EU Immigration and Asylum Law and Policy Droit et Politique de l'Immigration et de l'Asile de l'UE

POST 15 OF THE SERIES OF THE ODYSSEUS BLOG ON THE PACT ON MIGRATION & ASYLUM

By Daniel Thym, Director of the Research Center Immigration & Asylum Law, University of Konstanz

Complaints about secondary movements lie behind recurrent complaints by ‘Northern’ governments, in the same way as their ‘Southern’ peers express their frustration about the lack of meaningful solidarity. Secondary movements in this sense concern irregular onward movements of asylum applicants and beneficiaries of international protection. Whereas the exceedingly complex and half-hearted provisions on solidarity in the New Pact receive much academic and political attention, the contents and implications of the new rulebook for secondary movements is often neglected despite its practical and political significance.

Both themes had been intricately linked during the negotiations: their interaction presented the Gordian knot any compromise would have to cut. The end result fails to deliver much progress. Inspection of the small print will demonstrate that ‘more of the same’ is the motto of the day. The status quo remains essentially intact once we pierce the veil of legislative innovation. That is why the new legislation will be no game changer on secondary movements, mirroring the daunting outlook on solidarity.

Secondary Movements as an Everyday Reality

Statistics about secondary movements are unreliable (here, p. 4-6). Nevertheless, plausible indicators demonstrate that many move northwards. First, discrepancies between the number of asylum applications and administrative first instance decisions indicate that many do not stay in countries of first arrival (the mismatch is even higher if we include refusal on formal grounds at first instance, such as implicit withdrawal). Second, secondary movements occur after the completion of the asylum procedure; during 2023, Germany took no less than 16.500 asylum decisions on applicants who had received international protection in Greece (here, p. 14), out of 25.000 positive decisions in Greece that year. Third, it is widely known that the take back procedure under the Dublin system has never functioned particularly well (here, p. 157-163). Again, German data shows the extent of dysfunctionality. Italy was obliged to take back 15.514 applicants last year, but only 11 returned (here, p. 23), mostly voluntarily according to media reports; the ratio of success stood below 0.1 %. As a response, the German authorities have called upon EUAA to support its Dublin units in a desperate attempt to achieve better outcomes.

These snapshots deliberately focus on the lived experience of secondary movements, rather than the legitimate sense of frustration about the structural unfairness of the Dublin rulebook. The impasse of the policy debate has always been that both narratives are correct. For years, reciprocal complaints by countries of destination and first arrival about the absence of ‘solidarity’ and ‘responsibility’ have marred any debate about the future of EU asylum policy. To overcome this deep-rooted animosity, the CEAS is in urgent need of meaningful solidarity and better compliance.

A previous reform attempt failed in 2018 because it proved impossible to overcome reciprocal accusations. The Pact has been more successful, partly because the Commission proposed to continue the basic contours of the status quo. That paved the way for a political compromise whose outcome seems unlikely, as we shall see, to change much in practice. The new legislation tinkers with the small print without changing the structure. Corresponding provisions can be found in different pieces of legislation mentioned hereinafter. On one front, however, we may see substantial change. The revised Eurodac Regulation will track movements of migrants, provided they are registered diligently upon first entry and on later occasions (here, Article 12). This may give us a better picture of the intensity and scope of secondary movements once the new legislation becomes effective in 2026.

Driving Forces behind Secondary Movements

Social scientists teach us that there are no simple explanations why people leave home states, how they choose their destination, and to what extent these preferences change over time. Multiple ‘push’ and ‘pull’ factors overlap and their relative weight depends on the circumstances, including networks and ‘smugglers’ (here, p. 96-110). Secondary movements are, in other words, context-dependent, making it difficult to identify replicable patterns (here, here, here). Any assessment of the driving forces is complicated by the comparatively low level of information on the part of many asylum seekers, which ethnographic research has unveiled, thus rendering symbols, stories, and perceptions as relevant as ‘hard’ facts.

Notwithstanding these uncertainties, four conclusions help us assess to what extent the new rulebook will help prevent secondary movements. First, asylum seekers tend to have a low level of knowledge about asylum laws, meaning that the statutory small-print of EU asylum legislation, which defines the policy debate, will influence onward movements to a limited extent only. Individuals will not always have heard about the Dublin system. Second, reception and living conditions are relevant. Substandard reception conditions can be a reason for onward movements, while generous welfare states are one element amongst others rendering a country attractive, although migrants will rarely distinguish between social benefits sensu stricto and the general quality of public services, including education and healthcare.

Two other elements are the most relevant. Third, economic prospect and labour market success, real or perceived, vary significantly within the EU; they are undoubtedly an essential driving force. Finally, ethnic and family networks are generally a core factor determining the choice of destination. People go where relatives and compatriots are staying already. By contrast, all Member States are, fortunately, securing physical safety as a matter of principle, thus rendering this factor irrelevant within the EU—in contrast to movements elsewhere, while the social climate and the degree of xenophobia can differ between countries.

For the political debate, the multitude of ‘push’ and ‘pull’ factors entails that the EU institutions should strive for a smart legislative design which fosters compliance, instead of relying primarily on reactive law enforcement (here, p. 134-135). Note that this is not a normative claim to respect the preferences of asylum seekers but a matter of regulatory self-interest. Putting your regulatory faith in an inter-state transfer procedure is an exercise in ‘technocratic overreach’ in an area without internal border controls (here, p. 103-118). To overcome the dismal performance of the Dublin system, the EU should leave behind the false dichotomy between incentives and sanctions. Both can be combined: the proverbial ‘carrots and sticks’. We shall see that the EU legislature concentrates on the former.

‘Long Live Dublin’: Transfer of Responsibility

It is widely recognised that the basic contours of the Dublin III Regulation will be maintained behind the façade of the Asylum and Migration Management Regulation. This continuity is palpable for the ‘Dublin criteria’ determining the state responsible for examining an application, but it extends to seemingly technical provisions on the transfer of responsibility. The new legislation reaffirms explicitly that secondary movements are illegal (here, Article 17(4); here, Article 9(2)(h), here, Article 27). At the same time, however, onward movements will be effectively ‘legalised’ whenever states fail to return an applicant to the state responsible under the Asylum and Migration Management Regulation within a six-month time frame – exactly the same period as at present (here, Articles 37(1), 46(1), (2); and here, Article 29(2)).

The Commission had little choice but to propose such continuity, given that it had decided to leave the ‘Dublin criteria’ largely intact once quota-based relocation was beyond reach. ‘Frontline’ Member States would not have accepted the prohibition of multiple asylum applications in different Member States without a fundamental reversal of the ‘Dublin criteria’. Despite the grand promise of a ‘New Pact’, the basic contours of the dysfunctional Dublin system remain intact – in contrast to the innovative design of the Commission’s 2016 proposal (here, p. 13-15).

To be sure, there are minor amendments. Responsibility will be transferred after three years when the applicant, or a family member, absconds, instead of 18 months at present (here, Article 46(2)). Crucially, the definition of absconding is broadened significantly to include scenarios where the applicant is absent from a reception centre or fails to present to the competent authorities (Article 2(17)), thus effectively overturing a restrictive earlier CJEU judgment (here, paras 52-65). In scenarios of crisis and force majeure, destination countries benefit from an extended one-year deadline to perform the transfer instead of the normal six months (here, Article 12(2)(d)), whereas countries of first arrival may suspend Dublin transfers to their territory (Articles 12(4), 13) – something they have been doing de facto for years.

Responsibility ‘offsets’ are a novelty in the arsenal of solidarity measures; they were introduced during the negotiations. As a secondary level instrument, offsets become available when Member States do not pledge enough places for relocation (here, Recital 33, Article 63). Destination countries must officially resume responsibility for an asylum application, thus renouncing the—often hypothetical—option to transfer the person to the country of first arrival. Rules governing the activation of ‘offsets’ are complex, as they are the result of intense political debates. While ECRE welcomes them for respecting the choice of destination by asylum applicants, they can be described, from a policy perspective, as the pragmatic recognition of the factual situation. They give the transfer of responsibility, which has been an everyday reality in tens of thousands of cases for years, the aura of inter-state solidarity.

Family Networks: Irrelevance of an Essential Motivation

Family networks are, as we have seen, an essential driver for secondary movements. They are also the motivation which may be politically and ethically most compelling, even if relations with siblings and other relatives are not covered by the human right to family life (here, para 94). Against this backdrop, the Commission had proposed broadening the definition of ‘family member’ to include siblings. In practice, this would have entailed that a Syrian man applying for asylum on a Greek island would have been flown to Belgium where his adult sister is living. That did not happen with the exception of unaccompanied minors (here, Article 25), as well as where states decide voluntarily to assume jurisdiction following a request by the country of first arrival (Article 35(2)). A minor change concerns extensions to reunification with the core family, even if not wider relatives, formed during transit after having left the country of origin (Articles 3(8), 27-28).

Resistance by countries of destination against broadening the definition of family member illustrates how arduous negotiations on secondary movements have been. In practice, many asylum applicants will continue joining relatives anyway, meaning that it can be futile to try to prevent or sanction movements motivated by family ties. EU institutions might have been well advised to recognise the inevitable and invest scarce resources in improving other elements of the erstwhile Dublin system. Like responsibility ‘offsets’, unification with relatives might have been counted as a solidarity measure (here, p. 135-137). That did not happen, thus reaffirming that the new legislation will probably not help curtailing secondary movements.

Small ‘Carrots’ Incentivising Compliance

Non-governmental organisations have long argued that preferences should be respected. To do so may even, within limits, define a smart legislative design for the simple reason that it would foster compliance. In 2017, the European Parliament had proposed taking ‘meaningful links’ into account when implementing a quota-based relocation scheme. The idea survives for relocations under the solidarity mechanism, although its impact is reduced considerably by what is likely to be a limited overall number of relocation pledges (here, Article 67(3)); the same holds for the discretionary clause (Article 35(2)). EU institutions emphasise that respect for meaningful links entails no individual ‘right to choose’ (Recital 26). An additional responsibility criterion for diplomas obtained in a country before an application for asylum elsewhere is poised to concern few people (Article 23).

Another option is to legalise onward movement after status acquisition. Negotiations on the revision of the Long-Term Residents Directive 2003/109/EC, which will resume in the coming months after an earlier deadlock, will have to decide whether beneficiaries of international protection can obtain long-term residence status after three years, instead of five (here, Article 4(1a)). Another element was integrated into the asylum legislation already: the time period of the asylum procedure counts towards the qualification period for long-term residence if applicants await its outcome in the state responsible (here, Article 40). The practical effects of either amendment will be limited. The acquisition of long-term residence status presupposes economic self-sufficiency, and destination countries retain the authority to refuse legal onward movement on the same ground. Few refugees would benefit as a result, since they would have to find a job in Sweden before being allowed entry in that country on the basis of their Greek long-term residence status. Moreover, why should they leave Greece in the first place after having successfully integrated into its economy? Secondary movements mostly concern people who are not economically self-sufficient.

For the same reason, the existing option to apply for a blue card will cover few people. It permits, by way of example, a refugee residing in Spain to receive an entry visa for Ireland if she is highly qualified and has a job offer (here, Article 3(1), (2)(a), (b)). German statistics show that the level of qualification of people entering via the asylum system tends to be comparatively low; it takes years until they are economically self-sufficient, including in lesser qualified jobs. Provided these data are representative, the idea that labour mobility can be a realistic alternative to secondary movements is illusionary, on a larger scale at least. Having said this, such models might be worth exploring even if they concern a minority. As a member of the German Expert Council on Integration and Migration, I co-authored a report describing regulatory options (here, p. 41-45).

Streamlined Take-Back Procedures

Streamlined procedures are generally welcome. By way of example, most take back requests follow a ‘hit’ in the Eurodac database and are approved in the form of silence (here, Article 25(2)). ‘Frontline’ Member States rarely bother to examine incoming Dublin requests in full knowledge that they are often legally correct and that few transfers will take place anyway. That’s why it is pragmatic to replace the former ‘request’ by a ‘notification’ based on simplified anecdotal evidence and subject to a two-week deadline (here, Article 41). However, this change sounds more far-reaching than it is. ‘Hits’ in the Eurodac database will remain the primary evidence; they had been subject to a two-week deadline already.

Somewhat more relevant is the extension of the take back procedure to applicants who had been relocated under the solidarity scheme. Experience with the temporary Relocation Decisions of 2015/16 shows that quite a few asylum seekers did not stay in the state to which they have been transferred, indicating that solidarity-based distribution does not work in practice. In addition, Maiani shows that the new rules will cover beneficiaries of international protection, unlike the Dublin III Regulation. Note that neither change guarantees that the take back procedure will be executed in practice.

There are many reasons why transfers fail, including scarce administrative resources of destination countries and the lack of cooperation on the part of applicants. If, by contrast, countries of first arrival fail to cooperate, the Commission may suspend solidarity measures to their benefit (here, Article 60(3)(4)). This option indicates that the legislature did not have full confidence in the effectiveness of the new system. If this turns out to be correct, the tit-for-tat of reciprocal complaints might continue undermining inter-state trust in the post-Dublin world.

The most important modification concerns legal remedies against the transfer decision. Appeals will be confined to family links, new factual developments, and the real risk of inhumane or degrading treatment (here, Article 43(1)). All applicants will apparently be able to start a legal remedy, with only its scope being limited. As a result, the transfer can only be executed after the expiry of the time limit for seizing domestic courts or when the latter have declined a preliminary injunction (Article 43(2), (3)). Doctrinal analyses suggest that the CJEU will accept the decision of the legislation to return to a modified version of the status quo ante under former Dublin II Regulation No 343/2003 by limiting the scope of individual statutory rights to be compatible with Article 47 CFR (here, p. 163-169; and here).

‘Wishful Thinking’: Highlighting Obligations

It is declaratory from a legal perspective if the new legislation declares secondary movements to be illegal. The small print of the Schengen acquis explains that the absence of internal border controls within the Schengen area does not bring about a legal authority of cross-border mobility, not even for beneficiaries of international protection without sufficient resources (here, p. 316-319). EU legislation puts an emphasis on informing asylum applicants about this illegality and the consequences of non-compliance (here, Articles 18-19). While this may deter some people, information as such is unlikely to have a dramatic effect. Asylum applicants are intelligent enough to grasp the transfer of responsibility as an everyday reality. Secondary movements are poised to continue.

Of critical practical importance are the consequences of non-compliance. The original asylum application in the state responsible will be deemed implicitly withdrawn and any additional application, in the country of first arrival or any another country, will be treated as a subsequent application (here, Articles 41(1)(f) and 55(2) in conjunction with 3(8)). Member States are not obliged, in other words, to start a regular asylum procedure in response to an application after secondary movements.

Unfortunately, rules governing subsequent applicants have been designed without secondary movements in mind in what seems to be a lack of coordination during the drafting process. Subsequent applicants are subject to an admissibility assessment which revolves around the submission of ‘new elements’ which had not adequately been taken into account on the occasion of the first application (here, Article 55(3)). This proviso does not cover information on persecution or serious harm in countries of origin which has been submitted for the first application which is considered implicitly withdrawn; it will not usually constitute a ‘new element’.

Arguably this incoherent drafting should not prevent the asylum authorities from proceeding to the second stage in regular circumstances (see also here, paras 34-43), rather than rejecting the subsequent application as being inadmissible. During the second stage, a regular asylum procedure should be performed. Even if that interpretation was confirmed by the CJEU, the doctrinal uncertainties entail the real risk of misunderstanding and premature inadmissibility decisions, thus undermining the right to asylum enshrined in the Asylum Procedures Regulation.

Controversial ‘Stick’: Withdrawal of Reception Conditions

Countries of destination will not usually be obliged to provide reception conditions to asylum seekers who have engaged in secondary movements (here, Article 18; here, Article 21). A fourfold caveat constrains this drastic move. First, the withdrawal becomes effective once a transfer decision has been notified, not automatically when someone files another application. Second, the general scheme of the legislation indicates that withdrawal comes to an end with the transfer of responsibility, usually after six months. Do we expect the Netherlands or Austria to throw applicants out of a shelter upon the notification of the transfer decision, after they had been obliged to provide reception conditions following the initial application?

Third, the procedural requirements are uncertain as a result of incoherent legislative drafting. The wording of the Asylum and Migration Management Regulation hints at the need for an administrative decision to withdraw reception conditions, including an individualised proportionality assessment (here, Article 18(3), (4)). However, the Reception Conditions Directive indicates that withdrawal can be automatic (here, Article 21(2); also here, p. 269-270). In any case, potential legal remedies against withdrawal are governed by domestic laws (here, Article 29). The additional exception for substandard reception conditions in countries of first arrival is legally superfluous, since a take back decision will not usually be delivered in such scenarios (here, Articles 18(4), 16(3)).

Finally, it remains an open question whether complete withdrawal is compatible with ‘a standard of living in accordance with Union law, including the Charter, and international obligations’ (here, Article 21(1)). ECtHR judgments did not reflect upon whether reception conditions can be withdrawn if they are available in law and in fact in another country (here; here). The CJEU accepted that outcome for EU citizens, albeit without discussing the implications of the Charter (here, paras 88-90; also here). A potential middle ground has been upheld by German courts in light of the constitutional guarantee of human dignity in judgments on EU citizens: authorities must provide support for an interim two-week period and pay for voluntary return to the country where social benefits are available. That rule will be extended to Dublin cases within Germany in the coming months in a swift implementation of the new Reception Conditions Directive.

Conclusion

‘Wishful thinking’ has defined the debate about solidarity and responsibility for years. Insistence on structurally unfair ‘Dublin criteria’ contrasts with the reality of secondary movements. Non-compliance similarly defines the lived experience of ineffective transfer procedures. We can expect that the post-Dublin world will look similar. Secondary movements will continue and they will often be ‘legalised’ following the transfer of responsibility, contrary to the solemn promise that the ‘New Pact’ will overcome the structural imbalances which have haunted asylum policy for years.

The Commission is correct to focus on implementation these days. Even so, bona fide implementation might turn out to be insufficient for the simple reason that the new legislation fails to address the ‘original sin’ of the erstwhile Dublin system: allocating asylum jurisdiction to one country in an area without internal frontiers. While the basic contours of the legislation stay intact, several amendments epitomise the desire to prevent secondary movements through quite some sanctions and limited incentives. In light of the drivers discussed at the outset, these ‘carrots and sticks’ are unlikely to have a lasting impact. It seems inevitable that we will see ‘more of the same’ with regard to secondary movements.