What remains 'common' in the “European Asylum System” if Dublin fails?

16 April 2018

Just recently, discussions on the future of the Dublin Regulation have come to a halt. The Bulgarian presidency reacted by installing an expert group to elaborate a zero draft on the future of the Dublin system. To recall, the Dublin Regulation is one of the core instruments of what is altogether referred to as the “Common European Asylum System (CEAS)”. All key CEAS instruments (the Asylum Procedures Directive, the Qualification Directive, the Reception Conditions Directive as well as the Regulation on the European Asylum Support Office (EASO)), are under discussion again along with the Dublin Regulation. The new legal framework shall find an agreement by the end of the Bulgarian presidency in June 2018.

Currently there are two rather opposing proposals on the table: the European Commission proposal builds on the Dublin III Regulation by maintaining the so-called “first country responsibility” principle and adds a ‘corrective allocation’ system, which would be triggered once an EU MS received 150% of asylum seekers according to an EU wide “asylum seekers allocation quota”. The other proposal has been tabled by the European Parliament. It aims to end the “arbitrary” system of determining the responsibility based on the geographical location in which those MS at the external borders face disproportionate pressure. Instead, the European Parliament proposes to distribute asylum seekers across the EU according to a set distribution key. Thus, in this proposal, the EC’s ‘corrective allocation’ mechanism becomes the permanent feature of the system. Now, it is up to the Council to react in one or the other way on an issue that is intensely debated among scholars but even more among EU Member States, often in diametrically opposed directions. Border countries, in particular Greece and Italy, strictly refuse any system that (again) places the biggest portion of responsibility on them. The Visegrad group is known to oppose any mandatory distribution scheme. Countries having taken in high numbers of asylum applicants in the last years, would wish to see more solidarity. The rest of the EU countries were comparably less affected by refugee flows and may find no reason to change a system they benefitted from in the past.

The Dublin system has often been referred to as the “cornerstone” of the Common European Asylum System suggesting that the CEAS could not be maintained without Dublin. In fact this contains some truth even though in a different way than the statement suggests.

A Common European Asylum System without Dublin?

If we look back at the development of an EU regional asylum system, harmonisation has been utilised as a basis to realise the EU asylum project, launched by the European Council in 1999 in the Tampere Conclusions. It was soon labelled the “Common European Asylum System” suggesting its ultimate goal would be an emerging of a new “system” that is “common” within the European Union. Now, almost 20 years later, neither did a “new system“ emerge, nor did it become a “common” one. Instead, we still have 28 (+) national asylum systems, the practices and legal standards of which still vary as it has been critisised by many scholars and practitioners.

The legal instruments to achieve the “Common European Asylum System” already showed considerable conservatism as harmonisation of national systems has been chosen over unification: the majority of legal acts forming the CEAS are EU Directives aiming at “harmonising” different existing state practices such as the asylum procedure, the reception conditions or the qualification of asylum seekers as refugees or beneficiaries of subsidiary protection. Unification however, had to be used as a tool to develop those procedural areas which by then were not needed in any EU MS, namely the determination of responsibilities among and between EU MS (the Dublin System) and the exchange of fingerprints that would operationally support the implementation of the Dublin system through the Eurodac database. Without national schemes and practices, a directly applicable regulation had to fill this specific legal vacuum. Ironically, the Dublin procedure (and the associated Eurodac regulation) emerged as the only truly “common” and “European” element of the EU asylum system.

Why did unification work in the Dublin System?

The Dublin System introduced a concept which evolved in the early 1990s, namely the first country of asylum or the safe third country concepts. The understanding that the first country of asylum shall be considered responsible is partially based on an interpretation of Article 31 of the 1951 Geneva Refugee Convention. It states that refugees should not be punished for illegally entering a country if they are arriving directly from a country where they were under threat. The notion was also formally put forward in the context of the 1977 Diplomatic Conference on Territorial Asylum: back then, Denmark proposed that it would be reasonable and fair that asylum seekers should be called upon to request asylum from that State where it appeared that they already had a connection or close links with. UNHCR ExCOM determined though that asylum should not be refused solely on the grounds that it could have been sought elsewhere, but, however, also recognised a certain responsibility of the first country if asylum seekers and refugees move further from a country where they already have found protection.

While it is undisputed that states have to take back their own citizens, less clear are obligations towards third country nationals. The literature suggests that there is no such obligation to take back third country nationals. Principles like the “safe third country” are only unilateral concepts that, without an additional multi or bilateral agreement cannot develop any obligation to readmit third country nationals. As Moreno-Lax put it, “the notion of the safe third country concept constitutes a unilateral declaration by the removing State of the obligations owed by another country on implicit premises inferred from the terms of the 1951 Convention, but without the express agreement of the country in question to such interpretation”.

This expressed agreement to take back third country nationals is commonly vested in so-called “Readmission agreements” which – following Coleman – “facilitate the expulsion of unauthorised immigrants by establishing obligations and procedures regarding readmission between contracting parties.” The readmission concept exists since the beginning of the nineteenth century. In the beginning of the 1990s readmission agreements became a central policy instrument to manage migration flows. Initially European countries used readmission agreements amongst themselves, and only later the emphasis shifted to the readmission of migrants to transit countries outside the European Community. However, a specific situation refers to people in search for international protection. Asylum seekers who are returned to a country which is not their country of origin might risk being deprived of the possibility to submit an asylum application or to have it examined in substance, thus becoming subject to so-called “chain-refoulement”, or placed in an unsustainable situation in terms of social rights. Thus, the readmitting country not only needs to agree to take back an asylum seeker, but it also needs to implement standards and provide protection from refoulement. The Dublin system therefore is the regional application of the safe third country principle among the member states of the European Union. Without this complex Dublin system both the safe third country principle as well as the first country of asylum principle would remain toothless concepts within the EU because of the lack of consent of (safe) third (or first) countries of asylum to take back asylum seekers who moved on from their territories.

The EU concept – the Dublin System

The Dublin responsibility system is mainly based on the principle that the Member State through which the asylum seeker first entered the EU is preliminarily responsible for processing asylum applications submitted on its territory (although there are also other, higher, hierarchical criteria to determine the responsibility such as the best interest of the child or family unity, etc.). The EU approach to the determination of responsibility is strongly orientated around the principle that as soon as an asylum seeker reaches any EU country he or she is supposed to seek and find protection there. Dublin provides clear rules on how to proceed if an asylum seeker or irregular migrant moved on from a first country where he/she has entered irregularly or has asked for international protection. This comprises a specific procedure to determine responsibility, executing it if necessary by force, and to return the person to the responsible state, i.e. in most cases, the first EU country where the person first entered EU territory.

Since its creation, the Dublin system has been strongly criticised. It is said to be ineffective and inhumane, as well as time-consuming and cost-intensive (see as an example here, here and here). In addition, national and supranational courts have stopped Dublin returns to certain EU Member States in several cases, also in response to case-law of both the ECtHR and the CJEU, in particular because of insufficient conditions in the admission, accommodation and care of refugees, directly referring to Art 3 ECHR respectively Art 4 CFR, and the prohibition of torture and inhuman treatment as an impediment to return.


Dublin is arguably the only part of the CEAS that deserves the attribute “common” or “European”. While other areas of the system remained broadly national, based, however, on solid minimum and common standards, the EU responsibility determination process indeed became a “common” and “new” system, which has been introduced in each EU MS.

Against this background and the fact that significant financial commitments have been taken to set up this process, it is understandable that the EC holds on to this system, as imperfect as it may be. In some respect, however, Dublin contradicts the principles of solidarity, impedes a fair or purposeful sharing of responsibilities among the 32 Dublin countries and seems to be the stumbling block for reaching a consensus in June 2018. Determining responsibilities purely based on arbitrary reasons such as the geographical location of a State seem unsuitable to represent the cornerstone of a European System.

Dublin (in its current form) leaves little room for the countries to negotiate: countries at the EU external border such as Greece or Italy hardly have another choice than opposing a system that makes them responsible for (basically) all asylum claims submitted within the EU. In the absence of an EU wide solidarity mechanism, attractive destination countries such as Germany, Sweden or Austria which have been confronted with secondary movements will turn to solutions assigning the responsibility on the state of first entry. Less attractive destination countries, will opt for a status quo that guarantees them rather modest burdens to stem potentially even taking financial penalties into account.

20 years of Dublin could be finally enough. The EP proposal may offer a good way out of the dilemma: keeping the good parts of the system (only one country shall further be responsible) and skip the problematic ones (like the first country of asylum or the safe third country principle). Only a fundamental adjustment of the Dublin system can open the doors for further progress towards a common European asylum system. Thus, ironically, the key for building a true and comprehensive European Asylum architecture may only succeed if the – at present – corner stone of the European Asylum system, the Dublin system, is sacrificed.


ICMPD is partner of the CEAS-EVAL project under the EC’s Horizon2020 programme. Within this project ICMPD investigates on some of the key elements of the Common European Asylum System, namely “solidarity”, “responsibility sharing” and “harmonisation”. Those terms are researched as regards to how they emerged at EU level, how they have been discussed in the past and how (or whether) their acceptance and approach has shifted over the years. The research is accompanied by a series of blog posts on various aspects connected to these terms. Ultimately the research will lead to various deliverables such as briefing, thematic or policy papers, info graphs as well as more detailed reports. The present blog post is the first in this series and reflects on the stuck negotiations on the “Dublin reform”. The views expressed in the blog are those of the author, and neither necessarily those of the CEAS EVAL project nor those of ICMPD.