Solidarity under EU asylum policy with the New Pact on Migration and Asylum: Deal or Grail?
By Professor Philippe De Bruycker, Université Libre de Bruxelles
In recent decades, the EU has realised that having common borders means so much more than having a customs union: it also entails sharing responsibilities and burdens when it comes to asylum and migration issues. How has such solidarity manifest itself on paper and in practice? Philippe De Bruycker is a professor in the Institute of European Studies of the University Libre de Bruxelles and coordinator of the Odysseus Network for Legal Studies on Migration and Asylum in Europe. He holds the Jean Monnet Chair for European Law on Immigration & Asylum and has published extensively in this area. In this Long Read he analyses how solidarity has forged EU asylum policies and assesses how it should do so to address the EU’s future migration challenges.
Tackling migration issues alone or together?
More than thirty years ago, the EU made a significant choice with the conclusion, in 1990, of the Schengen agreement and the Dublin convention. These two international instruments, which later became EU law, rely on each member state taking individual responsibility rather than on solidarity between them, as I explain in the first part below. After 20 years of solidarity being considered a non-issue, member states were eventually prompted to face this debate by the refugee crisis in 2015, when about one million people entered the EU through Greece. In 2020, the European Commission’s New Pact proposed a flexible solidarity mechanism that has led to a harsh political debate. This shows that the quest for solidarity on asylum policy is one of the most difficult issues facing the EU, as I discuss in the second part of this article.
The choice against solidarity
Solidarity is nowadays one of the hottest topics in relation to EU asylum policy. It is headline news and discussed at length by the Council of Ministers, which managed to reach an agreement in June 2023 after years of discussions. It has even become the subject of debate as high as European Council level. This is a new development: despite the congenital defects of the Schengen area, solidarity had been a non-issue for the preceding 20 years.
The congenital defects of Schengen and Dublin
The Schengen Convention of 19 June 1990 is an invention of northern member states while the Dublin Convention of 15 June 1990 was clearly inspired by Schengen, which it replaced to allow the UK to join the system for determining responsibility for asylum applications while remaining outside the Schengen area. The problem with both these Conventions stems from the fundamental political choice made at the time to leave it up to each member state individually to deal with the external borders and asylum applications.
The southern member states are therefore legally obliged to, on their own, control their part of the external borders and examine the asylum applications lodged on their territory. This is problematic because the burdens are asymmetrical in the EU. Greece and Italy in particular are the member states most exposed to migration flows because of their geographical location. However, the same is not true of the eastern and northern member states, which, in principle1, have very limited responsibilities due to the absence of land borders with non-EU countries and only a limited number of asylum-seekers arriving directly on their territory.
The way the responsibilities are allocated among EU member states is thus extremely unequal and therefore flawed. It is surprising that such a principle was ever devised as it is totally unfair and therefore creates a huge risk of ineffective application by the most burdened member states. Its adoption is explained by the fact that the Schengen system, and the Dublin system it inspired, were the product of five northern member states (France, Germany and the three Benelux countries) within an intergovernmental framework and was subsequently imposed on the others once the Schengen and Dublin rules became EU law (2). There was almost no mechanism of solidarity accompanying these rules, leaving the southern member states to face their burden alone. It is a real surprise not only that such mechanisms were ever invented, but that they remain applicable in the EU today, 33 years after their adoption. This is because, until the crisis of 2015/16, solidarity was considered a non-issue.
Solidarity as a non-issue
Following the entry into force of the Amsterdam Treaty (Article 63(2)(b)), measures on refugees and displaced persons were supposed to be adopted within five years to ‘promote a balance of effort between member states in receiving and bearing the consequences of receiving refugees and displaced persons’. The Tampere conclusions adopted by the European Council in 1999 contained only a vague reference to an ‘agreement on the issue of temporary protection for displaced persons on basis of solidarity between Member States’ (point 16). The legal rules on solidarity (3) included in the Temporary Protection Directive (4) are weak and vague. Article 25 of Directive 2001/55 required member states to receive displaced persons ‘in a spirit of solidarity’, leaving it up to them to their reception capacity ‘in figures or general terms’. If the number of displaced persons exceeds their capacity, ‘The Council shall examine the situation and take appropriate action’. It is interesting to observe that the activation of this directive for the first time in 2022, due the influx of millions of Ukrainian refugees resulting from Russia’s aggression, has revealed the huge discrepancies between the number of displaced persons received by member states (5). In particular, it has shown that giving refugees free choice of destination member state leads to unfair distribution among them.
The authors of the Lisbon Treaty took the issue of solidarity seriously in 2007, not least by including Article 80 of the Treaty on the Functioning of the European Union (TFEU), which states: ‘The policies of the Union set out in this Chapter and their implementation shall be governed by the principle of solidarity and fair sharing of responsibility, including its financial implications, between the Member States. Whenever necessary, the Union acts adopted pursuant to this Chapter shall contain appropriate measures to give effect to this principle’. The language used is strong: it is no longer about ‘promoting’ solidarity as it was under the Amsterdam Treaty, but rather implementing it in EU legislation based on a clearly binding provision of primary law.
However, there have been discussions about its value (6), as the principle of solidarity and fair sharing of responsibility is rather vague and gives EU lawmakers much room for manoeuvre. Be that as it may, the provision nevertheless legally requires solidarity measures leading to ‘a fair sharing of responsibility’, according to its interpretation by Lilian Tsourdi (7). It goes further than the idea of burden sharing, often debated under international refugee law. If indeed what is fair cannot be defined precisely, a system like the Dublin one is clearly contrary to that provision as it is totally unfair (8).
Moreover, the Dublin system has not been accompanied by solidarity measures aimed at counterbalancing its unfair distribution of responsibilities as required by Article 80 TFEU. Though several measures have been adopted, their scope remains limited in comparison to the scale of the issue. This is because solidarity was not actually taken seriously until the crisis of 2015/16, despite the Lisbon Treaty adding Article 80 to the TFEU. It was before treated as a non-issue based on the mantra that Dublin — considered the cornerstone of the Common European Asylum System (CEAS) — is a responsibility mechanism, not a solidarity one.
The quest for solidarity
In the early days of EU asylum policy in the 2000s, some solidarity tools were nevertheless developed, but they remained modest. The crisis of 2015/16, when about one million asylum seekers entered the EU mainly through Greece, was the catalyst for mandatory solidarity with the adoption of relocation measures. Under the New Pact on Migration and Asylum, presented in 2020, the Commission proposed flexible solidarity.
The early days during the 2000s: modest solidarity
The Commission published a Communication on ‘enhanced intra-EU solidarity in the field of asylum’ in 2011. It insisted on practical cooperation through the European Asylum Support Office (EASO) (which later became the European Union Agency for Asylum (EUAA)) to assist member states that face particular pressure and on enhancing the added value of financial solidarity. Abandoning its 2008 proposal to temporarily suspend Dublin transfers when member states face a particularly urgent situation that places an exceptionally heavy burden on their asylum systems, the Commission eventually supported the idea of creating an early warning mechanism to prevent crisis. This proposal became a reality under the Dublin III Regulation in 2013, but such suspensions have never been used. The Justice and Home Affairs Council closed the process on 8 March 2012 by adopting conclusions ‘on a common framework for genuine and practical solidarity towards member states facing particular pressures due to mixed migration flows’. These were nothing more than a catalogue listing all the existing solidarity measures adopted to date. Two solidarity tools — funding and agencies — were developed during this period, while a third — relocation — emerged tentatively.
EU funding
Funding was the first solidarity tool used, with the creation of the European Refugee Fund (ERF) in 2000. Solidarity was even the first goal mentioned by ERF I and ERF II in 2004. The third ERF, adopted in 2007, was part of a framework programme explicitly entitled ‘solidarity and management of migration flows’. Regulations adopted in 2014 and 2021 respectively incorporated the ERF into one overarching fund — the Asylum, Migration, and Integration Fund (AMIF). The amounts available increased very significantly, from €216 million for the 2000–2004 period to almost €10 billion for the 2021–2027 period. The latter figure includes not only the budget allocated to asylum policy, but also to migration and integration.
However, EU funding remains modest, representing around 2% of the member states’ total expenses for their national asylum policies over the 2007–2013 period, when €614 million was allocated to the ERF (9). Moreover, the funds allocated to asylum are distributed on the basis of absolute numbers (of asylum seekers and protected persons in each member state) rather than, for example, their population or GDP as a proportion of the EU total. This is regrettable because the latter approach would multiply the solidarity effect of the EU funds available (10).
The redistributive effect of the EU funding cannot be denied — for instance, Greece, which is facing particular pressure in the area of asylum, received more (11.3% of the total funding for the 2014–2020 period) than Germany, the biggest member state (which received 9.1%) (11).However, it remains clear that financial solidarity at EU level does not compensate significantly for the unfair distribution of burdens between the member states (12).
EU agencies
Agencies were the second solidarity tool introduced, with the creation of Frontex in 2004 and the European Asylum Support Office in 2010. Though rarely presented as such, agencies are tools aimed at creating operational or better institutional solidarity, created to support member states. They create solidarity by virtue of being funded by the EU budget rather than by the member states. The best example of such solidarity is Frontex’s deployment of border guards or the EASO’s deployment of asylum officials to benefiting member states. These officials’ costs are paid by the agencies (vertical solidarity from the EU towards member states) while their salaries continue to be paid by their member state of origin (horizontal solidarity among member states). These agencies’ were set up to benefit, in particular, those member states in need of solidarity (13).
European relocation
The third solidarity tool, relocation, which emerged tentatively, concerns the transfer of beneficiaries of international protection (or even of asylum-seekers, as we will see below) from one overburdened member state to another. It is therefore a kind of resettlement scheme — already a familiar concept in international refugee law — but one that is internal to the EU. Initially, relocation was based on ‘double voluntariness’, meaning that both the person being relocated and member state to which they were being relocated had to agree to the transfer. A pilot project called ‘European Relocation Malta’ (Eurema) ran from 2009 until 2013. It involved a very limited number of protected persons, with fewer than 600 relocations (14).
Overall, only modest progress was made on solidarity in the 2000s. This is not surprising as, during this period, member states insisted continuously on responsibility, requiring all member states to honour their obligations under EU law, including the Dublin regulation with its unbalanced way of allocating responsibilities among member states. The Bratislava declaration epitomised this view despite having been adopted by the European Council as late as 2016. The declaration tries to reconcile solidarity and responsibility as two opposing political principles, whereas Article 80 TFEU is about one single legal principle of solidarity and fair sharing of responsibility (15) (see above).
The crisis of 2015/16: mandatory solidarity
The crisis of 2015/16, during which around one million of migrants and asylum seekers, including many Syrians, entered the EU from Greece after having transited through Turkey, led the Commission to adopt the EU agenda on migration in May 2015 (16). The Commission launched the idea to temporarily (for a period of two years) use the relocation mechanism to show solidarity with the overburdened member states. This led to the adoption, by the Council, of two decisions on 14 (17) and 22 (18) September 2015 aimed at relocating a total of 160 000 asylum seekers from Greece and Italy.
This time the system was based on ‘double mandatoriness’ — neither the asylum-seeker nor the member state to which they were to be relocated could refuse relocation. Interestingly, the number of people to be relocated was shared the member states using a distribution key based on population (40 %), GDP (40 %), resettled refugees (10%) and unemployment rate (10 %). The system did not produce very good results (19): only 34 700 asylum seekers were actually relocated (25 % of the initial target). The main explanation for this failure is that, to be eligible, asylum seekers had to come from a country of origin from which at least 75 % of applicants were granted protection.
However, this system triggered a political crisis within the EU as it was adopted in Council by a qualified majority, putting some eastern member states in the minority. Refusing forced relocation, Slovakia, and Hungary, supported by Poland, asked the EU Court of Justice to annul Decision 2015/1601. It refused to do, but without emphasising the principle of solidarity and fair sharing of responsibility in its reasoning (20).
In May 2016, the Commission adopted a proposal to reform the Dublin system. The proposal did not envisage reforming the Dublin system but rather accompanying it with a corrective mechanism based on mandatory relocation. Each member state would be responsible for its share of asylum seekers, calculated on basis of a reference key made up of population (50%) and GDP (50%). The system would be automatically triggered when the number of asylum seekers for which a member state is responsible exceeded 150 % of its fair share. The proposal included the possibility for member states to buy their way out of the relocation system by paying a sort of fine of €250 000 per asylum seeker allocated to them. The Council never adopted a common position on this proposal to enter into negotiations with the European Parliament (21).
In 2015 and 2016, EU funding was increasingly used, through the emergency reserve, to help member states deal with the crisis . Emergency funding amounted to almost €546 million during this period, around 20% of the total AMIF (22). This temporary form of financial support for member states showed that they were no longer capable of implementing EU migration and asylum policies during a crisis and that the EU was itself starting to implement these policies at national level by providing funding.
During the same period, Frontex was strengthened and became the European Border and Coast Agency in 2016 (23). The EASO later underwent a similar evolution, being replaced by the EU Asylum Agency in 2021 (24). Both agencies became active in hotspots in Greece and Italy, where the experts they deployed were tasked with the identification, registration, fingerprinting and debriefing of migrants (25).
Around 5 000 people were relocated from Greece between 2020 and 2022 based on the voluntary relocation scheme put in place under the plan of immediate measures to support Greece. In June 2022, 18 member states adopted the ‘modus operandi on the voluntary solidarity mechanism’. This declaration was supposed to pave the way for the adoption of the New Pact on Migration and Asylum. However, as of December 2022, the number of transfers had been very limited at 207 (26).
The New Pact of 2020: flexible solidarity
The Commission proposed the New Pact on Migration and Asylum (27) in 2020. The issue of solidarity was this time at the core of this enormous legislative package that included a proposal for an asylum and migration management regulation (AMMR). After the failure of its 2016 proposal, which was based on relocations and the choice to keep the Dublin system but accompany it with a solidarity mechanism, the Commission added the idea of flexibility to mandatoriness.
As the legislative process continues at the time of writing (October 2023), I base my analysis on the common position adopted by the Council on 13 June 2023. This position will be the basis for negotiations with the European Parliament on the adoption of the regulation, which is planned for spring 2024, just before the European Parliament elections.
Types of solidarity
In the Commission proposal for the AMMR, member states can choose between three types of solidarity measures besides relocation: return sponsorship, capacity building measures and operational support. The aim of introducing this flexibility was to accommodate the opposition to relocation expressed by the four Visegrad Group member states (the Czech Republic, Hungary, Poland, and Slovakia) by offering them other ways for showing solidarity with the other member states. The idea of sponsored return has been dropped by the Council, so the choice remains between two main types of physical (relocation) and financial solidarity, plus alternative measures (capacity building, services, staff support and technical equipment). Member states have full discretion to choose between these forms of solidarity.
Relocations concern asylum-seekers but could also concern beneficiaries of international protection if agreed bilaterally between the contributing and benefitting member states. Member states can buy their way out of the solidarity mechanism by paying €20 000 per asylum-seeker. Compared to the amount of €250 000 envisaged in the 2016 proposal (see above), which seemed like a fine, the new amount looks more like a solidarity contribution to the system. Financial solidarity is about direct financial help from the contributing member state for a specific project proposed by the benefitting member state. The projects to be funded concern migration, border management, asylum and even projects ‘in third countries that may have a direct impact on the flows at the external borders or may improve the asylum, reception and migration systems of the third country concerned’.
Deciding which member states will benefit
Solidarity is intended to benefit those member states under ‘migratory pressure’, defined as ‘a situation generated by the large number of arrivals of third-country nationals placing a disproportional burden on member states even on well-prepared asylum and reception systems, taking into account the overall situation in the Union and the specificities of the geographical location of a member state’. It is important to note that the identification of member states under migratory pressure is not based on figures such as population and GDP, as it was in the 2016 Commission proposal (see above). Instead, it is based on an assessment by the Commission. This assessment is based on a very large number of parameters, such as the number of asylum seekers, beneficiaries of international protection or migrants staying illegally, return decisions and returnees removed, and the number of people apprehended while attempting to cross the external borders irregularly. It also takes account of many other diverse elements such as the benefitting member states’ needs assessment, capacity and preparedness measures, their level of cooperation in the area of returns and readmission, the geopolitical situation in relevant non-EU countries and root causes of migration, the results of Frontex’s vulnerability assessments, and the scale and trends of secondary movements between member states.
This long list of elements to be considered by the Commission indicates that the choice has been made to evaluate the situation in the benefitting member states from a general point of view, rather than purely in terms of asylum. Moreover, the evaluation will cover not only the needs of the member states requiring solidarity but also the way they assume their responsibilities. This is evident in the way the principle of solidarity and fair sharing of responsibility provided for in Article 5 is defined in the Council’s common position on the AMMR. The definition includes many elements relating to responsibility (efficient national asylum and migration management systems, prevention and reduction of irregular migration, correct and swift application of the Dublin rules and measures to prevent and reduce unauthorised secondary movements between member states) rather than solidarity.
The Commission will have huge discretionary power to decide which member states are under migratory pressure, through a European migration management report. In this report, it will assess the situation along migratory routes, leading to the adoption of decisions identifying member states under migratory pressure.
The determination of the share of member states
Based on this report, the Commission will adopt a recommendation setting up a solidarity pool by taking into consideration relevant qualitative and quantitative criteria such as the number of arrivals, the average recognition rate, and the average return rate as quoted by Article 7c(3) of the Council’s common position. The pool must comprise a minimum of 30 000 relocations and €600 million for direct financial contributions.
Member states will pledge their contributions to the solidarity pool during a high-level migration forum bringing together their representatives. These contributions must be made using of a distribution key, based 50% on population and 50% on GDP. The Council of Ministers will eventually adopt the solidarity pool on basis of the Commission’s recommendation.
Future — towards ‘structural’ solidarity?
If the solidarity mechanism is adopted as planned in 2024, time will tell whether the Commission will be able to use its discretionary power by building a climate of consensus between benefitting and contributing member states. Implementing the system in practice will also pose a difficult challenge. Implementing forms of physical solidarity such as relocation of asylum-seekers is cumbersome when it must be done against their will, creating the risk of secondary movements between member states persisting, which will in turn feed distrust among member states. It also remains to be seen how the heavy system of financial solidarity relying on intergovernmental agreements between benefitting and contributing member states will work in practice.
The key question will of course be whether enough solidarity will be established towards countries designated as benefitting member states under the New Pact on Migration and Asylum to enable them to face their enormous responsibilities of controlling EU land borders efficiently, screening migrants, implementing asylum and return procedures at the borders, and applying the Dublin system rigorously. In particular, it remains to be seen to what extent the additional costs will be covered by EU funding (28). Though the target of 30 000 relocations per year is double the number achieved during 2015/2016 refugee crisis, it is very limited in comparison with the one million asylum applications submitted in the EU in 2022 (29). By contrast, the minimum amount of €600 million per year for funding solidarity projects seems more ambitious.
Finally, it will not only be about a question of the frontline member states’ willingness to implement these policies, but also their capacity to do so. If the solidarity system fails, the EU will once again face the same risk of chaos as it did during the 2015/2016 crisis. This could lead to the collapse of the Schengen area, to more member states failing to abide by the principle of not returning refugees to countries in which they face danger (‘non-refoulement’), and to a temptation for the EU to outsource its obligations under asylum law even more than it is doing at present. The alternative to such a dark scenario would be greater EU integration, with EU agencies gradually taking over member states’ role in implementing border and asylum policies, funded by an increasing share of the EU budget (30).
This article was first published on the 2/2023 issue of the ECA Journal. The contents of the interviews and the articles are the sole responsibility of the interviewees and authors and do not necessarily reflect the opinion of the European Court of Auditors.
(1) There are exceptions, however, e.g. due to the massive influx of Ukrainians into these member states following the Russian attack on Ukraine in February 2022.
(2) On the negotiation of the Dublin II Regulation 343/2003, see AUS, P., Logics of decision making on Community Asylum Policy, A case study of the evolvement of the Dublin II regulation, Arena Working Paper n°3, 2006; see also Tsourdi, L. and Costello, C, The evolution of EU law on refugees and asylum, in: CRAIG, P. and DE BURCA, G., OUP, 2021, Evolution of EU law, pp.807–808.
(3) See Solidarity in EU Asylum Policy, European Parliamentary Research Service (EPRS), January 2023, PE 649.344.
(4) Directive 2001/55.
(5) As of July 2023, according to Eurostat, out of 4 114 320 beneficiaries of temporary protection, 971 080 are in Poland and 357 540 in the Czech Republic, compared to 67 350 in France.
(6) And also whether it can it be used as a proper legal basis or only in combination with other ones (see about this controverse, European Parliament Research Service (EPRS), Solidarity in EU Asylum Policy, briefing, January 2023, pp. 7.
(7) Tsourdi, L., Solidarity at work? The prevalence of emergency-driven solidarity in the administrative governance of the Common European Asylum System, Maastricht Journal of European and Comparative Law, 2017, Volume 24, n° 5.
(8) See Kucuk, E., The Principle of solidarity and fairness in sharing responsibility: more than window dressing?, in: European Law Journal, 2016, Volume 22, No 4, p.448.
(9) Maiani, F., Sans solidarité point d’UE, Revue Trimestrielle de Droit Européen (RTDE), 2012, volume 48, n° 2, p.319, note de bas de page n° 105. For more details, see D’Alfonso, A., External border control and asylum management as EU common goods: a budgetary perspective, EUI working papers, RSCAS 2019/05.
(10) Distribution on a relative basis would imply that smaller or poorer member states taking up large numbers of asylum seekers would receive more funding.
(11) EU funds for migration, asylum, and integration policies, European Parliament, Policy Department for Budgetary Affairs, PE 603.828, April 2018, p. 18.
(12) Monar, J., Solidarity as a challenge for the EU: the case of justice and home affairs, in EU Studies in Japan, 2015, Volume 35, p.15.
(13) See Article37(2) of Frontex Regulation 2019/1896 and Article 1(2) last indent of EASO Regulation 2021/2303.
(14) According to on the fact-finding report on intra-EU relocation activities from Malta of the European Asylum Support Office.
(15) See De Bruycker, P., and Tsourdi, L., The Bratislava Declaration on Migration: European irresponsibility instead of solidarity, blog EU Migration and Asylum Law and Policy, 27 September 2016.:
(16) For a chronological analysis of this crisis, see Greubel, J., Towards a profound European Asylum System, On EU governance during the refugee crisis, in: Wurzburger Jean Monnet Papers, 2018, n°2.
(17) Council Decision 2015/1523.
(18) Council Decision 2015/1601.
(19) See Implementation of the 2015 Council decisions establishing provisional measures in the area of international protection for the benefit of Greece and Italy, European Parliament, Policy Department Citizens’ rights and constitutional affairs, PE 583 132, 2017
(20) See van der Mensbrugghe F.R., Crise migratoire: entre solidarité et minimalisme, observations sous CJUE 6 septembre 2017, Affaires jointes C-643/15 et C-647/15, Slovaquie et Hongrie c. Conseil, in : Revue de la Faculté de droit de l’Université de Liège, 2017, numéro 1, p. 121.
(21) For a critical analysis, see The Reform of the Dublin III Regulation, Study for the Libe Committee, European Parliament, Directorate General for Internal Policies, Policy Department citizens’ rights and constitutional affairs, 2016, authored by F. Maiani.
(22) Tsourdi, L., Solidarity at work? The prevalence of emergency-driven solidarity in the administrative governance of the Common European Asylum System, Maastricht Journal of European and Comparative Law, 2017, Volume 24, n° 5.
(23) Regulation 2016/1624, later replaced by Regulation 2019/1806.
(25) See ECA special report 06/2017: “EU response to the refugee crisis: the hotspot approach”.
(26) See the operational conclusions of the solidarity platform pact of 1 December 2022.
(27) See Commission Communication COM(2020)609 of 23 September 2020. About the new pact, see Thym, D., Reforming the Common European Asylum System: Opportunities, Pitfalls, and Downsides of the Commission Proposals for a New Pact on Migration and Asylum, Nomos, 2020; The European Commission’s new pact on migration and asylum, horizontal substitute impact assessment, European Parliamentary Research Service, PE 694.210, 2021; EU Pact on Migration and Asylum: state of play, European Parliamentary Research Service, PE 739.247, June 2023.
(28) See Goldner-Lang, I., Financial aspects of the EU’s new pact on migration and asylum: towards stronger EU-funded policy implementation?, p.239.
(29) EU Asylum Agency, Latest Asylum Trends, January — June 2023 Review, September 2023.
(30) See Tsourdi, L., The emerging architecture of EU asylum policy: insights into the administrative governance of the Common European Asylum System, in: EU law in populist times; Crises and prospects, Cambridge 2020, p. 191.
