepts, i.e. presence (the ability to shape expectations, understandings and behaviour of others), capability (structural prerequisites enabling external action) and opportunity (the external context). Emiliani (2017) finds varying levels of EU actorness across the policy domains of migration, asylum and external border management across time. Moreover, Dobrescu and Schumacher (2018) suggest that contested statehood in the case of Georgia has adversely affected EU actorness in the field of mobility and migration, especially vis-à-vis the two unrecognised entities, Abkhazia and South Ossetia. Reslow (2015), in turn, draws on a broader conceptualisation of actorness that also includes criteria developed by Jupille and Caporaso (1998), such as autonomy and cohesion. She reveals a mixed record of EU actorness concerning its international migration governance.

Drawing on debates as to whether the EU is a ‘normative’ (Manners 2002) or a ‘market power’ (Damro 2012), Lavenex (2019) finds that the normative power perspective, which is based on the vocation to universal (liberal) values advocated by the European Parliament and the Commission, and the market power identity, based on the traditional regulatory approach revolving around economic priorities advocated by the Commission’s Single Market-oriented directorates, are recurrently countered by the ‘statist’ identity focusing on control and security that has been advocated by member governments in the Council. In their analysis of the EU-Turkey deal, Gürkan and Coman (2021) suggest that the European Parliament – and initially also the Commission – framed the EU’ response corresponding to the Normative Power Europe image. By contrast, the European Council and the Council – which were eventually supported by the Commission – acted in accordance with the civilian power image, resting on economic and diplomatic cooperation to satisfy security references rather than normative concerns.

Europeanisation and diffusion beyond Europe are also the subject of important conceptual debates in EU external migration policy research (Lavenex and Uçarer 2004). In countries for which EU accession may be considered a rather unlikely prospect, such as Albania and Turkey, the EU’s impact on domestic migration polices was found to be weaker than in the Polish case, and yet weaker in countries with no enlargement perspective like Morocco and Ukraine (Ette and Faist 2007; Icduygu 2007; Kruse 2007). Wunderlich (2012) studies the implementation of external EU migration policies in Morocco and Ukraine, thus focusing on the implementation side of Europeanisation. His findings suggest that bottom-up processes are most important in the implementation of these policies. A more recent comparative study using the Europeanisation lens suggests that the EU has to some extent (and partly unintentionally) influenced the migration policies of Turkey and Morocco in terms of externalising control to these non-member transit countries (Yildiz 2016).

The implicit assumption in the literature on diffusion is that the EU acts as a diffuser of ideas, norms, policies and institutions and that other countries and regions may, to varying degrees, emulate the EU (Lavenex and Uçarer 2004). Few authors have conceptualised the EU’s migration relations with other regional organisations in terms of diffusion. Bisong (2018) suggests that some degree of policy diffusion from the EU towards ECOWAS occurs in terms of migration policymaking, for example through forums like the Rabat Process. Soennecken (2014) likewise identifies processes of diffusion between the EU and Canada with regard to safe third country policies. Acosta Arcarazo and Geddes (2014) develop a framework for the transnational diffusion of ideas, policies and institutions. Their framework distinguishes between two sets of factors: material vs. ideational and structural vs. agential. Based on this, they develop a two-by-two matrix on different potential drivers for diffusion. Empirically, however, they find little evidence of a diffusion of ideas and policies from the EU towards Mercosur, where in fact migration policy developed to some extent in opposition to the former (cf. also Brumat and Freier, this volume). An examination of the diffusion of policies and norms to the EU from elsewhere remains wanting in research.

This does not mean, however, that the EU is always perceived as the stronger actor in negotiations, as scholars using theories of migration diplomacy highlight (Adamson and Tsourapas 2019). Greenhill has shown for Libya at various times (Greenhill 2010) and for Turkey in 2015 (Greenhill 2016), for instance, that weaker states engineer and use mass migration strategically to blackmail stronger states into financial support or political concessions. Tsourapas (2019) explains what strategy refugee hosting states chose in order to extract benefits from Europe. Western democracies are especially vulnerable to this as they are internally divided: on the one hand, they have to uphold standards of human rights protection while, on the other hand, they need to respond to their divided publics, parts of which have anti-immigrant preferences (Gürkan and Coman 2021).

Another approach that has been used to conceptualise EU external migration policymaking is the principal-agent (PA) framework. Menz (2015) has drawn on this approach to analyse the relationship between member-state principals and the Commission as an agent in the context of Mobility Partnership negotiations. His analysis uncovers the phenomenon of ‘principal slippage’, i.e. certain member governments – in a parallel process – negotiated bilateral accords with North African countries. Using the PA framework, Ripoll Servent (2018) analyses the reform of Frontex to mitigate the shortcomings of the CEAS, including the agency’s new right to intervene. Drawing on Majone’s (2001) typology of delegation, she suggests that Frontex was supposed to act as a trustee, i.e. safeguarding cooperation and ensuring the observance of commitments by principalss with heterogenous preferences. However, given that Frontex was provided with insufficient autonomy, it is at risk of being seized by particular interests and of being used as a proxy by member states with a strong regulatory tradition.

Taking a step back, it seems that while some conceptual approaches remain underrepresented in the literature – such as historical institutionalism and (compared to its utilisation elsewhere) also neofunctionalism –, others have not been explicitly applied to the external dimension of EU migration policy (like post-functionalism, but see Börzel and Risse 2018 on the internal dimension). In addition, although concepts such as learning and diffusion have been made use of, they remain under-researched in this policy domain. Frameworks that highlight the normative power of EU partner countries seem entirely absent, and non-Eurocentric perspectives generally still merit further conceptualisation and usage. This special issue seeks to fill some of these gaps, as the next section indicates.

Contribution of this special issue

The special issue makes several broader conceptual, methodological and empirical contributions to research on EU external migration policies.

Conceptually, we propose a definition of EU migration governance and its relationship to EU migration policy. We define EU migration policy as the sum of all formalised instruments adopted by the EU. This includes internal migration policy, such as the EU asylum acquis under the Common European Asylum System, EU labour migration policies for third country nationals and policies towards irregular migrants like the return directive. It also includes external migration policies, such as visa policies, border policies, Mobility Partnerships or the European Union Trust Fund for Africa. All these instruments are ‘formalised’ policies (Lavenex and Uçarer 2004, 421) adopted with the aim of regulating migration in one way or the other. In the external dimension, this is often – but not always – attempted through cooperation with third countries which are either origin or transit countries of migrants. In this context, the EU often uses conditionality (Ibid.) to either force or incentivise third countries to alter their own approach to migration (e.g. by adopting tougher border enforcement or criminalising irregular migration).

In addition to these formalised policies, there are more informal ways of EU influence, which we subsume under EU migration governance. These are usually not intentional but result from other countries or regions responding to EU policies by either supporting them, emulating these policies and modelling their own policies on their basis (Ibid.) or – and this has so far received scant attention in the literature – by openly opposing them and adopting policies that are decidedly different from those of the EU. Brumat and Freier (2023) provide a case of open contestation as the flipside of emulation, showing how Mercosur member states decidedly adopt a policy model on internal labour migration that is more encompassing than that of the EU to criticise the latter’s approach. Still, this contestation is a reaction to European policies and hence part of the EU’s (unintended) migration governance. The main difference between formal and informal ways of EU influence is that informal influence does not necessarily have an impact on migratory movements to the EU. However, it arguably may have when the emulating state is proximate enough to the EU so that emulating or adopting a different policy may attract or deter migrants which consider the EU an alternative target region.

The EU’s overall (external) migration system is thus made up of both EU migration policy and EU migration governance. This is depicted in Figure 1.

Figure 1. The EU’s (external) migration system.

Zooming in on EU migration policy, this special issue investigates the different regulatory³ approaches underlying EU external migration policy. We argue that most policies employ ‘remote control’ – broadly defined as a ‘set of practices, physical structures, and institutions whose goal is [to] control the mobility of individuals while they are outside the territory of their intended destination state’ (FitzGerald 2020: 9; cf. Zolberg 2006). Thus, policies either engage third parties in the enforcement of EU borders, usually outside of European territory, or they focus on the incentives of migrants and third countries. This includes either providing incentives to states to cooperate in policies of remote control or providing incentives to migrants to stay in their respective country of origin instead of trying to reach European territory. The latter, regulatory approach gained more relevance post-2015, with the EU expanding its activities in the area of resettlement as well as job-creating policies for example in the context of the EUTF Africa. As Table 1 shows, all policies use either of the two regulatory approaches and some even use both.

Table 1. Different regulatory approaches of EU external migration policy.

	Focus on remote control	No focus remote control
Focus on third countries’ and migrants’ incentives	Job creation and economic development programmes under the EUTF Africa (Zaun & Nantermoz), mobility partnerships (Cardwell & Dickson, Tittel-Mosser, Stutz), resettlement policies (Czaika et al.), readmission and visa facilitation agreements (Stutz, Czaika et al.), Frontex Working Arrangements (Stutz), administrative and financial cooperation (Stutz)	Jordan Compact (Vaagland)
No focus on third countries’ and migrants’ incentives	Cooperation with CSDP actors (Bergmann & Müller), cooperation with Libyan border force (Cusumano & Riddervold), visa policies (Czaika et al.), readmission agreements (Czaika et al.)

Many of the EU’s migration-development policies, e.g. under the EUTF Africa, as well as readmission and visa facilitation agreements, Frontex Working arrangements and administrative and financial cooperation in the area of migration management, have both a remote-control element and either aim at providing incentives to migrants to not migrate (in the case of the EUTF Africa) or to states to cooperate in the area of migration control. Visa policies, readmission agreements and cooperation with CSDP actors and external partners such as the Libyan border force have no such focus on the interests of states, nor do they focus on migrants’ incentives – beyond pure deterrence. They primarily aim to control migration by adding further hurdles – either physical hurdles through border force or regulatory hurdles through visa policies and the costs of applying for a visa. They may in a secondary dimension also aim to deter migrants from coming. However, this deterrent function is usually an add-on and not the main function. This is quite different to the aforementioned policies which directly take into consideration the preferences of migrants and third countries and try to set positive incentives for cooperation and non-migration. The Jordan Compact is an example of a policy that entirely forgoes EU control aims, where the EU is a respondent to a Jordanian initiative, and which only aims to increase the incentives for migrants to stay. It does not have a ‘remote control’ dimension. Stutz’ contribution can be understood as an analysis of the interplay between remote control and the role of incentives. While his research is based on the premise that migration cooperation both explicitly and implicitly serves as a means to control irregular migration to the EU, the objective of the paper is to define the impact of several factors, but not explicitly the impact of incentives. In his comparative study, he finds that broad migration cooperation (i.e. cooperation that encompasses legal, irregular and return migration, borders and administrative and financial cooperation) can be explained by a strong EU leverage, which in turn can require (but does not necessarily focus on) incentives for cooperation set by the EU.

Additionally, the individual papers make specific theoretical and conceptual contributions, thus developing new research agendas or contributing to recently emerging agendas. For instance, several papers analyse the impact that policy nexuses have on the policies which they connect and what this means for the actors and institutions that are now prompted to cooperate. In this context, we find that policy nexuses can lead to a horizontal transfer of politicisation from one area to another (here, from migration to development policies; cf. Zaun and Nantermoz) and that pressures for further integration between two different policy areas can entail more conflict but also cooperation between institutions in the different areas (Bergmann and Müller). Bergmann and Müller focus on the security-migration nexus and the intensified link between policies in the area of the Common Security and Defence Policy and the Area of Freedom, Security and Justice since 2015. They find empirically that, although institutional overlap between both branches has existed for a long time, both areas have developed largely independently of each other. Building on neofunctionalist explanations, they show that an increased overlap through spillover dynamics resulting from the 2015/2016 crisis can impact the integration of both areas and increase the potential for both cooperation and conflict. Zaun and Nantermoz investigate the EU’s approach towards the migration-development nexus (Lavenex and Kunz 2008). They argue that by reframing migration as an issue of underdevelopment, the EU has attempted to depoliticise migration, which had not only caused substantive divisions between EU member states in 2015/16 but also put governments under pressure from public opinion and populist challenger parties. The contribution shows empirically that three factors can explain the success or failure of such depoliticisation strategies: the types of actors involved, the location where the policy is implemented, and the salience and polarisation of the policy domains involved. Yet, as the case of the European Union Emergency Trust Fund for Africa exemplifies, such an approach can also lead to horizontal politicisation, in which development policies become more politicised if linked to migration control.

Analysing the role of politicisation in external migration and development policies is also a first step towards a more post-functionalist approach to external EU migration policies. The latter focuses on demonstrating that electorates have mattered significantly in the strengthening of the external dimension of EU migration policies post-2015 (Zaun and Nantermoz 2023). This is a theory of European integration that has so far not received any attention in EU external migration policy.

Another theory of integration that has so far received scant attention in the field is historical institutionalism. Cusumano and Riddervold (2023) combine ‘failing forward’ and institutional perspectives to explain why the EU has relapsed to relying on Libyan border forces to control irregular migration. They argue that this is particularly puzzling given the instability of the Libyan regime and the negative human rights implications of that cooperation, as Libya is not a signatory of the Geneva Refugee Convention and genuine refugees will not find protection there. This suggests that the EU did not take any lessons from the past court decisions. Cusumano and Riddervold hold that this behaviour can be explained by a mixture of muddling-through processes in which ‘path-dependent incremental responses building on pre-existing institutional architectures’ are used. They link this approach to the literature on ‘failing forward’, suggesting that this is an ongoing process in which policies are adopted especially in response to crises. Yet, due to path-dependencies, these are never ambitious or reforming the system in a way that would make it functional. In contrast to most accounts in the historical institutionalist tradition, the authors show how stickiness and path-dependency can lead to backsliding into pre-existing institutional arrangements (rather than locking in a path towards further integration). The notion of ‘failing though’ (Cusumano and Riddervold) thus provides an innovative conceptual alternative to ‘failing forward’, which received significant attention in the literature (Lavenex 2018; Scipioni 2018).

Furthermore, the special issue also contributes to a yet understudied phenomenon in International Relations more generally – namely, the issue of policy learning, diffusion and epistemic communities. While EU migration policies are certainly a case where such learning has been observed more generally, including their external dimension (Soennecken 2014), we still know very little about the conditions and mechanisms for learning in this area. Tittel-Mosser (2023) argues that the influence of epistemic communities depends both on factors related to the epistemic communities themselves and on structural factors. Epistemic communities normally need a certain prestige to be influential. However, what is even more important is that the partner country is interested in their advice and input.

While most contributions in the special issue clearly focus on the EU and its preference formation, two contributions take a critical approach to the Eurocentrism of ‘EU’ external migration policy. They look into the perception of non-European countries and their preferences, while also contributing to a growing body of research on the reception and perception of EU migration governance in other parts of the world (Adam et al. 2020). Identifying phenomena of policy diffusion and learning, Brumat and Freier (2023) demonstrate that, while South American countries have readily followed the EU’s example regarding the introduction of internal freedom of movement, they have morally and politically opposed the EU’s restrictive external migration governance. Both processes have not necessarily occurred on the EU’s instigation but are rather unintended consequences of EU migration policies. South American countries are hence not passive policy-takers but active shapers that look into the world and adopt or contest policies that they support or oppose, underlining their agency. This takes previous research approaches highlighting the bottom-up implementation of policies a step further (Wunderlich 2012) and gives more room to the agency of non-EU countries.

In a similar vein, Vaagland (2023) studies the EU as a respondent to a Jordanian initiative, showing that the EU is not always in the driving seat in its external migration policymaking. At the same time, she brings the EU’s perspective back into research on migration diplomacy, demonstrating that the EU is not a passive recipient of threats by countries that use refugees as bargaining leverage for more economic support but that it has the capacity to assess these critically and has normative reasons to provide support to top hosting countries in the Global South. Moreover, Vaagland brings a fresh theoretical perspective to external EU migration policies by showing that moral arguments of EU partners matter and can bind the EU if they are credible, the EU had a long-standing relationship with the partner and the contested issue is important to both sides of the negotiation table.

Against the backdrop of little large- and medium-N research in EU external migration policy, Stutz (2023) explains why the EU cooperates on migration with some third countries rather than others and hence addresses an important empirical gap. Using a fuzzy-set Qualitative Comparative Analysis, a so far rarely used methodological approach in EU external migration policies (but see Guérin 2018), he finds that a third country’s existing relations with the EU or it being a democratic state and, at the same time, economically dependent on the EU explains intensified cooperation rather than being a transit or sending country for migration. This sheds critical light on EU external migration cooperation because it suggests that these policies and agreements are unlikely to meet their aims.

Cardwell and Dickson (2023) take a legal perspective on EU Mobility Partnerships to describe a phenomenon which they refer to as ‘formal informality’ and which characterises much of the EU’s external migration governance. Moreover, they argue that the informality approach chosen by the EU is problematic, as it allows EU member states to bypass the substantive and procedural formalities associated with law and thus undermines transparency and the protection of human rights. Indeed, the legal quality and implications of these policies are under-researched to a great extent because these policies have little legal quality and are therefore unusual suspects for legal analysis. However, given that these policies can have severe human rights implications, further legal study of them is needed. This is even more important against the background of Cusumano and Riddervold’s finding on the stickiness of restrictive dynamics and the re-emergence of the restrictive paradigm post-2015.

Going beyond the impact of particular policy instruments, Czaika, Bivand Erdal, and Talleraas (2023) analyse how policy instruments in the external dimension – such as visa, readmission and resettlement policies – develop and interact spatially, temporally and categorically across European states. By paying attention to these instruments and dimensions, we enhance our understanding of how external migration policies (also interacting with other migration-relevant ones) operate together in a broader European migration policy mix. They find strong evidence for spatial policy dependence across European countries for all three policy instruments (especially for visa policy), but also, among others, for categorical dependence between selected external instruments within European countries. The authors argue that the overall European external migration policy mix remains heterogeneous so that these instruments and policies may erode each other’s effectiveness and lead to unintended consequences.

Acknowledgements

We are grateful to the anonymous reviewer, Paul Statham and the special issue contributors for their helpful comments on earlier drafts. In addition, we would like to thank Svenja Budde, Alexandra Bumcke and Charlotte Stüber for their competent research assistance.

Disclosure statement

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

Notes

1 We thank an anonymous reviewer for this point. Also cf. Niemann and Zaun (2018).

2 Also cf. Cusumano and Riddervold this volume and Bergmann and Müller this volume for related work in these traditions.

3 “Regulatory” is not to be understood here in the legal sense as denoting regulations, but to be understood more broadly in the sense of steering, governing or managing.