This week the EU Commission published its new package of proposals on asylum and (non-EU) migration – consisting of proposals for legislation, some ‘soft law’, attempts to relaunch talks on stalled proposals and plans for future measures. The following is an explanation of the new proposals (not attempting to cover every detail) with some first thoughts. Overall, while it is possible that the new package will lead to agreement on revised asylum laws, this will come at the cost of risking reduced human rights standards.
Since 1999, the EU has aimed to create a ‘Common European Asylum System’. A first phase of legislation was passed between 2003 and 2005, followed by a second phase between 2010 and 2013. Currently the legislation consists of: a) the Qualification Directive, which defines when people are entitled to refugee status (based on the UN Refugee Convention) or subsidiary protection status, and what rights they have; b) the Dublin III Regulation, which allocates responsibility for an asylum seeker between Member States; c) the Eurodac Regulation, which facilitates the Dublin system by setting up a database of fingerprints of asylum seekers and people who cross the external border without authorisation; d) the Asylum Procedures Directive, which sets out the procedural rules governing asylum applications, such as personal interviews and appeals; e) the Reception Conditions Directive, which sets out standards on the living conditions of asylum-seekers, such as rules on housing and welfare; and f) the Asylum Agency Regulation, which set up an EU agency (EASO) to support Member States’ processing of asylum applications.
The EU also has legislation on other aspects of migration: (short-term) visas, border controls, irregular migration, and legal migration – much of which has connections with the asylum legislation, and all of which is covered by this week’s package. For visas, the main legislation is the visa list Regulation (setting out which non-EU countries’ citizens are subject to a short-term visa requirement, or exempt from it) and the visa code (defining the criteria to obtain a short-term Schengen visa, allowing travel between all Schengen states). The visa code was amended last year, as discussed here.
For border controls, the main legislation is the Schengen Borders Code, setting out the rules on crossing external borders and the circumstances in which Schengen states can reinstate controls on internal borders, along with the Frontex Regulation, setting up an EU border agency to assist Member States. On the most recent version of the Frontex Regulation, see discussion here and here.
For irregular migration, the main legislation is the Return Directive. The Commission proposed to amend it in 2018 – on which, see analysis here and here.
For legal migration, the main legislation on admission of non-EU workers is the single permit Directive (setting out a common process and rights for workers, but not regulating admission); the Blue Card Directive (on highly paid migrants, discussed here); the seasonal workers’ Directive (discussed here); and the Directive on intra-corporate transferees (discussed here). The EU also has legislation on: non-EU students, researchers and trainees (overview here); non-EU family reunion (see summary of the legislation and case law here) and on long-term resident non-EU citizens (overview – in the context of UK citizens after Brexit – here). In 2016, the Commission proposed to revise the Blue Card Directive (see discussion here).
The UK, Ireland and Denmark have opted out of most of these laws, except some asylum law applies to the UK and Ireland, and Denmark is covered by the Schengen and Dublin rules. So are the non-EU countries associated with Schengen and Dublin (Norway, Iceland, Switzerland and Liechtenstein). There are also a number of further databases of non-EU citizens as well as Eurodac: the EU has never met a non-EU migrant who personal data it didn’t want to store and process.
The Refugee ‘Crisis’
The EU’s response to the perceived refugee ‘crisis’ was both short-term and long-term. In the short term, in 2015 the EU adopted temporary laws (discussed here) relocating some asylum seekers in principle from Italy and Greece to other Member States. A legal challenge to one of these laws failed (as discussed here), but in practice Member States accepted few relocations anyway. Earlier this year, the CJEU ruled that several Member States had breached their obligations under the laws (discussed here), but by then it was a moot point.
Longer term, the Commission proposed overhauls of the law in 2016: a) a Qualification Regulation further harmonising the law on refugee and subsidiary protection status; b) a revised Dublin Regulation, which would have set up a system of relocation of asylum seekers for future crises; c) a revised Eurodac Regulation, to take much more data from asylum seekers and other migrants; d) an Asylum Procedures Regulation, further harmonising the procedural law on asylum applications; e) a revised Reception Conditions Directive; f) a revised Asylum Agency Regulation, giving the agency more powers; and g) a new Resettlement Regulation, setting out a framework of admitting refugees directly from non-EU countries. (See my comments on some of these proposals, from back in 2016)
However, these proposals proved unsuccessful – which is the main reason for this week’s attempt to relaunch the process. In particular, an EU Council note from February 2019 summarises the diverse problems that befell each proposal. While the EU Council Presidency and the European Parliament reached agreement on the proposals on qualification, reception conditions and resettlement in June 2018, Member States refused to support the Presidency’s deal and the European Parliament refused to renegotiate (see, for instance, the Council documents on the proposals on qualification and resettlement; see also my comments on an earlier stage of the talks, when the Council had agreed its negotiation position on the qualification regulation).
On the asylum agency, the EP and Council agreed on the revised law in 2017, but the Commission proposed an amendment in 2018 to give the agency more powers; the Council could not agree on this. On Eurodac, the EP and Council only partly agreed on a text. On the procedures Regulation, the Council largely agreed its position, except on border procedures; on Dublin there was never much prospect of agreement because of the controversy over relocating asylum seekers. (For either proposal, a difficult negotiation with the European Parliament lay ahead).
In other areas too, the legislative process was difficult: the Council and EP gave up negotiating amendments to the Blue Card Directive (see the last attempt at a compromise here, and the Council negotiation mandate here), and the EP has not yet agreed a position on the Returns Directive (the Council has a negotiating position, but again it leaves out the difficult issue of border procedures; there is a draft EP position from February). Having said that, the EU has been able to agree legislation giving more powers to Frontex, as well as new laws on EU migration databases, in the last few years.
The attempted relaunch
The Commission’s new Pact on asylum and immigration (see also the roadmap on its implementation, the Q and As, and the staff working paper) does not restart the whole process from scratch. On qualification, reception conditions, resettlement, the asylum agency, the returns Directive and the Blue Card Directive, it invites the Council and Parliament to resume negotiations. But it tries to unblock the talks as a whole by tabling two amended legislative proposals and three new legislative proposals, focussing on the issues of border procedures and relocation of asylum seekers.
Screening at the border
This revised proposals start with a new proposal for screening asylum seekers at the border, which would apply to all non-EU citizens who cross an external border without authorisation, who apply for asylum while being checked at the border (without meeting the conditions for legal entry), or who are disembarked after a search and rescue operation. During the screening, these non-EU citizens are not allowed to enter the territory of a Member State, unless it becomes clear that they meet the criteria for entry. The screening at the border should take no longer than 5 days, with an extra 5 days in the event of a huge influx. (It would also be possible to apply the proposed law to those on the territory who evaded border checks; for them the deadline to complete the screening is 3 days).
Screening has six elements, as further detailed in the proposal: a health check, an identity check, registration in a database, a security check, filling out a debriefing form, and deciding on what happens next. At the end of the screening, the migrant is channelled either into the expulsion process (if no asylum claim has been made, and if the migrant does not meet the conditions for entry) or, if an asylum claim is made, into the asylum process – with an indication of whether the claim should be fast-tracked or not. It’s also possible that an asylum seeker would be relocated to another Member State. The screening is carried out by national officials, possibly with support from EU agencies.
To ensure human rights protection, there must be independent monitoring to address allegations of non-compliance with human rights. These allegations might concern breaches of EU or international law, national law on detention, access to the asylum procedure, or non-refoulement (the ban on sending people to an unsafe country). Migrants must be informed about the process and relevant EU immigration and data protection law. There is no provision for judicial review of the outcome of the screening process, although there would be review as part of the next step (asylum or return).
The revised proposal for an asylum procedures Regulation would leave in place most of the Commission’s 2016 proposal to amend the law, adding some specific further proposed amendments, which either link back to the screening proposal or aim to fast-track decisions and expulsions more generally.
On the first point, the usual rules on informing asylum applicants and registering their application would not apply until after the end of the screening. A border procedure may apply following the screening process, but Member States must apply the border procedure in cases where an asylum seeker used false documents, is a perceived national security threat, or falls within the new ground for fast-tracking cases (on which, see below). The latter obligation is subject to exceptions where a Member State has reported that a non-EU country is not cooperating on readmission; the process for dealing with that issue set out under the 2019 amendments to the visa code will then apply. Also, the border process cannot apply to unaccompanied minors or children under 12, unless they are a supposed national security risk. Further exceptions apply where the asylum seeker is vulnerable or has medical needs, the application is not inadmissible or cannot be fast-tracked, or detention conditions cannot be guaranteed. A Member State might apply the Dublin process to determine which Member State is responsible for the asylum claim during the border process. The whole border process (including any appeal) must last no more than 12 weeks, and can only be used to declare applications inadmissible or apply the new ground for fast-tracking them.
There would also be a new border expulsion procedure, where an asylum application covered by the border procedure was rejected. This is subject to its own 12-week deadline, starting from the point when the migrant is no longer allowed to remain. Much of the Return Directive would apply – but not the provisions on the time period for voluntary departure, remedies and the grounds for detention. Instead, the border expulsion procedure would have its own stricter rules on these issues.
As regards general fast-tracking, in order to speed up the expulsion process for unsuccessful applications, a rejection of an asylum application would have to either incorporate an expulsion decision or entail a simultaneous separate expulsion decision. Appeals against expulsion decisions would then be subject to the same rules as appeals against asylum decisions. If the asylum seeker comes from a country with a refugee recognition rate below 20%, his or her application must be fast-tracked (this would even apply to unaccompanied minors) – unless circumstances in that country have changed, or the asylum seeker comes from a group for whom the low recognition rate is not representative (for instance, the recognition rate might be higher for LGBT asylum-seekers from that country). Many more appeals would be subject to a one-week time limit for the rejected asylum seeker to appeal, and there could be only one level of appeal against decisions taken within a border procedure.
The revised proposal for Eurodac would build upon the 2016 proposal, which was already far-reaching: extending Eurodac to include not only fingerprints, but also photos and other personal data; reducing the age of those covered by Eurodac from 14 to 6; removing the time limits and the limits on use of the fingerprints taken from persons who had crossed the border irregularly; and creating a new obligation to collect data of all irregular migrants over age 6 (currently fingerprint data for this group cannot be stored, but can simply be checked, as an option, against the data on asylum seekers and irregular border crossers). The 2020 proposal additionally provides for interoperability with other EU migration databases, taking of personal data during the screening process, including more data on the migration status of each person, and expressly applying the law to those disembarked after a search and rescue operation.
Dublin rules on asylum responsibility
A new proposal for asylum management would replace the Dublin regulation (meaning that the Commission has withdrawn its 2016 proposal to replace that Regulation). The 2016 proposal would have created a ‘bottleneck’ in the Member State of entry, requiring that State to examine first whether many of the grounds for removing an asylum-seeker to a non-EU country apply before considering whether another Member State might be responsible for the application (because the asylum seeker’s family live there, for instance). It would also have imposed obligations directly on asylum-seekers to cooperate with the process, rather than only regulate relations between Member States. These obligations would have been enforced by punishing asylum seekers who disobeyed: removing their reception conditions (apart from emergency health care); fast-tracking their substantive asylum applications; refusing to consider new evidence from them; and continuing the asylum application process in their absence.
It would no longer be possible for asylum seekers to provide additional evidence of family links, with a view to being in the same country as a family member. Overturning a CJEU judgment (see further discussion here), unaccompanied minors would no longer have been able to make applications in multiple Member States (in the absence of a family member in any of them). However, the definition of family members would have been widened, to include siblings and families formed in a transit country. Responsibility for an asylum seeker based on the first Member State of irregular entry (a commonly applied criterion) would have applied indefinitely, rather than expire one year after entry as it does under the current rules. The ‘Sangatte clause’ (responsibility after five months of living in a second Member State, if the ‘irregular entry’ criterion no longer applies) would be dropped. The ‘sovereignty clause’, which played a key part in the 2015-16 refugee ‘crisis’ (it lets a Member State take responsibility for any application even if the Dublin rules do not require it, cf Germany accepting responsibility for Syrian asylum seekers) would have been sharply curtailed. Time limits for detention during the transfer process would be reduced. Remedies for asylum seekers would have been curtailed: they would only have seven days to appeal against a transfer; courts would have fifteen days to decide (although they could have stayed on the territory throughout); and the grounds of review would have been curtailed.
Finally, the 2016 proposal would have tackled the vexed issue of disproportionate allocation of responsibility for asylum seekers by setting up an automated system determining how many asylum seekers each Member State ‘should’ have based on their size and GDP. If a Member State were responsible for excessive numbers of applicants, Member States which were receiving fewer numbers would have to take more to help out. If they refused, they would have to pay €250,000 per applicant.
The 2020 proposal drops some of the controversial proposals from 2016, including the ‘bottleneck’ in the Member State of entry (the current rule, giving Member States an option to decide if a non-EU country is responsible for the application on narrower grounds than in the 2016 proposal, would still apply). Also, the sovereignty clause would now remain unchanged.
However, the 2020 proposal also retains parts of the 2016 proposal: the redefinition of ‘family member’ (which could be more significant now that the bottleneck is removed, unless Member States choose to apply the relevant rules on non-EU countries’ responsibility during the border procedure already); obligations for asylum seekers (redrafted slightly); some of the punishments for non-compliant asylum-seekers (the cut-off for considering evidence would stay, as would the loss of benefits except for those necessary to ensure a basic standard of living: see the CJEU case law in CIMADE and Haqbin); dropping the provision on evidence of family links; changing the rules on responsibility for unaccompanied minors; retaining part of the changes to the irregular entry criterion (it would now cease to apply after three years; the Sangatte clause would still be dropped; it would apply after search and rescue but not apply in the event of relocation); curtailing judicial review (the grounds would still be limited; the time limit to appeal would be 14 days; courts would not have a strict deadline to decide; suspensive effect would not apply in all cases); and the reduced time limits for detention.
The wholly new features of the 2020 proposal are: some vague provisions about crisis management; responsibility for an asylum application for the Member State which issued a visa or residence document which expired in the last three years (the current rule is responsibility if the visa expired less than six months ago, and the residence permit expired less than a year ago); responsibility for an asylum application for a Member State in which a non-EU citizen obtained a diploma; and the possibility for refugees or persons with subsidiary protection status to obtain EU long-term resident status after three years, rather than five.
However, the most significant feature of the new proposal is likely to be its attempt to solve the underlying issue of disproportionate allocation of asylum seekers. Rather than a mechanical approach to reallocating responsibility, the 2020 proposal now provides for a menu of ‘solidarity contributions’: relocation of asylum seekers; relocation of refugees; ‘return sponsorship’; or support for ‘capacity building’ in the Member State (or a non-EU country) facing migratory pressure. There are separate rules for search and rescue disembarkations, on the one hand, and more general migratory pressures on the other. Once the Commission determines that the latter situation exists, other Member States have to choose from the menu to offer some assistance. Ultimately the Commission will adopt a decision deciding what the contributions will be. Note that ‘return sponsorship’ comes with a ticking clock: if the persons concerned are not expelled within eight months, the sponsoring Member State must accept them on its territory.
The issue of managing asylum issues in a crisis has been carved out of the Dublin proposal into a separate proposal, which would repeal an EU law from 2001 that set up a framework for offering ‘temporary protection’ in a crisis. Note that Member States have never used the 2001 law in practice.
Compared to the 2001 law, the new proposal is integrated into the EU asylum legislation that has been adopted or proposed in the meantime. It similarly applies in the event of a ‘mass influx’ that prevents the effective functioning of the asylum system. It would apply the ‘solidarity’ process set out in the proposal to replace the Dublin rules (ie relocation of asylum seekers and other measures), with certain exceptions and shorter time limits to apply that process.
The proposal focusses on providing for possible exceptions to the usual asylum rules. In particular, during a crisis, the Commission could authorise a Member State to apply temporary derogations from the rules on border asylum procedures (extending the time limit, using the procedure to fast-track more cases), border return procedures (again extending the time limit, more easily justifying detention), or the time limit to register asylum applicants. Member States could also determine that due to force majeure, it was not possible to observe the normal time limits for registering asylum applications, applying the Dublin process for responsibility for asylum applications, or offering ‘solidarity’ to other Member States.
Finally, the new proposal, like the 2001 law, would create a potential for a form of separate ‘temporary protection’ status for the persons concerned. A Member State could suspend the consideration of asylum applications from people coming from the country facing a crisis for up to a year, in the meantime giving them status equivalent to ‘subsidiary protection’ status in the EU qualification law. After that point it would have to resume consideration of the applications. It would need the Commission’s approval, whereas the 2001 law left it to the Council to determine a situation of ‘mass influx’ and provided for the possible extension of the special rules for up to three years.
The Commission has also adopted four soft law measures. These comprise: a Recommendation on asylum crisis management; a Recommendation on resettlement and humanitarian admission; a Recommendation on cooperation between Member States on private search and rescue operations; and guidance on the applicability of EU law on smuggling of migrants – notably concluding that it cannot apply where (as in the case of law of the sea) there is an obligation to rescue.
On other issues, the Commission plan is to use current legislation – in particular the recent amendment to the visa code, which provides for sticks to make visas more difficult to get for citizens of countries which don’t cooperate on readmission of people, and carrots to make visas easier to get for citizens of countries which do cooperate on readmission. In some areas, such as the Schengen system, there will be further strategies and plans in the near future; it is not clear if this will lead to more proposed legislation.
However, on legal migration, the plan is to go further than relaunching the amendment of the Blue Card Directive, as the Commission is also planning to propose amendments to the single permit and long-term residence laws referred to above – leading respectively to more harmonisation of the law on admission of non-EU workers and enhanced possibilities for long-term resident non-EU citizens to move between Member States (nb the latter plan is separate from this week’s proposal to amend this law as regards refugees and people with subsidiary protection already). Both these plans are relevant to British citizens moving to the EU after the post-Brexit transition period – and the latter is also relevant to British citizens covered by the withdrawal agreement.
This week’s plan is less a complete restart of EU law in this area than an attempt to relaunch discussions on a blocked set of amendments to that law, which moreover focusses on a limited set of issues. Will it ‘work’? There are two different ways to answer that question.
First, will it unlock the institutional blockage? Here it should be kept in mind that the European Parliament and the Council had largely agreed on several of the 2016 proposals already; they would have been adopted in 2018 already had not the Council treated all the proposals as a package, and not gone back on agreements which the Council Presidency reached with the European Parliament. It is always open to the Council to get at least some of these proposals adopted quickly by reversing these approaches.
On the blocked proposals, the Commission has targeted the key issues of border procedures and allocation of asylum-seekers. If the former leads to more quick removals of unsuccessful applicants, the latter issue is no longer so pressing. But it is not clear if the Member States will agree to anything on border procedures, or whether such an agreement will result in more expulsions anyway – because the latter depends on the willingness of non-EU countries, which the EU cannot legislate for (and does not even address in this most recent package). And because it is uncertain whether they will result in more expulsions, Member States will be wary of agreeing to anything which either results in more obligations to accept asylum-seekers on their territory, or leaves them with the same number as before.
The idea of ‘return sponsorship’ – which reads like a grotesque parody of individuals sponsoring children in developing countries via charities – may not be appealing except to those countries like France, which have the capacity to twist arms in developing countries to accept returns. Member States might be able to agree on a replacement for the temporary protection Directive on the basis that they will never use that replacement either. And Commission threats to use infringement proceedings to enforce the law might not worry Member States who recall that the CJEU ruled on their failure to relocate asylum-seekers after the relocation law had already expired, and that the Court will soon rule on Hungary’s expulsion of the Central European University after it has already left.
As to whether the proposals will ‘work’ in terms of managing asylum flows fairly and compatibly with human rights, it is striking how much they depend upon curtailing appeal rights, even though appeals are often successful. The proposed limitation of appeal rights will also be maintained in the Dublin system; and while the proposed ‘bottleneck’ of deciding on removals to non-EU countries before applying the Dublin system has been removed, a variation on this process may well apply in the border procedures process instead. There is no new review of the assessment of the safety of non-EU countries – which is questionable in light of the many reports of abuse in Libya. While the EU is not proposing, as the wildest headbangers would want, to turn people back or refuse applications without consideration, the question is whether the fast-track consideration of applications and then appeals will constitute merely a Potemkin village of procedural rights that mean nothing in practice.
Increased detention is already a feature of the amendments proposed earlier: the reception conditions proposal would add a new ground for detention; the return Directive proposal would inevitably increase detention due to curtailing voluntary departure (as discussed here). Unfortunately the Commission’s claim in its new communication that its 2018 proposal is ‘promoting’ voluntary return is therefore simply false. Trump-style falsehoods have no place in the discussion of EU immigration or asylum law.
The latest Eurodac proposal would not do much compared to the 2016 proposal – but then, the 2016 proposal would already constitute an enormous increase in the amount of data collected and shared by that system.
Some elements of the package are more positive. The possibility for refugees and people with subsidiary protection to get EU long-term residence status earlier would be an important step toward making asylum ‘valid throughout the Union’, as referred to in the Treaties. The wider definition of family members, and the retention of the full sovereignty clause, may lead to some fairer results under the Dublin system. Future plans to improve the long-term residents’ Directive are long overdue. The Commission’s sound legal assessment that no one should be prosecuted for acting on their obligations to rescue people in distress at sea is welcome. The quasi-agreed text of the reception conditions Directive explicitly rules out Trump-style separate detention of children.
No proposals from the EU can solve the underlying political issue: a chunk of public opinion is hostile to more migration, whether in frontline Member States, other Member States, or transit countries outside the EU. The politics is bound to affect what Member States and non-EU countries alike are willing to agree to. And for the same reason, even if a set of amendments to the system is ultimately agreed, there will likely be continuing issues of implementation, especially illegal pushbacks and refusals to accept relocation.