Report | Doc. 13866 | 10 September 2015

After Dublin – the urgent need for a real European asylum system

Committee on Migration, Refugees and Displaced Persons

Rapporteur : Mr Michele NICOLETTI, Italy, SOC

Origin - Reference to Committee: Doc. 13592, Reference 4083 of 3 October 2014. 2015 - Fourth part-session

Summary

The Dublin regulation is a European Union legal instrument establishing a system for definitive identification of the participating State responsible for examining a particular asylum application. Since the system was introduced in 1990, however, the scale, nature and geographical focus of mass migration into the European Union have changed significantly. In addition, the distribution of asylum applicants between participating States is extremely uneven; the Dublin system is, however, not intended or capable of functioning as a “burden-sharing” mechanism to counteract this inequity, which on the contrary is exacerbated by Dublin transfers of asylum applicants. It has thus become a symbol of unfairness and lack of solidarity in European asylum policy. Furthermore, the Dublin system has given rise to serious violations of asylum seekers’ human rights.

The Dublin system is thus dysfunctional: ineffective and certainly inefficient in achieving its basic aims, and at an unacceptably high human cost and resource cost. Indeed, it is difficult to see how it could operate as intended. The Parliamentary Assembly should therefore propose a series of reforms to the implementation of the current Dublin system, the wider context in which it operates and on which it is dependent, and to the content of the Dublin regulation itself.

A. Draft resolution

(open)

1. As Parties to the 1951 Convention relating to the Status of Refugees, all Council of Europe member States have an obligation to protect refugees. In order to determine who should benefit from this protection and thereby avoid breaches of the obligation, it is necessary to examine individual asylum applications.

2. The Dublin Regulation is a European Union legal instrument establishing a system for definitive identification of the participating State responsible for examining a particular application. Alongside other instruments and mechanisms, it forms part of the European Union’s wider Common European Asylum System (CEAS). Its essential purpose is to address the problems of “asylum shopping” and “refugees in orbit”. Addressing these problems remains crucial for the CEAS to function as it should.

3. Since the original Dublin Convention was adopted in 1990, however, the scale, nature and geographical focus of mass migration into the European Union have changed significantly, with especially dramatic developments in recent years. In addition, the distribution of asylum applicants between participating States is extremely uneven, in many cases simply because of their geographical location – for example, in 2014, five participating States dealt with 72% of all applications.

4. The Dublin system is not intended or capable of functioning as a “burden-sharing” mechanism to counteract this inequity. On the contrary, the inequity is exacerbated by Dublin transfers of asylum applicants based on the criterion of “irregular border crossing” – the most commonly evoked, by which asylum seekers are transferred to the country of first entry to European Union territory. The Dublin system has thus become a symbol of unfairness and lack of solidarity in European asylum policy, in particular the CEAS, which lacks an effective compensatory mechanism for redistributing the burden.

5. Furthermore, the Dublin system has given rise to violations of asylum seekers’ human rights, including the right to family life, the right not to be subjected to inhuman or degrading treatment, the right not to be subjected to refoulement, the right not to be detained arbitrarily and the right to an effective remedy. Some of these violations have been the subject of court judgments, including of the European Court of Human Rights. The Dublin system may also subject asylum applicants to a lengthy period of uncertainty and precarity before the State responsible for examining their application is identified; this may be further prolonged where the person concerned is transferred to a country with lengthy delays in its asylum procedures.

6. Many of these problems are caused by deficiencies in implementation of other elements of the CEAS, in particular the Eurodac Regulation (on fingerprinting of asylum applicants), the reception conditions directive and the asylum procedures directive, which may obstruct proper implementation of the Dublin system. Indeed, the fact that transfers of asylum applicants to a particular country constitute a burden is a perverse incentive not fully to comply with these other instruments. In the case of the reception conditions and asylum procedures directives, this comes at great cost to the protection afforded to asylum applicants.

7. The Parliamentary Assembly concludes that the Dublin system is dysfunctional. It is not effective and certainly not efficient in achieving its basic aims, and its operation in practice has an unacceptably high human cost for asylum applicants and resource cost to participating States in complying with its lengthy and complicated procedures, whilst only aggravating the unfair repartition of responsibility. Furthermore, it is difficult to see how the Dublin system can operate as intended, as it depends on the still unfulfilled assumption that all participating States are able to ensure protection of asylum applicants and properly cope with the number of applications they receive. Indeed, recent events in Germany, Austria, Hungary and elsewhere show that the Dublin system has already collapsed and must be reformed as a matter of urgency.

8. The Dublin system, including both the Regulation itself and its implementation in practice, is thus in urgent need of reform – as, indeed, is the wider CEAS. Without far-reaching reform, there is a risk of participating States suspending or withdrawing from the Dublin system, which would cause chaos and confusion. The Assembly welcomes the fact that the European Commission will conduct an evaluation of the Dublin system in 2016 and takes note of the European Council conclusions of 26 June 2015. Nevertheless, many necessary measures can and should be taken now, and the need for other, more far-reaching ones is already apparent. Whilst reform of the Dublin system cannot provide answers to all of the questions raised by today’s mixed migratory flows to and around Europe, failure to reform it will undermine all other efforts to address the issue and place the entire CEAS in jeopardy. European political solidarity is at stake.

9. The Assembly therefore recommends, as regards implementation of the Dublin system and related instruments:

9.1. strict application of the family-related responsibility criteria and provisions concerning the best interests of the child, notably those relating to unaccompanied minors;

9.2. application of the dependent persons and humanitarian clauses in a fair, humane and flexible manner, taking greater account of asylum seekers’ preferences;

9.3. rigorous application of the discretionary clauses where a transfer would be incompatible with obligations under international law;

9.4. avoidance of reflexive recourse to the criterion of “irregular border crossing” in order to return asylum applicants to the country of first entry;

9.5. aggressively proactive use of the mechanism for early warning, preparedness and crisis management in anticipation of critical pressure on or problems in the functioning of a participating State’s asylum system;

9.6. prompt and full implementation of relevant judgments of the European Court of Human Rights and the Court of Justice of the European Union;

9.7. full and effective implementation of the reception conditions and asylum procedures directives by relevant States;

9.8. enhancing the resources available to and making greater use of financial mechanisms such as the Asylum, Migration and Integration Fund;

9.9. greater sharing of expertise and provision of technical assistance, including through the European Asylum Support Office.

10. As regards the wider context within which the Dublin system operates and on which it is dependent, the Assembly recommends:

10.1. greater harmonisation and stronger supervision of national refugee-status determination procedures, leading to mutual recognition of positive national status-determination decisions; or otherwise “joint” processing of asylum applications, which would have the advantage of contributing to burden sharing;

10.2. ensuring, through appropriate, automatic procedures, relocation of recognised refugees on a level adequate to ensure equitable burden sharing amongst the participating States;

10.3. introducing a status of “European refugee” for beneficiaries of international protection, allowing transfer of residence and exercise of other rights between States; or otherwise, shortening to two years the qualification period for beneficiaries of international protection to qualify under the long-term residents directive.

11. As regards revision of the Dublin Regulation, the Assembly recommends:

11.1. prompt adoption of the European Commission’s legislative proposal intended to resolve the ambiguity in the provision regulating responsibility for asylum applications lodged by unaccompanied minors;

11.2. considering removal of the criterion of “irregular border crossing” as a basis for determining which State is responsible for an asylum application;

11.3. taking account of the foregoing recommendations, undertaking an immediate evaluation of the Dublin system, which takes a holistic view of the overall effects of the Dublin system and of the wider context within which it operates.

B. Explanatory memorandum by Mr Nicoletti, rapporteur

(open)

1. Introduction

1. The Dublin Regulation is a European Union legal instrument intended to establish a technical mechanism for definitive identification of the member State responsible for examining a particular asylum application. Alongside other instruments and mechanisms, it forms part of the European Union’s wider Common European Asylum System (CEAS). In recent years, the “Dublin system” has been heavily criticised on various grounds, notably for causing violations of asylum seekers’ human rights; producing inequitable distribution of asylum applications across EU member States, since it provides for transfer of responsibility to countries of irregular first entry, which tend to be on the European Union’s external borders, especially to the south; and being slow, costly and ineffectual.

2. In its recently announced “European Agenda on Migration”, the European Commission acknowledged that “the ‘Dublin system’ is not working as it should. In 2014, five member States dealt with 72% of all asylum applications EU-wide.” This disparity is indeed a matter of concern, although it is not necessarily the fault of the Dublin system, which was never intended to be a burden-sharing mechanism. The Commission also noted that “[w]hen the Dublin system was designed, Europe was at a different stage of co-operation in the field of asylum. The inflows it was facing were of a different nature and scale”.

3. This latter point suggests an important consideration. The scale, nature and geographical focus of mass migration into the European Union have indeed changed significantly since the original Dublin Convention was adopted in 1990, with especially dramatic developments in recent years. As can be seen from the graph below,

there has been an almost constant growth in the number of asylum applications made within EU member States, with record figures reached in 2013 and 2014 that are fully expected to be significantly exceeded this year. Indeed, Europe is today faced with migration and in particular asylum challenges of a new and different nature. Several of the main countries of origin, located close to Europe’s external borders, are currently experiencing intense and protracted internal armed conflicts involving extreme violence and persecution towards civilians on ethnic, religious or political grounds. Those fleeing such situations are extremely likely to qualify as refugees under the 1951 Convention relating to the Status of Refugees, as shown by the very high recognition rates for the nationalities concerned. Furthermore, the passage to Europe is now being driven and facilitated by novel factors: these include the current situation of near-anarchy in Libya, previously a country of destination but now a point of embarkation for the maritime crossing to Italy and Malta; the near-saturation of Turkey, neighbouring Greece; and the emergence of sophisticated, dynamic migrant-smuggling networks, often employing ruthless (but effective) methods.

Source: “Asylum in the EU: Facts and Figures”, European Parliament Research Service, March 2015

4. On the one hand, reform of the Dublin system will not reduce the number of asylum applications made in the European Union and will certainly not provide answers to all of the questions raised, for example, by the flow of irregular migrants and asylum seekers, notably from Syria, Afghanistan, Eritrea, Pakistan, Iraq and West Africa. These challenges call for far more extensive technical and political solutions than could be found through revision of the Dublin Regulation. On the other hand, the radically different situation of today does give rise to legitimate questions as to whether or not the Dublin system, at least in its current form, is a useful or even appropriate element of the CEAS.

2. The history of the Dublin Regulation in the context of the European Union’s common European asylum system

5. The starting point for the European Union’s work on asylum policy may be taken as the 1985 Schengen Agreement between Belgium, France, Germany, Luxembourg and the Netherlands and, even if this was outside the then-European Communities treaty framework. Its main aim was the gradual abolition of checks at common “internal” borders between the participating States. Supplemented in 1990 by the Schengen Implementation Agreement, the Schengen acquis was integrated into the European Union treaty framework by the 1999 Treaty of Amsterdam (see below). The abolition of internal borders means that once an asylum seeker enters the Schengen space, s/he can circulate freely within it. Today, all EU member States (except Bulgaria, Cyprus, Ireland, Romania and the United Kingdom), as well as non-member States Iceland, Liechtenstein, Norway and Switzerland, are part of the Schengen system.

6. The 1990 Dublin Convention, which entered into force in 1997, was also adopted outside the European Communities treaty framework by a larger group (initially 12) of member States. It was intended to determine the State responsible for examining asylum applications and in doing so, to respond to the phenomena of “asylum shopping” (multiple applications in different countries) and “refugees in orbit” (endless transfers of asylum seekers with no country accepting responsibility). Under the Dublin Convention, responsibility was allocated according to a hierarchy of criteria, including the protection of unaccompanied minors, reunification with family members in a particular country, possession of a visa or residence permit for a particular country, illegal entry to or stay in a particular country, and country of first application. In addition, a so-called “humanitarian clause” allowed for voluntary acceptance of responsibility. The Dublin Convention depended on the existence of compatible standards and was based on mutual trust in implementation of those standards, although at the time there was no codification of these standards in the acquis communautaire. It contained no provision for mutual recognition of decisions on recognition of status.

7. The European Union progressively developed new competences in the area of asylum, notably as a result of the Treaty of Amsterdam, the European Council’s Tampere Programme in 1999 and the 2001 Treaty of Nice. Under the 2004 Hague Programme, the first phase of the resulting CEAS was completed in 2006, with further evaluation and reform of existing acquis and adoption of second-phase measures by the end of 2010.

8. The 2003 “Dublin II” Regulation brought the Dublin Convention within the framework of EU treaty law. The drafting process acknowledged and sought to address certain deficiencies, including slow operation of the system, uncertainty for applicants and member States, insufficient remedies for the “refugees in orbit” phenomenon, the risk of “chain refoulement”, lack of proper readmission rules and supervision, and the disproportionate burden imposed on member States with external EU borders.

9. Following the 2007 Treaty of Lisbon, the aims of the CEAS were further promoted by the 2008 European Pact on Migration and Asylum and the European Council’s 2009 Stockholm Programme (for 2010-2014), which aimed at establishing a “Europe of responsibility, solidarity and partnership in migration and asylum matters”; it also reiterated the importance and expanded the scope of the CEAS, incorporating issues such as access to the European Union, the resettlement and integration of refugees, external processing of asylum claims, regional protection programmes and mechanisms for sharing responsibility between EU member States.

10. As part of this process, a “recast” Dublin III Regulation was adopted in 2013. This includes the following improvements with respect to Dublin II:

an early warning, preparedness and crisis management mechanism, geared to addressing the root dysfunctional causes of national asylum systems or problems stemming from particular pressures;

a series of provisions on protection of applicants, such as a compulsory personal interview, at which the applicant must be informed that they may provide information about family members in other member States, guarantees for minors (including detailed guidance on assessing a child's best interests) and extended possibilities for reuniting them with relatives (now taken to include also grandparents, uncles or aunts), and provision to applicants of a standard information leaflet and a specific leaflet for unaccompanied minors;

a guaranteed right of appeal, with suspensive effect, against a transfer decision;

an obligation to ensure legal assistance free of charge upon request;

a single ground for detention in case of significant risk of absconding, and strict limitation of the duration of detention;

more legal clarity of procedures, for example clearer, exhaustive deadlines for different stages of the procedure.

11. Alongside the Dublin Regulation, the other main components of the CEAS include the following:

The “Eurodac” fingerprint registration system, intended to assist in determining which member State is to be responsible under the Dublin Regulation for examining an asylum application. Member States are required promptly to take the fingerprints of all asylum applicants aged 14 or over and within 72 hours to transmit them, along with information on the application, to the Eurodac central system;

The “reception conditions” directive, intended to establish common standards for the reception of asylum applicants in member States in areas such as access to housing, food, health care and employment, as well as medical and psychological care;

The “asylum procedures” directive, intended to establish common procedures for granting and withdrawing international protection so that all member States examine applications to a common, high quality standard;

The “qualification” directive, intended to establish common standards for qualification for international protection, for a uniform status for refugees or for persons eligible for subsidiary protection, and for the content of the protection granted. Such persons will benefit from a series of rights on protection from refoulement (residence permits, travel documents, access to employment, access to education, social welfare, health care, access to accommodation and access to integration facilities), as well as specific provisions for children and vulnerable persons;

The European Asylum Support Office (EASO), intended to help to improve the implementation of the CEAS, to strengthen practical co-operation among Member States and to provide and/or co-ordinate the provision of operational support to Member States subject to particular pressure on their asylum and reception systems.

12. The Dublin III Regulation is binding on all EU member States, as well as Iceland, Switzerland, Norway and Liechtenstein, which as noted above are also part of the Schengen system. As regards other elements of the CEAS, it should be noted that Denmark, Ireland and the United Kingdom have opted out of the recast asylum procedures, reception conditions and qualification directives.

13. In response to the further deterioration in the situation this year, the European Commission issued a “European Agenda on Migration” in May. This included proposals for reform of the Dublin system, amongst which “a temporary distribution scheme for persons in clear need of international protection to ensure a fair and balanced participation of all Member States to this common effort”, involving a “distribution key … based on objective, quantifiable and verifiable criteria that reflect the capacity of the Member States to absorb and integrate refugees, with appropriate weighting factors reflecting the relative importance of such criteria”.

Although the member States subsequently rejected the idea of “quotas”, the European Council did agree in June to “the temporary and exceptional relocation over two years from the frontline Member States Italy and Greece to other Member States of 40 000 persons in clear need of international protection, in which all Member States [except the UK] will participate”.

Whilst the introduction of a relocation system is to be welcomed as a potential step towards reform of the Dublin system, its temporary and exceptional nature and the relatively small number of people involved raise serious doubts as to its sufficiency in any more than the most immediate term. It would be extremely disappointing were this to prove no more than an ad hoc political palliative. Indeed, it is already a matter of great concern that the European Union was subsequently unable to meet even the 40 000 target, with agreement to relocate only 32 000: despite the Council conclusions referring to participation by “all Member States”, Austria and Hungary refused to accept any refugees, and there were great disparities between other States, with several accepting far fewer than expected.

This outcome does not augur well for the future: as Commissioner Avramopoulos subsequently commented, “This shows that a voluntary scheme is difficult to implement and whenever it was tried before, it has failed”.

3. Criticisms of the Dublin system

3.1. Previous Assembly criticisms

14. The Parliamentary Assembly has, in several resolutions, been critical of the way in which the Dublin system operates and even of its very rationale. Most recently:

in Resolution 1918 (2013) “Migration and asylum: mounting tensions in the eastern Mediterranean”, the Assembly called on the European Union to “revise and implement the ‘Dublin’ Regulation in a way that provides a fairer response to the challenges that the European Union is facing in terms of mixed migration flows”;

in Resolution 2000 (2014) on the large-scale arrival of mixed migratory flows on Italian shores, the Assembly, recalling its Resolution 1820 (2011), called on the European Union to “modify the Dublin system, … both to ensure fair treatment and appropriate guarantees for asylum seekers and beneficiaries of international protection and also to assist individual member States to face possible situations of exceptional pressure”.

15. A more detailed analysis of the situation can be found in the Assembly report on “Asylum seekers and refugees: sharing responsibilities in Europe”. This report notes that: “While the intention was to achieve a fairer division of responsibilities among European countries, the Dublin system has placed a disproportionate burden on countries such as Greece, Malta, Italy, Spain and Cyprus at the external borders of the European Union… It is clear that a simple mechanism putting the responsibility on the first State of arrival is not a complete solution. It was based on the assumption that all EU countries were safe and able to cope. They were not. There are also wide variations in a person’s chances of being granted asylum in particular countries.”

To a large extent, these criticisms remain valid today.

3.2. Judicial criticism

16. In the case of M.S.S. v. Belgium and Greece (judgment of 21 January 2011) concerning Dublin transfers from Belgium to Greece, the Court found, inter alia, that given the availability of numerous credible reports, the Belgian authorities “must” have been aware of the deficiencies in the asylum procedure in Greece. They had nevertheless proceeded without the Greek authorities having given any individual guarantee, when they could easily have refused the transfer. The Belgian authorities should not simply have assumed that the applicant would be treated in accordance with standards of the European Convention on Human Rights (ETS No. 5, “the Convention”) but rather have verified how the Greek authorities actually applied their asylum legislation. There had therefore been a violation by Belgium of Article 3 (prohibition of degrading treatment) of the Convention. There was also a violation of Article 13 (right to an effective remedy) because of the lack of an effective remedy against the applicant’s expulsion order. As regards Greece, the Court found a violation of Article 13 because of the deficiencies in the Greek authorities’ examination of the applicant’s asylum application and the risk he faced of being removed directly or indirectly back to his country of origin without any serious examination of the merits of his application and without having had access to an effective remedy. Greece was found also to have violated Article 3 because of both the applicant’s detention conditions and his living conditions in Greece.

17. Tarakhel v. Switzerland (judgment of 4 November 2014) concerned the Swiss authorities’ refusal to examine the asylum application of an Afghan couple and their six children and the decision under the Dublin Regulation to send them back to Italy. The Court found that the Swiss authorities would violate Article 3 were they to transfer the applicants back to Italy without having first obtained individual guarantees from the Italian authorities that they would protect the applicants, as appropriate to the age of the children, and keep the family together. Given the current situation regarding the reception system in Italy, and in the absence of detailed and reliable information concerning the specific facility of destination, the Swiss authorities did not possess sufficient assurances to these ends.

18. In A.M.E. v. the Netherlands (judgment of 13 January 2015), however, the Court declared the applicant’s complaint under Article 3 inadmissible, finding that he had not established that if returned to Italy, he faced a sufficiently real and imminent risk of material, physical or psychological hardship severe enough to fall within the scope of Article 3. Distinguishing it from the case of Tarakhel v. Switzerland (see above), which involved a family with six minor children, in this case the applicant was an able young man with no dependants. The current situation in Italy for asylum seekers could in no way be compared to the situation in Greece at the time of the M.S.S. judgment (see above): the structure and overall situation of the reception arrangements could not in themselves act as a bar to all removals of asylum seekers to Italy.

19. Similarly, in A.S. v. Switzerland (judgment of 30 June 2015), the Court found that the applicant, who had been diagnosed with severe post-traumatic stress disorder, would not be at risk of inhuman or degrading treatment if removed to Italy, as he was not at present critically ill and there was no indication that he would not receive appropriate psychological treatment in Italy or that he would not have access to an anti-depressant such as he was receiving in Switzerland. The Court also distinguished the case of D. v. the United Kingdom, in which the applicant, in the final stages of AIDS, had no prospect of medical care or family support in his country of origin. Whilst in Tarakhel the Court had raised serious doubts as to the capacities of the Italian reception system, the overall situation could not in itself justify barring all removals of asylum seekers to Italy. (The Court also found that the applicant had no claim under Article 8 (family life) arising from the presence of his two sisters in Switzerland.)

20. The Court of Justice of the European Union (CJEU) has also issued important judgments concerning certain aspects of the Dublin system. In N.S. v Secretary of State for the Home Department (judgment of 21 December 2011), the CJEU held that EU member States may not transfer an asylum seeker to the member State responsible under the Dublin Regulation where they cannot be unaware that “systemic deficiencies” in the asylum procedure and in the reception conditions of asylum seekers in that member State amount to substantial grounds for believing that the asylum seeker would face a real risk of being subjected to inhuman or degrading treatment within the meaning of Article 4 of the Charter of Fundamental Rights of the European Union. The CJEU also clarified that the EU asylum system cannot operate on the basis of a “conclusive presumption” that all EU member States observe the fundamental rights of the European Union.

21. There is some uncertainty as to whether the CJEU’s test in N.S. differs from that applied by the European Court of Human Rights in M.S.S., and if so, how the two may be reconciled. Further examination of this highly technical legal issue is beyond the scope of the current report: the essential point is that both courts have clarified that an EU member State intending to transfer an asylum applicant to another member State cannot simply rely on an untested assumption that the latter will apply the reception conditions and asylum procedures directives to the standard required to ensure effective protection of an asylum applicant’s fundamental rights.

22. In MA, BT, DA v. Secretary of State for the Home Department (judgment of 6 June 2013) concerned an ambiguity in the Dublin II Regulation, which provided that where a minor had no family member legally present in a member State, the member State responsible is that in which the minor has applied for asylum, but did not specify whether that is the first application which the minor lodged in a member State or the most recent application lodged in another member State. The CJEU ruled that, in such circumstances, the member State responsible will be that in which the minor is present after having lodged an application there. The European Commission has since proposed a Regulation amending the relevant provision of the Dublin III Regulation aimed at addressing the ambiguity identified in the CJEU judgment.

4. The current situation – persistent criticisms of the Dublin system

23. Despite the suggestion that the Dublin system places a disproportionate burden on southern European States, by far the largest number of new asylum applications in 2014 were submitted in Germany (172 945), followed by Sweden (74 980), Italy (63 655), France (57 230), Hungary (41 215), the United Kingdom (31 070) and the Netherlands (23 645). As regards relative burdens, the number of asylum applications per head of population is highest in Sweden (7 775 per million), followed by Hungary (4 175), Malta (3 000), Denmark (2 585) and Germany (2 140); the figures for non-EU member States Switzerland and Norway were 2 695 and 2 485 respectively.

24. Figures for 2013 show that Italy received by far the largest number of requests under the Dublin Regulation to take responsibility for an asylum application (15 532), followed by Poland (10,599), Hungary (7,756), Belgium (5,441) and Germany (4,532).

These figures seem to suggest that for certain countries, the Dublin system as it currently operates has a significant impact on their share of the overall burden of responsibility. It should, however, be noted that these figures relate to requests for transfers, not all of which are realised. The EASO has found that during the period 2008-2012, although around 80% of “outgoing” requests for Dublin transfers were accepted by the “incoming” State, only around 25% resulted in the physical transfer of the asylum applicant. Set against the total number of asylum applications made in the European Union, “outgoing” requests were made in about 12%, and transfers realised in about 3% of cases.

25. It has thus been argued that claims that the Dublin system “has prompted a transfer of asylum-processing responsibilities from Europe’s north to its southern borders … [are] not borne out by the evidence. While northern European States clearly send more transfer requests than do southern ones, and those in the south are most likely to be on the receiving end of requests, the disparity in numbers of actual transfers is relatively small”.

(Looked at the other way around, one could say that the absence, or minimal extent, of such transfers is evidence that the Dublin system is not working: given their geographical location and recent trends in irregular migration, one would expect that member States on the European Union’s southern external borders would be responsible for even higher numbers of asylum applications.) Similarly, a study prepared for the European Parliament concluded that “much effort and expenditure goes into maintaining a database of fingerprints which reveals increases in secondary movement of asylum seekers but not on a particularly dramatic scale; and a fairly desultory number of actual transfers of asylum seekers back from one member State to another. In relation to the overall numbers of asylum claims, this activity is quite minor. Further, the top sending and receiving countries under the system are, in a number of prominent cases, the same. So … the end result to the overall number of asylum seekers for which the State is responsible does not change much. For individual asylum seekers, by contrast, the human cost may be enormous”.

26. The Dublin system was never intended as a burden-sharing mechanism of solidarity in itself: its aim is to ensure definitive identification, based on common objective criteria, of the member State responsible for processing an asylum application. Operation of the Dublin system in practice involves neither consideration of equity in the resulting overall distribution of responsibility nor any mechanism specifically intended to address the resulting burdens or inadequacies in national capacity. The wider CEAS, however, does make provision for certain responses to these issues: since 2009, Article 80 of the Lisbon Treaty has provided that the CEAS is governed by “the principle of solidarity and fair sharing of responsibility”. This is expressed notably through the EASO and the Asylum, Migration and Integration Fund (worth €3 137 billion over the period 2014-20; in 2015, for example, €25 million was made available for emergency assistance to States).

27. In addition, being part of the Common European Asylum System, the effectiveness and consequences of operation of the Dublin system in practice depend also on the content and implementation of other aspects, notably the asylum procedures and reception conditions directives. The European Parliament study noted that “[t]he Dublin system is built on an implicit assumption that asylum seekers will be able to enjoy access to similar standards of treatment and rights in all participating States, but this goal, which is also the objective of the CEAS as a whole, is yet to be achieved in practice. The lack of trust that asylum seekers have for the system – and for the likelihood that it will ensure them access to similar standards of treatment and rights in all participating States – means that secondary movements persist, contrary to Dublin's implicit aim of preventing what is characterised negatively and simplistically as 'asylum shopping'. In many cases, Member States are unwilling or unable to comply with its provisions.”

Implementation of the Dublin system is also undermined by failure promptly to register as asylum applicants and fingerprint newly-arrived irregular migrants, as required by the Eurodac Regulation. Such persons are then free to move to and apply for asylum in another EU Member States without leaving any evidence of where they had irregularly crossed the European Union’s external border, thus defeating application of the relevant criterion for identifying the Member State responsible for examining the application.

28. Despite the high level of integration of the EU system, national implementation of the Dublin system is still ultimately dependent on domestic political will, and its reform even more so. Moral dilemma and domestic electoral considerations inevitably play a part. For example, certain judgments of the European Court of Human Rights and the Court of Justice of the European Union mean that until the necessary measures are taken by the domestic authorities concerned, Greece is in effect excluded from receiving incoming transfers of responsibility under the Dublin system and in certain circumstances, transfers to Italy require special arrangements. The dilemma lies in the fact that, on the one hand, there is a legal obligation to implement these judgments, whereas on the other, delay in implementation defers resumption of the responsibility to accept Dublin transfers, along with the resulting administrative and financial burden. This has already been noted in an Assembly report, which stated that “the Dublin Regulation de facto discourages the southern member States from improving their standards on reception and procedures for asylum seekers, and thus threatens the aim of a Common European Asylum System”;

the same could be said for improving implementation of the Eurodac Regulation (see paragraph 26 above). As to the significance of domestic political/electoral considerations, the growth in popular support for political parties that are critical of or hostile to immigration in general may deter politicians from advocating or responding positively to proposals for more equitable burden sharing.

5. Proposals for improved implementation or revision of the Dublin system

29. The Dublin system already contains provisions whose more extensive application could help mitigate many of the negative consequences at present resulting from its operation in practice. For example, EU member States could take the following steps, requiring only changes in practice without any need for revision of the existing texts:

ensure strict application of the family-related responsibility criteria through careful adherence to the established hierarchy, and ensuring full respect for the best interests of the child and the principle of family unity, in accordance with obligations under wider international law;

apply the dependent persons, sovereignty and humanitarian clauses in a fair, humane and flexible manner, taking greater account of asylum seekers' preferences;

avoid automatic recourse to the criterion of “irregular border crossing”;

apply the sovereignty clause where a transfer would be incompatible with obligations under international law, including those arising under the European Convention on Human Rights as interpreted in the case law of the European Court of Human Rights, as required in the preamble to the Regulation itself;

ensure that the best interests of the child are given full effect as the primary consideration in practice, including by making greater efforts to trace family members of unaccompanied minors and giving the benefit of any doubt in age-dispute cases;

fully and effectively implement the Eurodac Regulation, reception conditions and asylum procedures directives.

30. A more radical proposal, requiring revision of the Dublin Regulation, would be to abandon the criterion of “irregular border crossing” as a basis for determining which State is responsible for an asylum application. As a result, the State responsible would be that in which the application was lodged, in effect allowing asylum applicants freely to choose the member State in which their application would be examined.

This would not ensure equitable distribution of responsibility; there would be no guarantee that individual freedom of choice would lead to this result.

That said, the figures presented in paragraphs 23 and 24 above show that the current operation of the Dublin system does not necessarily result in an extreme concentration of asylum applications at the European Union’s southern external borders; and furthermore, a relatively small number of Dublin transfers are realised, especially when set against the total number of applications. This suggests that in practice, asylum seekers are already to a large extent making their applications in the countries of their choice. It is most likely that these are the countries in which they have family, friends and relations or cultural or linguistic reasons for preferring to be. This leads to two observations: first, States should more often be accepting responsibility under the family, humanitarian and sovereignty clauses; second, allowing freedom of choice might in practice make little difference to the distribution of asylum applicants, but it would relieve States of a costly, cumbersome and somewhat ineffectual administrative burden, whilst also avoiding much of the human cost to asylum seekers. Indeed, the procedural improvements introduced in the Dublin III Regulation, in particular the right to information and the personal interview, provide a clear basis for progress in this area, to which end they should be fully exploited.

31. Looking beyond the Dublin Regulation, many of its deleterious effects in terms of inequitable burden sharing could be mitigated by establishing provisions to ensure mutual recognition of positive national status-determination decisions and the possibility of transfer of international protection status between EU member States; in other words, creating a “European refugee” status. Indeed, this is already foreseen in the Lisbon Treaty, Article 78 of which calls for “a common European asylum system comprising a uniform status of asylum … valid throughout the Union [and] a uniform status of subsidiary protection”; mutual recognition could be seen as a concrete expression of the solidarity principle set out in Article 80 of the Treaty. Mutual recognition alone would not change the way in which the Dublin Regulation identifies the member State responsible for determining an asylum application, but it would allow the longer-term responsibility for meeting an individual's protection needs to be transferred between member States.

32. Improvement of implementation of other elements of the CEAS could also mitigate the current ill effects of operation of the Dublin system. As noted above, one incentive for asylum seekers to prefer making their application in one member State rather than another is the differing levels of implementation of, in particular, the reception conditions and asylum procedures directives. Ensuring full and effective implementation of these instruments would reduce this incentive and thereby also the resulting secondary movements. This in turn could reduce the pressure on other member States to take a more restrictive overall approach to accepting instead of seeking to transfer responsibility. The CEAS as a whole cannot operate as intended unless it is coherent and realistic, and all of its constituent parts function properly.

33. Indeed, one could well say that the CEAS as a whole is inadequate to the scale and nature of the challenges with which it is currently confronted. One way in which it could be further developed would be through greater harmonisation, or even centralisation of procedures for refugee status determination, going beyond improved national implementation of the existing asylum procedures directive. Possible measures to this end include a mechanism for more effective European supervision of national procedures, a status of “European asylum seeker” or joint, European-level processing of asylum applications. Such measures would enhance mutual trust and thereby provide a reliable foundation for introduction of a “European refugee” status.

34. Some of the practical benefits of a “European refugee” status are already partly realised through the long-term residents directive,

which inter alia allows exercise of the right of residence in another member State and whose application was in 2011 extended to beneficiaries of international protection.

The burden-sharing advantages of this mechanism could be further enhanced by shortening the qualification period for refugees from five to two years, in recognition of their particular situation.

35. On a more immediate, practical level, various possibilities already exist also for giving practical effect to the principle of solidarity and alleviating the inequities generated by the Dublin system. These include financial mechanisms such as the Asylum, Migration and Integration Fund; sharing of expertise, whether through the EASO or through development of some form of joint processing; greater targeted EASO assistance, for instance in the processing of “take charge requests”; and redistribution of beneficiaries of international protection, including through resettlement from one EU member State to another, on the basis of some form of distribution key, and enhancing their intra-EU freedom of movement.

Again, the solidarity principle set out in Article 80 of the Lisbon Treaty, which refers to the “financial implications” of fair sharing of responsibility, provides a solid legal basis for such actions. As regards joint processing, it must be recognised that there may be practical difficulties arising from a lack of mutual confidence in the quality of national asylum procedures and the absence of common legal frameworks, which depend on national legislation. The difficulties in ensuring common standards in practice could, however, be alleviated by pairing national officials from one country with those from another.

36. One could also imagine a mechanism allowing better management and distribution of asylum applications through balanced resettlement of certain categories of refugee – in particular those of nationalities generally recognised as being in need of international protection – following status determination procedures conducted in countries of transit. Such persons could apply for protection via the Office of the United Nations High Commissioner for Refugees (UNHCR) or European embassies. Clearly, there are various legal, practical, and above all human rights issues that would have to be resolved before such a policy could be implemented. This approach could, however, avoid some of the need for recourse to migrant smugglers and the dangerous routes and methods they employ to reach Europe. (Such a mechanism would be seen as the “other side of the coin” to “white lists” of countries whose nationals are presumed not to be in need of international protection.)

37. As regards the wider context and alleviation of the overall pressure of asylum seekers on Europe, another proposal would be to improve and strengthen protection possibilities outside Europe, notably in countries of transit. For this to work, however, it would be necessary significantly to reduce the duration of status determination procedures, as refugees may be unwilling to wait for long periods, especially when their goal had been to reach Europe.

38. Although the European Commission has yet to undertake its evaluation of the Dublin III Regulation, the “European Agenda on Migration” already includes, inter alia, the following points:

member States should allocate the resources needed in order to increase the number of transfers and cut delays, proactively and consistently apply the clauses related to family reunification, and make a broader and regular use of the discretionary clauses, allowing them to examine an asylum application and relieve the pressure on the frontline member States;

the European Asylum Support Office (EASO) will support member States by establishing a dedicated network of national Dublin Units.

39. The rapporteur cannot but agree with the above points, although he observes that problems of delay, inefficiency and ineffectiveness have been present since the beginning and were supposed to have been addressed in the Dublin II Regulation. He would also underline that the human cost to asylum applicants and the system’s ability to respond to emergency situations are of at least equal concern. The Rapporteur therefore looks forward to the Commission’s 2016 evaluation to see whether the improvements made in the Dublin III Regulation (see paragraph 10 above) have had the desired results.

40. Evaluation of the Dublin system should not be limited to a narrow consideration of statistics on transfers. The dramatically different and unforeseen situation in which it now operates should form the backdrop for a careful reflection on its very rationale. Bearing in mind also the system’s side effects and general ineffectiveness, has the simple, original purpose – to ensure that a single, easily identifiable country is responsible for examining a particular asylum application – been defeated by the creation of an overcomplicated system for its realisation?