Print this article
By Daniel Thym, Research Centre Immigration & Asylum Law, University of Konstanz, Germany.

Trust is an essential prerequisite for a functioning area of freedom, security and justice – as the Court of Justice coined it so well: ‘At issue here is the raison d’être of the European Union and the creation of …, in particular, the Common European Asylum System, based on mutual confidence.’ Our theme is not the controversial case law on fundamental rights, which judges dealt with when emphasising the relevance of mutual trust, but the more generic question of how countries in northern and southern Europe interact when asylum seekers take advantage of the border-free Schengen area to relocate themselves autonomously.

That phenomenon is usually referred to as ‘secondary movements’, even though the Commission evades the term in the ‘pact’, which nevertheless referred to the issue indirectly in the title of the accompanying press release with its call for a ‘balance between responsibility and solidarity.’ When it comes to policy debates among the Member States in the Council, enhanced rules on relocation (solidarity) and the prevention of secondary movements (responsibility) are two sides of the same coin.

EU Asylum Reform: Two Competing Narratives

In the debate about EU asylum policy, we are confronted with two competing narratives which underlie the breakdown of mutual trust among ‘southern’ and ‘northern’ states: While countries at the external border complain about having to shoulder the ‘burden’ without adequate solidarity, politicians further north often decry the alleged incapacity of their peers in running functioning asylum systems and in preventing onward movements. The first narrative is fed by the well-known pictures of arrivals at the southern and eastern shores of the Mediterranean. In relation with the second narrative, German or Dutch politicians, by contrast, will highlight statistical data: throughout the past five years Germany received more asylum applications than Italy, even though the Italian data for the latter include people who later moved elsewhere. Similarly, the numbers for the Netherlands have been between one-third and half of the figures for Greece. Germany made roughly 27 thousand take-back requests under the Dublin III Regulation towards Italy, Spain and Greece in 2019, of which about 3500 or less than 15% resulted in an actual transfer.

This blogpost discusses those elements of the Commission proposals on the reform of EU asylum policy which address the phenomenon of secondary movements. In doing so, it complements the discussion of solidarity measures in the contribution by Francesco Maiani. The initial criticism of laxness on secondary movements by the deputy chairman of Angela Merkel’s CDU/CSU parliamentary group in the German Bundestag with a responsibility for migration policy, Torsten Frei, shows exemplarily how relevant these rules can be for the policy debate. To assess the draft legislation, it is useful to start with more generic comments.

Driving Forces behind ‘Secondary Movements’

A compromise among the Member States is the main hurdle for any reform of EU asylum policy. Without it, no new legislation will be adopted. The focus on political negotiations should not detract us from other challenges, such as the inquiry whether the proposals can function in practice. As explained in the introductory blogpost, multiple aspects of the reform package require more than statutory change. Secondary movements are one area amongst others in relation to which changing the laws does not guarantee success on the ground for the simple reason that Member States and individuals might not comply with statutory obligations (here, p. 103-118).

Social sciences teach us that it can be notoriously difficult to identify the reasons why people leave their home states, how they choose destination countries and in what respect these preferences may change over time, for instance when individuals stay in a first state for several months or years before moving on to a second state, thus turning the initial destination into a ‘transit country’. Multiple ‘push’ and ‘pull’ factors overlap and their relative weight always differs depending on the circumstances. Any assessment is made more complex by the comparatively low level of information on the part of most asylum seekers, which ethnographic research has repeatedly unveiled, thus rendering symbols and stories relevant that are often not controlled by the EU and the Member States.

Notwithstanding these uncertainties, common features define the sociological analysis. Refugees and migrants generally have a low level of knowledge about the specificities of asylum laws or the intricacies of supranational legislation. Statutory details of domestic or supranational asylum legislation, which define the policy debate, will influence decision-making only to a limited extent only. While the level of social benefits can be a factor amongst others, other elements are more important: besides questions of physical safety (which, fortunately, all Member States guarantee as a matter of principle), the economic prospect, including perceived labour market success, and general living conditions are significant. Moreover, ethnic and family networks are a core factor determining where people want to go.

The length of procedures can similarly influence decisions, not least when it comes to the respect for an obligation to leave a country. The longer procedures last, the more likely individuals will have taken roots in a country. Against this background, the administrative inefficiency of the take back procedure under the Dublin III Regulation is one element amongst others, even though the abovementioned factors are certainly more relevant in terms of influencing the behaviour of applicants.

Implications for the Reform Debate

The essential lesson we can draw from the driving forces behind secondary movements is that the EU institutions should strive for a smart legislative design in order to optimise compliance. Note that this is not a normative claim to respect the preferences of asylum seekers as a matter of justice (even though some may want to argue that), but a matter of regulatory self-interest. An asylum system which optimises compliance will work better in practice – and we desperately need a better functioning regime given the dismal performance of the Dublin system.

From the point of view of economic rational choice theory, individual decisions can be influenced by means of either positive incentives or negative sanctions (the proverbial ‘carrots and sticks’). Unfortunately, policy debates about secondary movements are often framed in a binary manner. While NGOs plead for positive incentives, states concentrate on sanctions. This either/or-logic is a false dichotomy: positive and negative incentives can be combined. Doing so will not only facilitate compliance (which is in the interest of everyone), but may facilitate political agreement if different positions coalesce, also among the EU institutions.

Reliable statistics are an important objective to strive for. It is widely known that EU asylum statistics count applications, not persons. That can inflate numbers as a result of double counting and does not provide reliable information about how many individuals are residing in a country at any point in time (or have moved elsewhere). Of course, migration statistics will never be perfect, since authorities will not always track people reliably, but better data can help to rationalise the debate.

The Commission aims at bridging the information gap by upgrading its 2016 Proposal for the Eurodac Regulation. Eurodac is to become a genuine migration and asylum database, allowing Member States to track individuals (instead of counting applications) and facilitating the identification of the state responsible. This initiative will most likely be politically rather uncontroversial – notwithstanding concerns over data protection. The bone of contention will not be the database, but the solidarity mechanism as well as carrots and sticks preventing secondary movements.

Family Life: Recognising an Essential Motivation

In light of the driving forces discussed above, an extended definition of ‘family members’ including siblings and families formed in third states appears as a crucial move, which, moreover, can be considered to respond to the normatively most compelling reason why asylum seekers should be allowed to choose the place of residence (Article 2(g) of the Proposal for an Asylum and Migration Management Regulation). The effects are evident: countries with a residual population of refugees would have to assume responsibility if the legislature recognised what social scientists call ‘chain migration’ along family lines. That move had been proposed by the Commission in 2016 already (here, Article 2(g)), but it was rejected by the Member States. By contrast, extended jurisdiction based on previous stays or studies (Articles 19(4) and 20) will be less controversial.

It seems to me that the significance of the extended family criterion is crucial. Firstly, maintenance of the status quo might quite simply not prevent secondary movements along broader family networks. It is, in other words, a pragmatic recognition of driving forces to change existing rules. Secondly, the initial rejection of the amendment, amongst others by Austrian and German politicians, could possibly be mitigated in the political negotiations. There is ample room for compromise: responsibility for siblings could be counted towards the solidarity quota or it might be accepted politically in return for other changes, for instance the re-introduction of a variant of stable jurisdiction discussed below.

‘Other Carrots’: Incentivising Compliance

While an extended definition of ‘family member’ arguably constitutes the single most relevant incentive, the Commission adds further novelties. In the context of relocation under the solidarity mechanism described by Francesco Maiani, Member States ‘shall’ take into account ‘meaningful links’ when determining which people to relocate (Articles 57(3)(1) and 49(2) of the Asylum and Migration Management Regulation). The proposal does not define the notion of ‘meaningful link’, which, however, can be expected to mirror the original proposal of the European Parliament to base the decision ‘in particular on family, cultural or social ties, language skills or other meaningful links which would facilitate his or her integration into that other Member State’ (Article 24b in conjunction with Article 19(2) of the LIBE Committee’s Report on the Proposal for a Dublin IV Regulation).

It should be noted that the practical relevance of the ‘meaningful link’ criterion depends on whether the legislature extends the definition of ‘family member’ and broadens jurisdiction based on previous stays and studies. In practice, the rules on relocation and, thus, the ‘meaningful link’ criterion applies only to those who are not covered by leges speciales on asylum jurisdiction. Siblings, for instance, would be transferred to other Member States on the basis of the wide definition of family life already, which takes priority over solidarity-based relocation (Article 57(3)(3) of the Asylum and Migration Management Regulation).  The latter could be used, for instance, to consider language skills or, controversially, religious affiliation. It will certainly not prevent secondary movements in itself, but it takes up driving forces why people should be assigned to a specific state.

Another positive incentive is legal onward movement for beneficiaries of international protection with long-term residence status, which the Commission wants to permit after 3 instead of 5 years (Article 71 of the Asylum and Migration Management Regulation).[1] In line with existing rules, the three-year period would be calculated from the date of the asylum application, even though half of the Member States use the statutory option of counting only half of that period (Article 4(2)(3) of the Long-Term Residents Directive as amended by Directive 2011/51/EU). Moreover, long-term residents need to be economically self-sufficient, i.e. onward movement would only be allowed to those who do not depend on social benefits (Article 5(1)(a) of Directive 2003/109/EC). The freedom to live and work elsewhere in Europe would be subject to a number of statutory caveats, which the destination country may prescribe, such as labour market tests (Articles 14-17 of Directive 2003/109/EC). As a result, the amendment could be expected to concern comparatively few people.

It would be possible to broaden the opportunities of mobility for economic purposes during the legislative process. The legislature could consider, for instance, to amend the Blue Card Directive or the Seasonal Workers Directive, which, at present, do not cover beneficiaries of international protection. It could also introduce new rules on labour mobility specifically for refugees. Doing so would not result in ‘free choice’ and could be subject to predefined criteria, such as a work contract with economic self-sufficiency – as explained in an annual report of the German Expert Council on Integration and Migration, of which I am the vice chairperson (here, p. 41-45).

Streamlined Procedures

At present, the Dublin system often results in lengthy and ineffective procedures, which, moreover, often fail in practice. German statistics show that the initial designation of asylum jurisdiction by the authorities takes up to four months, followed by another five months domestic courts currently need to decide on the suspensive effect of appeals. Thus, asylum seekers are legally obliged to return to the state responsible almost one year after having arrived in Germany.

Against that background, it is apparent why the Commission wants to streamline procedures: the takeback procedure is turned into a simple notification subject to much shorter time limits (Article 31), the scope of the Regulation is extended to beneficiaries of international protection and those who had been relocated (Article 26(1)(c), (d)) and legal remedies are considerably curtailed, including a decision on suspensive effect within one month (Article 33(3)(1)). In future, only those challenging the transfer decision on grounds of family links or the real risk of inhumane or degrading treatment are given a legal remedy (Article 33(1)) – in line with the 2016 reform proposals (here, Article 28(4), (5)).

The latter amendment effectively returns to the status quo ante of the Dublin II Regulation, which had similarly provided for legal remedies with suspensive effects under restrictive conditions only. It is to be expected that the ECJ would accept this legislative about-turn for the simple reason that the case law extending legal remedies was based on the contents of the Dublin III Regulation. If the latter is reversed, the case law can be expected to return to the judicial status quo ante. Article 13 of the ECHR does not prevent that restriction, since it applies only to those with an arguable claim that another human right was violated (here, para 288) – a condition the Proposal complies with, since the Commission continuously foresees a remedy in relation to human rights matters under Articles 4 and 8 of the Charter. Article 47 of the Charter does not require further protection when the legislature from limits the scope of statutory rights (here, paras 74-84).

‘Sticks’: Sanctioning Disrespect for EU Rules

In line with our earlier observation that positive and negative sanctions can be combined, the Commission reiterates some of the sanctions which had featured in its 2016 Proposal for a Dublin IV Regulation. Those moving elsewhere will be subject to an accelerated asylum procedure (Article 40(1)(g) of the 2016 Proposal for an Asylum Procedures Regulation read in conjunction with Article 9(1) of the Asylum and Migration Management Regulation). Moreover, it suggests introducing a preclusion period for submitting relevant information (Article 10(2)). The additional express obligation not to engage in secondary movements and to comply with transfer decisions (Article 9(4)(a), (5) of the Asylum and Migration Management Regulation) cannot be expected to change much in practice.

The most significant sanction is the reduction of social benefits in line with the ongoing negotiations on Article 17a of the Reception Conditions Directive. The Commission proposes to introduce, however, a double caveat limiting the scope of the sanction ratione materiae. Firstly, the reduction is to be effective once a transfer decision has been notified to the individual; it will no longer apply automatically whenever someone files a second asylum application in the case of secondary movements. Secondly, the general scheme of the draft legislation indicates that the reduction of social benefits will come to an end, once the jurisdiction is transferred to the second Member State in line with the comments in the next section.

Notwithstanding these caveats, the effects of the reduction in social assistance should not be overestimated. While the level of benefits can influence secondary movements in line with previous comments, other factors are more relevant, including the prospect of labour market success. These broader pull factors, such as the general economic outlook, cannot be influenced by secondary legislation. Moreover, I have explained elsewhere that the Commission is in a regulatory dilemma when harmonising the social rights during the asylum procedure and after recognition, which draft legislation orientates at the level of support for citizens that varies greatly among the Member States (here, p. 1561-62).

Continuation of the Status Quo: Transfer of Jurisdiction

The most surprising novelty of the Pact on Migration and Asylum is the absence of ‘stable’ or even ‘permanent’ asylum jurisdiction. Instead, the Commission retains the individual right to a second asylum procedure (Articles 27(1) and 35(1), (2) of the Asylum and Migration Management Regulation). In particular, the Commission proposes to maintain the six-month rule during which the transfer must take place, while abolishing the three-month time limit for the initiation of the take back notification, which is less relevant in practice. It also proposes to extend the six-month period whenever asylum seekers ‘abscond’, which, in line with ECJ case law, does not cover ‘simple’ scenarios of non-transfer when the authorities know where asylum seekers are living (here, paras 52-65). It seems that in cases of absconding the take back option endures indefinitely.

This move effectively reiterates the legislative status quo under the Dublin III Regulation that allows for double (and threefold) asylum applications in cases of secondary movements. That move is a departure from the 2016 Proposal on a Dublin IV Regulation, which had suggested perpetuating asylum jurisdiction: the failure of the tack back procedure was no longer supposed to entail a transfer of responsibility. Asylum seekers were expected to return to the state responsible instead of receiving a second, albeit accelerated procedure elsewhere. This political about-turn is even more surprising if we remember that both the European Parliament and negotiations in the Council (here, Article 9a) had supported that change.

There is, however, one scenario in relation to which the Commission wants to introduce permanent asylum jurisdiction: beneficiaries of international protection. In contrast to the Dublin III Regulation, they are to be covered by the take back obligation under the new Asylum and Migration Management Regulation (Art. 26(1)(c)), thus extending its scope rationale personae. What is more, the Commission generally excluded beneficiaries of international protection from the cessation rule under Article 27(1)(2). The practical effects of this rule would depend on the length of asylum procedures. It would not cover those whose application was rejected and complementary humanitarian statuses under domestic law, such as the recently reintroduced Italian ‘special protection’. If they moved on, they would still benefit from the transfer of jurisdiction described above.

The question of permanence will feature prominently in the political debate. The German Presidency invited the Member States to comment on this aspect in an informal working paper in the aftermath of the meeting of home affairs ministers on 8 October 2020: ‘In this context, it is also necessary to discuss ways to prevent unauthorised movements, such as stable responsibilities and unbureaucratic transfers, prevention of multiple applications, and in principle granting reception benefits only in the Member State responsible’ (Council doc. WK 10621/2020 INIT).

It seems to me that the continued availability of double asylum applications can be read as an implicit recognition, on the part of the Commission, that – notwithstanding the insistence on a ‘fresh start’ through a ‘new’ pact – the practical effects of the proposals would not differ decisively from the status quo. The solidarity mechanism remains feeble and the novel emphasis on return might not work. Instead, we might be confronted with much of the same as before, albeit under changed circumstances: a bit more solidarity and slightly less secondary movements, which, nonetheless, result in a transfer of jurisdiction to states further north. That may not be the reset button some many had hoped for, but it could be a realistic assessment of what can reasonably be achieved at this juncture.

Conclusion: Overcoming the Vicious Circle

For many years, the Commission had pursued a negotiation strategy based on alleged win-win scenarios: its proposals were presented to satisfy the needs and desires of (almost) everyone. The ‘pact’ follows the reverse scenario. Commissioner Johansson predicted: ‘I will have zero Member States saying it’s a perfect proposal.’ She was certainly correct, as the initial reaction of southern states to the solidarity mechanism and northern states to the phenomenon of secondary movement indicated. However, she might be equally right that the proposal is ‘balanced’ in terms of Member States interests and that they might eventually say: ‘let’s work on this.’

Let’s assume the proposals were adopted without major modification: would the situation be satisfactory in the medium run? I doubt that the cleavages between northern and southern states would be overcome. On the one hand, the solidarity mechanism depends on a quasi-permanent negotiation of state contributions, which could facilitate the legislative procedure but at the cost of continued disputes among Member States about respective solidarity contributions. On the other hand, the new rules on secondary movements cannot be expected to be a game changer. Irregular movements would persist, including the transfer of jurisdiction.

This could result in a vicious circle with continued reciprocal accusations about the lack of solidarity and the persistence of secondary movements. Instead of mutual trust, reticence among Member States would be enhanced. The end result may be convergence on the lowest common denominator: prevent refugees from entering Europe in the first place to avoid poisonous debates about solidarity and secondary movements.

For those who do not want such an outcome, the lesson for the policy debate stands out: overcome the false dichotomy between either more solidarity or less secondary movements and between either positive incentives or negative sanctions to prevent the latter. Try to optimise compliance by designing rules that might work reasonably well in practice. This would not bring about a brave new world of European asylum law, but it may be a pragmatic move to prevent the Common European Asylum System from becoming dysfunctional by means of external closure.

Further Reading

odysseus blog

Evelien Brouwer, ‘Migration flows and the reintroduction of internal border controls: assessing necessity and proportionality’, EU Immigration and Asylum Law Blog of 12 November 2015.


Francesco Maiani, ‘Responsibility Allocation and Solidarity’, in: Philippe De Bruycker/Marie De Somer/Jean-Louis De Brouwer (eds.), From Tampere 20 to Tampere 2.0. Towards a new European consensus on migration (EPC, 2019), p. 103-118.


Markus Wagner, Jimy Perumadan and Paul Baumgartner, ‘Secondary Movements’, CEASEVAL Research on the Common European Asylum System No. 34/2019.

eu acts

Commission Proposal for a Regulation of the European Parliament and the Council on asylum and migration management and amending Council Directive concerning the status of third-country nationals who are long-term residents, COM(2020) 610 of 23 September 2020.

Commission Proposal for a Regulation of the European Parliament and of the Council addressing situations of crisis and force majeure in the field of migration and asylum, COM(2020) 613 of 23 September 2020.