The views and opinions expressed in this article are those of the authors and do not necessarily reflect those of the institution to which they are affiliated
1Over the past twenty years, the European Union (EU) has developed a policy aimed at integrating migrants in their receiving societies. It was first advanced without a legal competence, through a patchwork of soft law instruments. The entry into effect of the Treaty of Lisbon in 2010 formally conferred on the EU a role in the matter, thus suggesting an expansion of its integration policy. The start, last year, of the negotiations for the EU’s forthcoming multiannual financial framework seemed to confirm a tendency towards the consolidation of the policy. Yet, a careful analysis unravels a more nuanced process, which poses the question of the direction the EU’s policy is going. This article investigates the trajectory the EU’s policy on the integration of third country nationals has followed thus far in order to assess whether the locus of policy-making in the domain is shifting towards the EU.
- 1 There is quite a wealth of publications on the matter but see, inter alia, Castles et al., 2013.
- 2 Realising the potential of migration has been a recurring topic at the EU level since at least 2003 (...)
2As migration has become a structural feature of European society, integration has gained central relevance, be it for migrants themselves or for the EU’s economy as a whole. As Europe ages, the proportion of workers compared to that of pensioners shrinks, posing obvious problems in sustaining both the EU’s economy and member states’ social policies (European Commission, 2011). For several years now, immigration has been the main driver of population growth (Lanzieri, 2010; see also OECD, 2016). Projections show that influxes are not likely to cease in the near future. Yet, migrants face considerable obstacles on their path towards full labour market integration1, thus undermining the value they can add to the EU and to national economies. If EU member states wish to make the most of the potential migration holds, integration is a key issue and an issue that they seem to have acknowledged2.
3When they gathered in 1999 in Tampere, Finland, to set the basis of the Area of Freedom, Security and Justice (AFSJ), the member states called immediately for a “more vigorous integration policy” to be put in place at the EU level (European Council, 1999). Indeed, as much as environment or transportation, immigration—and integration as one of its subsets—has transnational stakes, thus justifying the Union’s action in the policy realm. However, no legal competence existed on the matter. A policy on the integration of migrants unfolded nonetheless, from the early 2000s up until the present day. Comprising a common conceptual framework relating to integration, networks of officials, benchmarks and funding opportunities, the policy took the shape of an ensemble of soft instruments, a patchwork of tools forming a fragmented but coherent framework. With the entry into force of the Treaty of Lisbon in 2010, the EU was endowed with a formal competence on integration. Although limited in scope, such a competence hinted at a plausible expansion of the policy in time, through the adoption of new policy instruments. Instead, the development of the policy in subsequent phases displays ambiguous signals as to whether it is expanding or shrinking. This article aims to answer the following question: where is the EU’s policy on the integration of migrants heading? Is it moving towards a greater role for the EU or is the EU’s intervention in the domain retrenching?
4Whereas scholarship has focused on the effects of single EU outputs on national policies, and thus concentrated on single events, this piece of research conceives the EU’s policy as a process. By linking the dots of its progressive development, it seeks to identify the trajectory this policy has followed over time. Drawing upon neofunctionalist theory, I seek to reconstruct the process that has led to today’s integration policy at the EU level. I take as an indicator of policy dynamics the evolution of the funding structure for integration, as established by the succession of multiannual financial frameworks (MFFs). The passage from the 2007-2013 to the 2014-2020 MFF as well as the debate over the forthcoming 2021-2027 MFF offer a unique opportunity to compare different phases of development of the policy and assess whether its evolution is headed towards more or less EU involvement in integration policies. I find that the process at play is dual: whilst the EU’s role consistently shrinks on substantive directions, it appears to be thriving on the procedural side.
5The first section of the paper provides a brief review of the growing literature on the topic in order to refine this article’s contribution to the debate. I also clarify the perspective endorsed in this article, namely, neofunctionalism. The second section indicates the epistemological choices made to characterise the process under scrutiny. It also outlines the method and technique at the basis of the findings presented. Sections 3 through 5 analyse each of the three MFFs in turn, providing a comparison of their specific features with the previous funding scheme. I also provide some context to underline some of the functional pressures exerted (e.g. the asylum crisis). Lastly, I conclude in the sixth section by taking stock of the findings and addressing the research question.
6Despite the fact that the first reference to an EU integration policy was made approximately 20 years ago, political scientists and public policy scholars have only recently started to consider it as a research object. Most of the scholarly attention to migrant integration policy has, thus far, focused on nationally and sub-nationally defined public policies, relegating the EU’s intervention in this policy realm to a secondary aspect to be taken account of in a multilevel governance structure (Geddes and Achtnich, 2015; Scholten and Penninx, 2016). In contrast, the development of the Area of Freedom Security and Justice (AFSJ) has given rise to three different strands of study focusing on the EU level. The first one, mostly consisting of lawyers, stressed the potential conflicts between EU and national legal orders that the AFSJ, and the provisions relating to legal migration (and thus somehow concerning integration), could generate (Kostakopoulou, 2002a; Groenendijk, 2004; Murphy, 2009; Handoll, 2012). A second strand has concentrated on the effects of the provisions contained in the various Directives relating to legal migration, notably in the family reunification Directive (Oosterom-Staples, 2007; Groenendijk et al., 2007; Hailbronner, 2010), and in the long-term residence Directive (Halleskov, 2005; Kostakopoulou, 2002b). A third strand has developed whilst the policy, facing the reluctance of member states, has taken the shape of soft law instruments.”. Whilst some scholars have studied the Commission’s attempt to launch an Open Method of Coordination (OMC; Caviedes, 2004; Szyszczak, 2006; Velluti, 2007), others have studied the effects of the EU’s outputs on national policies, notably with regard to the Common Basic Principles on integration (Mulcahy, 2011; Carrera and Faure Atger, 2011) and the European Integration Fund (Van Wolleghem, 2017; 2019).
7On the other hand, little attention has been paid to the original pattern the policy as a whole formed. Although some EU and national officials who took part in the policy-making process over the 2000s have provided interesting factual elements, a wider theoretical perspective has been lacking (Urth, 2005; Pratt, 2015). Carrera (2008) has framed the whole policy as a “quasi-OMC”, thus alluding to the coherent patchwork of soft law instruments. Van Wolleghem (2019a), for his part, has traced the process that led to the emergence of the EU’s integration policy. However, these studies have proposed accounts that are either static or focused on a limited timespan; therefore, they have to date been unable to produce a dynamic analysis of the trajectory of the policy as whole. This is despite the fact that the persistence of the policy in time and the changes it seems to have undergone pose the pertinent question of the direction the policy is taking.
- 3 Early versions of Neofunctionalism were formulated by Haas (1958) and Lindberg (1963). They were su (...)
- 4 The confusion in terms between “policy for the integration of migrants” and “European integration” (...)
- 5 Note though that this is a minimal definition of European integration. Haas (1958) originally theor (...)
8In order to understand these changes and hypothesise the policy’s trajectory for the years to come, I borrow from neofunctionalist theory. Despite the vivid criticisms it has received over the years, particularly relating to its early formulations3, neofunctionalism proposes useful concepts for the analysis of European integration4. In this line of thought, European integration is understood as a process of gradual shift5 from political activities towards a new centre (Haas, 1958). It is characterised by diverse and multiple actors whose interaction leads to incremental decision-making (as opposed to grand designs or teleological dynamics), itself driven by marginal adjustments resulting from the unintended consequences of previous decisions (Niemann and Schmitter, 2009). Importantly, neofunctionalism also distinguishes between the depth and breadth of European integration; the former being a matter of degree to which a given issue is governed at the EU level, and the latter being a question of the scope of issues in which the EU has a say (Niemann, 2006). The purpose of this article is to capture the dynamics at play in the development of the EU’s integration policy to assess whether we are witnessing a process of European integration. Therefore, I focus on the depth of the phenomenon; that is, the degree to which a particular policy realm becomes an EU matter, rather than on the breadth of the phenomenon (or how the EU gained competence on the matter in the first place). According to neofunctionalist orthodoxy, one would expect a process of spillover to take place, irresistibly leading to further European integration (Haas, 1958). Revised versions of the theory have, however, tended to reject the automaticity of spillover effects, hypothesising the existence of other responses and, notably, the reversibility of regional integration (Schmitter, 1970; 2004).
9Capturing policy dynamics is a complex undertaking, in that the perspective from which it is observed has far-reaching consequences on the research design, choice of variables and methodology, and, of course, on research results (Capano and Howlett, 2009). The risk is therefore high that one sees a certain level of European integration in every policy in which the EU has a say. Epistemological choices, therefore, ought to be clarified in order to adequately grasp the essence of the phenomenon under study (Capano, 2009). While it is clear that the purpose of this article is to establish whether a process of integration of the policy in question is underway, it is less clear what is regarded as a case of European integration. European integration is generally understood as the expansion of the role of supranational institutions, but such an expansion may manifest in different ways. For the purpose of this article, I consider two main possible indicators of European integration. On the one hand, there is European integration in cases in which the Union acquires the ability to orientate member states’ substantive policies. On the other hand, I consider there is European integration when the Union is capable of altering procedural features in a way that either diminishes member states’ discretion or increases the EU’s.
10From the methodological standpoint, the perspective endorsed implies choices as to the research object, which allows an empirical study of the policy’s chronology.. I consider the multiannual financial frameworks (MFFs), and the funds they cover, as indicators of the pace and direction of change. MFFs constitute the medium-term budget of the EU; they plan the spending of its resources for seven years in accordance with the EU’s political priorities. Each MFF is then broken down into specific funds, each with their own priorities and spending rules. In this way, MFFs (and the funds they cover) indicate how, how much and when the EU is to intervene in a given policy field. The present research spans three MFFs: i) the 2007-2013 MFF, long concluded, and for which a wealth of documents is available; ii) the 2014-2020 MFF, which is still being implemented; and iii) the 2021-2027 MFF, which is still being negotiated and for which only the Commission’s proposal is available. For each of these, I propose a comparison of the funds explicitly addressing the integration of migrants. Table 1 below summarises the main changes. I also underline elements of situational and financial context in order to outline the functional pressures exerted on the policy.
Tab. 1 – Evolution of the EU’s funding scheme for the integration of migrants
- 6 Featured in Regulation 514/2014/EU, article 3(1), which reads “The Specific Regulations shall provi (...)
- 7 Featured in COM(2018) 471 final, article 7(1), which reads “Support provided under this Regulation (...)
Item
|
2007-2013 MFF
|
|
2014-2020 MFF
|
|
2021-2027 MFF (COM proposal)
|
Fund name
|
European Fund for the Integration of third country nationals (EIF)
|
|
Asylum, Migration and Integration Fund (AMIF)
|
|
Asylum and Migration Fund (AMF)
|
Scope of the fund
|
Narrow, only integration
|
|
Wide, with safeguards for asylum and integration
|
|
Wide, with no safeguard
|
Targeting principle
|
Exclusive, only third country nationals who are not beneficiaries of international protection
|
|
Inclusive, non-EU citizens
|
|
Inclusive, non-EU citizens with focus on short-term integration
|
Amounts (as per original legal act)
|
825 million
|
|
3,137 million
|
|
10,415 million
|
Of which explicitly for integration
|
825 million (100%)
|
|
478 million (15.2%)
|
|
0
|
Co-financing principle (EU’s share)
|
50%
75% maximum
|
|
75%
90% maximum
|
|
75%
100% maximum
|
Implementation method (as per legal act)
|
93% shared management
7% direct/indirect management
|
|
88% shared management
12% direct/indirect management
|
|
60% shared management
40% shared or direct/indirect management
|
Substantive policy directions
|
High, attempt to initiate convergence through implementation of the CBPs and some level of coercion/incentive
|
|
Low, widely defined
|
|
Low, widely defined
|
Procedural constraints
|
Low, high discretion of central governments, weak partnership principle, no additionality principle.
|
|
Medium, stronger partnership principle, weak additionality principle6
|
|
High, reinforced partnership principle, weak additionality principle7
|
Source: The Author
11This article borrows mostly from case-study methods, a heterogeneous set of approaches that, however, have in common the intensive study of a bounded system (or systems) over time through detailed data collection and analysis (see Creswell, 2007, for an interesting account). As for the technique used, the findings presented rely on document analysis to map the evolution of the EU’s integration policy, from the outset to the present day. Data was obtained via the EU Commission’s various platforms from which official documents were downloaded. They mostly consist in Communications from the Commission (and technical appendices), European Council Conclusions and adopted pieces of legislation. From the bulk of documents retrieved, I have focused my attention on those specifically concerned with budget matters. For ease of reading, only the most relevant documents are cited in the main text.
- 8 In this respect, see Bigo, 1996; 2002; Guiraudon, 2003; Huysmans, 2000.
- 9 The list of publications in this respect is rather long. See inter alia Schnapper (1994) and Wische (...)
12Further to the entry into force of the Treaty of Amsterdam and the forthcoming creation of the Area of Freedom Security and Justice (AFSJ), the programme member states agreed upon in Tampere provided the European Commission with a mandate to develop an integration policy. Despite member states’ commitment in principle, implementing the “more vigorous integration policy” impetus proved harder than expected (Urth, 2005; Van Wolleghem, 2019a). Notwithstanding, the general context was somehow favourable to the emergence of an integration policy. Even though migration was mainly framed as a security issue8, all the more so after the attacks on the Twin Towers in 2001, the social inclusion paradigm pervaded the EU policies of the time, to the point that it was enshrined in EU primary law (Vanhercke, 2012; Mathieson et al., 2008). Accordingly, migrants appeared as a group at risk of exclusion that should therefore be the object of specific attention. Integration had already been on the agenda of most member states since the late 1980s/early 1990s9, such that the context at the national level was favourable to integration policy too. With a policy space available and a window of opportunity swinging open, the conditions were met to develop an integration policy at the EU level. But member states were resistant when it came to moving forward formally, notably with the adoption of Directives, thus showing that, if there were to be an EU integration policy, the policy space available for it would be very narrow.
- 10 This paragraph summarises some of the findings presented in Van Wolleghem (2019a).
13The policy took shape thanks to the interaction of three conditions10. Firstly, to circumvent member states’ reluctance, an integration policy would necessarily display great flexibility; soft law thus became the best alternative to normal policy instruments (notably Directives). Secondly, the policy had to pass from national agendas to the EU’s agenda. The succession of several Presidencies of the Council with similar preferences proved successful in transferring these preferences to the EU level. Of course, such intent does not always pay off since policies or national agendas are in competition with others and not all of them end up on the EU agenda. But because any initiative in this domain necessarily involved soft law – so that member states would have the possibility to adopt a text without ever having to implement it – no member state had an interest in standing in the way of adoption and breaking the culture of consensus that reigns in the Council (see Aus, 2008, for an interesting example). Lastly, the third condition consisted in the readiness of the Commission to occupy a policy space without ever impinging on member states’ exclusive competence. In so doing, the Commission systematised and transformed the progress made in (soft) policy instruments.
- 11 Decision 2007/435/EC.
- 12 The financial incentive only applied to the member states not eligible for the cohesion fund. Membe (...)
14It is clear from the foregoing that the space for an EU integration policy was very narrow. In spite of this, a number of policy instruments were adopted. The most important ones in the first years were: i) the National Contact Points on Integration (NCPI), created in 2002 as a network of officials for the exchange of information and practices amongst member states; ii) the Common Basic Principles on integration (CBPs), adopted in 2004 as a set of 11 short-worded non-binding principles aimed at guiding member states policies; and iii) the European Integration Fund (EIF)11, adopted in 2007 as a fund of approximately €825 million to be spent over the period 2007-2013. The NCPI, the CBPs and the EIF were strongly interconnected. The NCPI functioned as a forum on which the Commission could rely to develop and discuss new initiatives at the EU level before starting the official procedure (Urth, 2005). This is where the CBPs were first presented as well as the first ex-ante impact assessment and the draft Decision relating to the European Integration Fund. The CBPs were thought of as the conceptual framework on which further policy instruments relating to integration should be based. Lastly, the EIF provided for systematic funding opportunities to support national policies implementing the CBPs. The implementation of these principles was consistently encouraged through a mix of “soft obligations” and financial incentives. For soft obligations, member states would need to state how and the extent to which they would implement the principles in their multiannual spending programme. Likewise, some priorities, albeit vaguely defined, would be mandatory but broad enough to fit different ambitions (Van Wolleghem, 2019a). As for financial incentives, EU money would support projects addressing a set of specific priorities by increasing its co-financing share from 50 percent to 75 percent12. In summary, the EIF had a strong convergence ambition; it provided a series of substantive policy points and encouragements for the member states to follow them.
- 13 In her study, Mulcahy (2011) finds that the CBPs had no effect whatsoever on member states’ integra (...)
15However, without a strong foothold on integration matters at the EU level, the adoption of a fund that would consistently enforce a European view of integration was unlikely. The absence of a clear legal basis would necessarily imply a limited role for the Commission in the definition of the priorities, as well as very limited control mechanisms over member states’ implementation choices. Accordingly, it would be up to the state to decide whether or not to use EU funding to give effect to the CBPs13. As the Decision establishing the EIF reads:
The general objective of the Fund is to support the efforts made by the Member States in enabling third-country nationals of different economic, social, cultural, religious, linguistic and ethnic backgrounds to fulfil the conditions of residence and to facilitate their integration into the European societies (Decision 2007/435/EC, article 2; emphasis added).
16From the outset, the EIF was thus placed in a limbo between the exclusive national competence and the European framework. Even though the CBPs were designed so that they would focus on key areas and shared problems rather than on national practices in order to increase their acceptability (Pratt 2015), and even though they were adopted at the unanimity of the member states, their practical implementation as per Multi-Annual Programmes was found to be underwhelming (Carrera and Faure Atger, 2011). Similarly, evidence regarding the Annual Programmes shows that implementation of the priorities set at the EU level was very modest, despite the existence of financial incentives (Van Wolleghem, 2017).
Fig. 1 - Implementation of EU indications by the member states in percent of total funding, by financial incentive, mean percentage and standard deviation, 2007-2013
Source: Van Wolleghem, 2017, p. 1131
- 14 The priorities in question were established in Commission Decision C (2007) 3926 final.
- 15 See Van Wolleghem (2017 and appendix) for more detail and calculation methods.
17Figure 1 displays the means and standard deviations of member states’ intention to implement the priorities set by the European Commission14, as per their annual programmes, for the years 2007-2013. The values are expressed in percentage of total funding so that, for instance, Germany planned to use, on average, about 40% of its total allocation to address EU priorities whilst Poland did not intend to address them at all15. Note that some member states were financially incentivised to address these priorities (on the right-hand side of the figure) whilst the member states eligible for the cohesion fund were not (on the left-hand side). In total, a stark 28.4 percent of the total funding over the seven years of implementation and across countries actually planned to tackle EU indications. The financial incentive provided by the EU proved incapable of dragging member states’ preferences towards the indications set at EU level.
18On the procedural side, the EIF, despite being designed on the model set by structural funds, hardly resembled the latter. Indeed, it would grant the Commission control mechanisms to ensure lawful and sound financial management, but national governments would still enjoy wide discretion as to how to spend their allocations. The principles—now well established—introduced with the reform of the EU’s cohesion policy in 1988 (Bachtler and Mendez, 2007) would have little or no bearing. This meant that no additionality principle was provided for, so that member states could substitute national funding with their EIF allocations. Likewise, the partnership principle, which associates a series of actors to the management of the funds, was made optional in the case of the EIF. As a matter of fact, most member states considered the principle as a one-off consultation on the occasion of the drafting of the Multi-Annual Programme (Carrera and Faure Atger, 2011).
19All in all, it appears that the member states were capable of retaining a firm grip on the use of the fund and on the policy altogether. The end of the 2007-2013 financial framework and the entry into effect of the 2014-2020 financial framework opened up on a new phase of policy development that offers the possibility to compare the design of the EU’s intervention on integration matters in time and see whether its role has improved or receded (Table 1 above summarises the most relevant changes).
- 16 Decision 2007/435/EC, article 2.
- 17 Regulation 514/2014/EU.
20The EIF was created as a means to “support member states’ policies”, as the Decision establishing the fund read16, but was supposed to give effect to the Common Basic Principles on integration, a set of principles aimed at framing member states’ policies in order to, eventually, reach some level of convergence (Mulcahy, 2011). The entry into force of the new financial framework brought to bear its sequel, the Asylum, Migration and Integration Fund17 (AMIF). Before exploring its features, some preliminary remarks are in order as important changes occurred between the adoption of the EIF and the adoption of the AMIF.
- 18 Integration was first mentioned in primary law in the Constitutional Treaty. It notably excluded le (...)
- 19 Article 79(4) TFEU. Note that the Constitutional Treaty, rejected in 2005, featured the same provis (...)
- 20 Even though no legal basis was provided, measures on integration were adopted on the basis of artic (...)
21First of all, the overall institutional framework had changed. The Treaty of Lisbon, which entered into effect in 2010, finally provided a legal basis for the EU’s policy on integration18. However, such a legal basis was limited to a single paragraph that circumscribed the EU’s action to providing “incentives and support for the action of member states”19, thus constraining the scope of a sound EU integration policy. Another major change, introduced by the treaty of Lisbon, is the change in voting rules that ended the legal migration special regime. Decisions relating to integration would no longer be taken at the unanimity of the member states20 but at qualified majority voting.
22The question that arises is whether these changes translate into a more consistent EU integration policy or the other way around. Considering the Treaty of Lisbon, the introduction of a clear reference to integration may be construed as a further step towards a stronger EU integration policy whereas the limited scope of this very same provision may be read as a step backwards. More importantly, the changes in the overall structure of the migration funds send ambiguous signals, both money-wise and priority-wise.
- 21 Border management was not included in the AMIF. It was made part of the Internal Security Fund, ano (...)
- 22 I write “initially” because it was considerably increased during the asylum crisis in order to tack (...)
- 23 The amount calculated here is lower than 20 percent of the AMIF’s total amount (€3,137 million). Th (...)
- 24 It should also be noted, as Westerby (2018, p. 22) points out, that the minimum percentages to be d (...)
23The EIF’s evaluation reports have consistently underlined the complexity of the fund’s spending rules (Ramboll, 2011; 2018b). On the one hand, the administrative burden was deemed too heavy compared to the amounts distributed; on the other, the existence of two funds with overlapping, yet mutually exclusive targets, made its use difficult (see below). In response to these issues, the structure of the funds for the EU’s migration policy was overhauled with the 2014-2020 MFF. Whereas the previous financial framework counted four distinct funds—each for a specific item, namely borders, refugees, return and integration—with clear allocations for each of them, the 2014-2020 financial framework merged them into a unique fund21, thus granting more flexibility in its use on the part of the member states (Malmström, 2014; Ramboll, 2018a; Westerby, 2018). In order to ensure flexibility would not translate into too much focus on one item or another, safeguards were introduced. For integration, it was provided that a minimum of 20 percent of the fund be earmarked for the integration of migrants. This safeguard opened the possibility that less funding be dedicated to integration than with the EIF. Considering that the AMIF initially22 amounted to €3,137 million, in the event that every member state would decide to limit their allocations to integration to the 20-percent threshold, the amount that would go to integration would be about €478 million23, slightly more than half as much as the EIF. Going over this minimum would then be up to the state and how it intended to distribute its allocation between integration, asylum, solidarity and return24.
- 25 The reports available on the implementation of the ERF mention the share of projects that addressed (...)
24Furthermore, the benchmark for comparison in this situation is not the EIF but the EIF together with a part of the European Refugee Fund (ERF). The EIF was a fund strictly targeting the integration of third country nationals who are not beneficiaries of international protection. Instead, the ERF was dedicated to a wide range of interventions relating to asylum, comprising the integration of beneficiaries of international protection. With a total of €628 million, it is hard to appraise the share of the ERF that was actually used by the member states to ease the integration of beneficiaries of international protection25 and thus that which should be added to the EIF to build a proper benchmark against which to compare the AMIF minimum amount for integration. The fact remains that the AMIF presents the clear opportunity to stick to a minimum of €478 million dedicated to integration.
25The data available thus far shows that member states exceeded this threshold. The AMIF interim evaluation report calculates that around 32 percent of the fund was used to address the specific objective “Integration/legal migration” over the period 2014-2017 (Ramboll, 2018a, p. 33). This summary figure, however, hides differences between groups of countries with, for instance, the UK and Ireland displaying an average of 24 percent of their allocations to the issue, and Italy, Greece and Malta around 22 percent. Greece obtained a derogation to the 20-percent minimum allocation rule and foresaw a sheer 12.5 percent of its total to integration and legal migration (Westerby, 2018, p. 22).
- 26 One report has been published thus far, that of Ramboll (2018a) on the midterm evaluation of the AM (...)
- 27 For instance, in Radaelli’s (2003, p. 30) conception of Europeanisation, it is likely that the ways (...)
26Beyond the amounts provided for, another aspect is fundamental to assess where the EU integration policy is heading: the definition of priorities. With the EIF, there was a clear attempt to generate convergence through the definition of principles as its use was supposed to contribute to the implementation of the CBPs. Numerous were the references in the EIF Decision directly or indirectly referring to the implementation of the CBPs. Resultantly, the EIF’s evaluation reports also emphasised the priorities addressed by member states while spending EU resources (Ramboll, 2011; 2013). No such an attempt appears to have been at the basis of AMIF’s design. Instead, the AMIF only includes loose references to substantive policy points. If one considers as an example the attention paid to the CBPs, the only direct reference to them is Recital (20) of Regulation 516/2014/EU, even though indirect references may be seen in the reading of article 9 that specifies the kind of integration measures that the fund supports. Consequently, CBPs do not feature in member states’ spending programmes or in their reporting activities and nor in monitoring and evaluation activities26. This translates into less attention being paid to the common priorities established at the EU level for integration and more attention to member states’ priorities. This does not mean that the CBPs are ignored altogether, as they may be part of the shared beliefs that orientate member states’ policy making27, but it does mean that they lose visibility, which makes it hard to appraise whether they are being followed or even heeded.
- 28 For more detail, see Van Wolleghem, 2019a, p. 115.
- 29 See article 12, Regulation 514/2014/EU. The provision also specifies that ‘the composition of the p (...)
27On the procedural side, significant changes occurred in the passage from the EIF to the AMIF. The principle of partnership was significantly reinforced with the AMIF. Originally, the partnership principle was introduced by the Commission with the reform of the EU’s cohesion policy in 1988 (Bachtler and Mendez, 2007) and provided for the participation of national, supranational, but especially subnational actors, in the design and implementation of programmes adopted in the framework of said policy. As already mentioned, the EIF comprised a partnership principle in its design, albeit a much less stringent one. Because the principle, as formulated in the EIF, left great room for manoeuvre regarding the choice of the stakeholders to be associated and the phase of the policy cycle in which they would participate, it rather consisted in an invitation to consult bodies other than governments in the management of the fund28. Differently, the AMIF features a stronger partnership principle, which forces the association of subnational bodies to the “preparation, implementation, monitoring and evaluation of national programmes”29. In so doing, it consistently decreases central governments’ grip on the management of the fund.
- 30 See European Commission, 2014, for more on EU budget implementation methods.
28Lastly, yet importantly, the AMIF has greatly increased the share of the fund the EU Commission can use with some level of discretion. According to financial Regulations30, EU money may be spent following three methods: i) shared management: the member states spend, the Commission controls; this is the management method used for member states’ allocations; ii) direct management: the Commission (or EU agencies) spends directly; iii) indirect management: the Commission delegates by entrusting budget implementation to third parties (other than member states in the case at issue). Whilst shared management puts the member states in charge, so to speak, direct and indirect management confers on the Commission greater room to express its preferences, through the adoption of its annual work programmes, within the limits set by specific funds’ Regulations. With regard to the EIF, 7 percent of the total funding (€58 million) was to be spent through direct management for Community actions; i.e. transnational projects or actions of interest to the Community as a whole. The AMIF considerably extended both the scope and width of the direct and indirect spending method. The Regulation earmarked 12 percent of the total (€385 million) to be spent for Union actions, emergency assistance, the European Migration Network and technical assistance of the Commission.
29Everything considered, it appears that the AMIF has initiated a dual process of European integration characterised by less substantive directives but more procedural pressure and more room for the Commission’s preferences. Pursuing the objective of easing implementation and increasing spending (Malmström, 2014; Ramboll, 2018a; Westerby, 2018), the AMIF has shed most of the EIF’s substantive objectives, thereby reducing its ambition to generate convergence. At the same time, the new fund has increased both the EU’s and subnational bodies’ weight in the management of integration funding. This logic appears to have been accentuated with the outbreak of the asylum crisis.
- 31 See for instance the failure of the Commission’s proposal to create a refugee quota system (Zaun, 2 (...)
- 32 See SWD (2018) 339 final.
- 33 Ibid., p. 19.
30The sudden increase in asylum cases over the years 2014-2016 put the EU’s ability to propose a concerted response to the test. The uneven distribution of inflows and the unanswered calls for more solidarity31 generated tensions among member states. In the midst of the crisis though, the AMIF was topped up with significant resources to face the emergency32. By the end of 2017, this sum amounted to €6,888 million, a 120 per cent increase compared to the original budget. Interestingly, the decision taken in the passage from the 2007-2013 to the 2014-2020 MFF to merge the funds dedicated to integration and asylum, combined with the response to the asylum crisis, has had the unintended effect of increasing the resources allocated to integration altogether. With the increase of the AMIF, an additional €2,305 million was made available for the integration of third country nationals (irrespective of their protection status). Such an increase was, however, lower than the increase dedicated to asylum (solidarity, relocation and reception), such that, even if the absolute amount for integration increased, its relative importance decreased to 19 percent of the total funding33.
31At the same time, the percentage of funding at the disposal of the Commission under direct and indirect management—thus with some level of discretion—was increased to 22 percent of the new total (that is, €1,491 million instead of the 12 percent originally allocated). A good share of the amount at the Commission’s disposal served to fund the emergency assistance (€1,029 million, or 15 percent of the total funding) but significant resources were attributed to Union actions (€462 million, some 7 percent of the total). Unfortunately, the data available does not allow a disaggregation of the figure per item but it appears that around 76 percent of the total amount spent for Union actions covered either legal migration/integration or returns34.
32In summary, the new structure of EU migration funding introduced with the 2014-2020 MFF, together with the response to the asylum crisis, resulted in both more resources for integration and more resources to be spent at the Commission’s initiative. This incremental development seems to have been followed up with the Commission’s proposal for the 2021-2027 MFF.
- 35 With, on the one hand, the Visegrád Group coalition against relocation and, on the other hand, the (...)
33The asylum crisis has generated tensions between member states which are, paradoxically, likely to facilitate the emergence of a consensus on the need for more funding for an integration policy. More precisely, member states’ reluctance to manifest solidarity by relocating migrants in their territory (Zaun, 2017; Ortensi et al., 2018) may very well translate into them preferring financial solidarity35. In this perspective, the Commission’s proposal for a renewed MFF brings about two novelties: a significant increase of migration funds and of the AMIF’s substitute, the Asylum and Migration Fund (AMF - see below); and an attempt to mainstream the integration of migrants by including it as a clear objective for the European Social Fund Plus (ESF+) and the European Regional Development Fund (ERDF).
- 36 See COM (2018) 375 final.
- 37 See COM (2018) 471 final.
- 38 See COM (2018) 382 final, article 7(3).
34Starting with the latter point, the European Social Fund (ESF) has always, more or less explicitly, concerned migrants as a group prone to exclusion and whose labour market integration should be facilitated. However, ESF resources have never been either distributed to member states in accordance with their share of foreign residents or specifically targeted at third country nationals (CSES, 2011). With the forthcoming MFF, these two aspects are likely to change. On the one hand, the MFF provides for a share of the ESF+ and ERDF to be distributed according to the number of third country nationals residing in member states’ territories. More precisely, as the proposal reads, to member states’ respective allocation should be added “a premium of €400 per person per year, applied to the population share of the region of net migration from outside the EU to the Member State since 1 January 2013”36. On the other hand, the MFF provides for the ESF+ to explicitly target third country nationals. Namely, the Commission’s proposal foresees a distinction in roles between the ESF+, aimed at migrants’ medium and long-term integration, and the AMF, aimed at supporting early integration measures37. That said, the ESF+ proposal does not compel member states to earmark the share of their allocation they will dedicate to integration, thus losing the ability to ascertain a share of the ESF+ to be actually used for migrant integration38. In light of the foregoing, the mainstreamed approach to integration as proposed by the Commission does not seem to constitute an expansion of the EU’s integration policy as of yet.
- 39 Over the years, the EU’s budget has stabilised at around 1% of the EU’s GDP. According to forecasts (...)
- 40 See COM (2018) 471 final, articles 8 and 9.
35Altogether, the 2021-2027 MFF, as drafted by the Commission, foresees a significant increase of funds for migration (2.6 times more than with the 2014-2020 MFF), thus indicating a shift in EU priorities, at least from the Commission’s point of view39. With respect to the AMF, the Commission has proposed a total of €10,415 million dedicated to asylum, migration and integration. Whereas the overall structure set with the previous MFF is replicated with the AMF, the new fund somewhat reshuffles the cards by blurring the distinction between management methods. The AMIF provided a clear split between national programmes, to be spent through the shared management method (88 percent of the fund), and actions at the initiative of the Commission, following a direct or indirect management method (12 percent). The AMF, as put forth by the Commission40, proposes that 60 percent of the total amount (€6,249 million) be implemented under shared management; the remaining 40 percent would be allocated to a so-called “thematic facility” implemented through a mix of shared and direct/indirect management. It is not clear as of yet how this mix will operate in practice but, considering the trend established in the course of the two previous MFFs, an increase of the percentage allocated under the direct/indirect method compared to the AMIF is likely. If it were not so, there would, at any rate, be an increase of the absolute sum allocated in this manner, given the increase of the total amount.
36Another significant change rests with the absence of any safeguard as to the use of money between the different strands of the fund, thus even questioning the integration objective of the AMF. The AMIF set a bottom limit whereby at least 20 percent of member states’ allocations should tackle integration and another 20 percent asylum. In its pursuit of greater flexibility, the AMF proposal does not replicate these thresholds. This is a significant step backwards for the development of an EU integration policy in its own right, as it further accentuates the EU’s withdrawal from substantive policy points by taking it to a whole new level.
- 41 See COM (2018) 375 final.
- 42 See Regulation 516/2014/EU.
37From the procedural standpoint though, the discretion enjoyed by central governments is further reduced through the reinforcement of the partnership principle, now completely aligned with the structural funds. In fact, it is not provided for in the AMF proposal itself but higher up, in the umbrella Regulation that covers the provisions common to all the structural funds41. As it stands in the Commission’s proposal, the principle explicitly designates the range of actors necessary to constitute the minimum partnership where the AMIF vaguely referred to “relevant authorities and bodies”42. These actors are: urban and public authorities; economic and social partners; relevant bodies representing civil society. They shall be involved through a Partnership Agreement, drafted together with national governments and submitted to the Commission for approval, and participate in the preparation and implementation of programmes.
38The forthcoming MFF was presented by the EU Commission in May 2018. It is still a proposal and negotiations could still lead to significant alterations, so no conclusion can be drawn as of yet. It remains that the proposal put forth by the Commission, if it grants more discretion to the Union for migration policy-making as a whole, marks a significant step backwards for an EU integration policy in its own right.
39Twenty years after the first reference to an EU integration policy, this article has aimed to shed light on the trajectory this policy has followed thus far. If the need for integration policies is hardly debatable, the role that the EU should have in this respect is more delicate a question. The creation of a common immigration policy with the treaty of Amsterdam eluded the question: integration of migrants was not mentioned in EU primary law. This did not prevent member states from calling for a more vigorous integration policy at the EU level on the occasion of the Summit in Tampere in 1999. Thereafter, the Commission set a policy programme in motion that encountered strong resistance on the part of the member states: if they agreed to collaborate in principle, giving effect to their commitment proved a harder task. A policy on integration unfolded nonetheless.
40Drawing on neofunctionalist theory, this article reconstructs the process that has led to today’s integration policy at the EU level. I have notably focused on the financial schemes explicitly designed for the integration of migrants, in light of the fact that they constitute valuable indicators of the direction, breadth and stringency of a given policy. The passage from the 2007-2013 to the 2014-2020 MFF as well as the debate over the forthcoming 2021-2027 MFF offer a unique opportunity to compare the different phases of development of the policy and assess whether its evolution heads towards a deeper level of European integration. Starting with the 2007-2013 Multiannual Financial Framework (MFF), which gave birth to the first European Integration Fund (EIF), I have shown that the EU has displayed a clear attempt to generate convergence through the design of a fund intended to implement the Common Basic Principles on Integration. The passage from the 2007-2013 MFF to the 2014-2020 one has brought about significant change. Whilst substantive policy directives largely faded with the Asylum, Migration and Integration Fund (AMIF), procedural developments have increased the importance of subnational bodies and that of the EU, much to the detriment of national governments’ discretion. In addition, the share of the fund the Commission is able to use with some level of discretion significantly increased. With the 2021-2027 MFF under negotiation, it is still too soon to ascertain the direction the EU’s integration policy is actually taking. The proposal put forth by the Commission, however, seems to confirm the tendency initiated with the AMIF; that of dissociating substantive and procedural aspects with the former being almost entirely removed and the latter gradually reinforced. Likewise, the share of the fund to be used at the Commission’s initiative would appear set to increase yet again. Accordingly, a likely scenario for the near future would see the EU’s integration policy merged with the larger migration policy. Whilst the EU would lose the capacity to ascertain some of the funding going to integration, it would gain stronger procedural power over the entire migration financial scheme.