The need to balance humanitarian responses and legal obligations while ‘ending’ irregular migratory journeys has overwhelmed the EU for the past three years. A patchwork of policies emerged as a response to the ‘refugee crisis’ of 2015. In the span of less than a year in 2015, Europe witnessed the rise of the 'Welcome' culture, the opening of the Western Balkan corridor and a growing awareness of the responsibility to save lives but also the raising of fences and border closures, political divisions within the Union and a growing determination of Member States to focus on deterrence and prevention of irregular migration cloaked in humanitarian discourse. The proposals submitted for the revision of the Common European Asylum System (CEAS) indicate a turn towards harsher, more restrictive and more difficult asylum processes. However, the crucial questions of who undertakes the responsibility for search and rescue, disembarkation and asylum processing remain unsolved. As the idea of disembarkation platforms and asylum processing centres outside of the EU take hold in the policy debate, it is important to note that this idea is neither new nor entirely untested. Externalisation of asylum processing has already taken its hold on Europe, through the EU-Turkey Statement of 18 March 2016. Hailed as a success by policy makers and Member States, the EU-Turkey Statement serves as a reminder and lesson that the EU will go to considerable lengths to avoid bold decisions on migration and asylum internally but will happily externalise responsibility to its external borders and to non-EU countries.
The EU-Turkey Statement and Greece
The refugee ‘crisis’ of 2015, was for the first half of the year an entirely Greek ‘crisis’. The country was on the brink of financial collapse with Grexit looming on the horizon. As the islands of the northern Aegean bore witness to the humanitarian crisis unfolding, little assistance or indeed few solutions came from the mainland of Greece and from Brussels.
In October of 2015 alone 211,663 migrants reached the Greek islands with approximately 80% of those disembarking on Lesvos . The overwhelming number of those were from Syria, Iraq and Afghanistan seeking to apply for asylum in Northern Europe. Greece has always been a transit country for asylum seekers. Coupled with the financial crisis and absence of any sustainable solutions for asylum applicants, the only solution for most arriving, was to move onward. A de facto transit corridor was quickly established from the islands to Athens, on to Thessaloniki and from there to Idomeni. The journey continued to Former Yugoslav Republic of Macedonia onward to Serbia and Hungary with final destinations diverging: Austria, Germany, Sweden, Netherlands, Finland etc.
The EU-Turkey Statement of 18 March 2016 (henceforth the Statement) proposed an unorthodox solution: reduce the incentives for journeys at sea by returning those who arrive on the Greek islands to Turkey. In return for accepting returnees, but also enforcing stronger border controls, a Facility for Syrians in Turkey was established to the tune of 3+3 billion euros. To further encourage accepting returnees, Member States committed that for every Syrian returned another Syrian would be resettled from Turkey to the EU. The core of the Statement is returns, primarily as a measure of deterrence for future arrivals. The Statement was combined with the closure of the Balkan corridor, effectively trapping migrants in Greece. Relocation was proposed as a way of assisting Greece (and Italy), however the number of those who could participate was limited since eligible nationalities had to meet the 75% threshold of positive recognition in asylum applications. 
The proponents of the Statement argued it was the only pragmatic way forward since it combined (in theory) speedy asylum processing, returns of those rejected, with resettlement and improvement of conditions in Turkey. Perhaps more crucially, a moral dimension was added; the deal would ensure reduction in loss of life at sea.
The opponents of the Statement argued it endangered how asylum is approached, made journeys longer and even riskier. More importantly those who opposed the deal argued it was a Faustian bargain with Turkey, a normative and moral loss for Europe. More than two years on, its success on the ground remains contested. Nonetheless, it is currently the blueprint for the way forward regarding the partnerships the EU will seek to pursue with third countries. Additionally, the Statement has been hugely influential in reopening the debate about who is a migrant and who is a refugee and whether protection is indeed a right or perhaps a privilege.
For Greece, the implementation of the Statement resulted in two important changes. First, the islands transformed overnight into detention sites for those who arrived post 20 March 2016 from Turkey, with important implications for the migrants as well as the local communities. Secondly, the Statement de facto altered the implementation of the Common European Asylum System with Greece currently applying two different asylum procedures, one in the islands and another in the mainland.
A complex implementation
There has always been an aspect of externalisation of protection responsibilities in the Common European Asylum System, specifically in the Dublin Regulation. The Dublin Regulation rules determine responsibility of asylum processing, thereby de facto transferring the overwhelming burden to the front-line states at the external borders of the Union .
The EU-Turkey Statement reinforces this externalisation and takes a step further, requiring that Greece and Turkey share responsibility. Greece has to walk a fine line between the 1951 Convention and speedy returns. The deal proposes individualised processing of every asylum applicant, with a caveat. Before the merit of the case is decided, the Asylum Service should examine the (in)admissibility of claims on the basis of the ‘safe third country’ and ‘first country of asylum’ rules. Those whose claims are found inadmissible, or whose asylum has been rejected on merit, are in theory returnable to Turkey that also bears responsibility for ensuring access to protection similar to what is offered in the 1951 Convention.
Significant changes took place in Greece when it started to implement the deal creating an administrative labyrinth impossible to navigate. To this day, it is the administrative complexity of implementing the Statement that carries a significant portion of the blame for the situation in the hotspots. If what the ‘founding fathers’ of the deal envisaged would be swift asylum processing and removals, it is safe to say they had little understanding for the legal and administrative intricacies.
On 3 April 2016, Law 4375/2016 appeared in the Greek Official Government Gazette seeking to define the legal standing of the operations on the five islands designated as ‘hotspots’. The facilities applying the hotspot approach were renamed ‘Reception and Identification Centres’ (RICs). The government incorporated in the Law the EU recast Asylum Procedures Directive (APD)13 to ensure that the (in)admissibility process is applied and returns take place based on the Greece-Turkey bilateral readmission agreement (signed in 2001) as well as the EU-Turkey Readmission Agreement of 2013. The asylum officer has to decide in each case whether the individual has sufficient ties to the third country or is able to return to that third country and apply for asylum there. The European Commission has argued that returns are then possible, particularly for the Syrians (covered by the Temporary Protection status in Turkey) who can apply for and receive protection in Turkey. In practice, unless a Syrian can prove vulnerability and/or fear of persecution in Turkey, his or her claim will be deemed inadmissible and a return order issued. For other nationalities, a mixed procedure is followed where admissibility and eligibility procedures interchange. Greece is the only country at present that follows different procedures for different nationalities. It is also the only country to follow an (in)admissibility procedure on the islands and a regular procedure in the mainland. In other words, Greece by virtue of the Statement is a testing ground for what asylum could look like in the future.
As part of the support measures to Greece, FRONTEX and the European Asylum Support Office  increased their presence in the hotspots while Law 4375/2016 added a rapid procedure to the RICs at times of emergency, different from the accelerated border procedure. Under the rapid procedure, EASO’s personnel is allowed to conduct asylum interviews and recommend decisions on (in)admissibility as regards returns to Turkey. The arrangement has been highly criticised as ‘problematic’ , with individual assessment of each asylum case sidestepped for quick decisions on admissibility. A decision on admissibility usually does not result in return to Turkey. The option of appeal is utilized by most, resulting in a wait of at least two years for a final decision. In fact, since the implementation of the Statement, most migrants are still on the islands .
If the applications of Syrians are usually deemed inadmissible, all others are often found admissible, since it is difficult to argue safeguards exist for non-Syrians in Turkey. Applications are examined on merit, with many likely to be rejected. The process takes months, if not years including appeal, and while waiting for a decision, migrants are expected to remain trapped on the islands.
Excluded from the procedure on the islands are vulnerable populations. Vulnerability is one of two ways to leave the islands . Vulnerability assessment focuses on medical reasons (and less on psychological), age and gender. Those deemed the most vulnerable are eligible for transfer to the mainland and one of the accommodation facilities in place. In a tragic twist, vulnerability and particularly physical vulnerability (which is easier to identify) is much sought after by asylum seekers. However, even if vulnerable, transfer to the mainland does take often months.
The implementation of the Statement has created a unique situation where thousands of people wait for months and years on the five islands, either for their case to receive positive recognition, or to be deemed vulnerable and transferred to the mainland, or for their return to Turkey to take place.
Greece is no longer facing an emergency, but geographical restriction of movement continues. The argument of the Greek government is that it is necessary to achieve returns, since Turkey will only accept returnees that have entered through the sea border . Furthermore, allowing movement from the islands to the mainland would be seen as a pull factor for more migrants to enter. Despite having no evidence to this, the policy is maintained on both grounds.
A completely different situation exists on the mainland and land border. Those on the mainland had either arrived prior to 20 March 2016 or were transferred in preparation for the Statement. Initially, camps had been set up across the mainland. Technically the responsibility for the camps lay with the Ministry of Migration. The new Ministry lacked both the human and financial resources, as well as expertise in site management. As a result, most camps remain under the International Organisation for Migration (IOM) or various NGOs with previous experience in refugee emergencies. Overall camps did not meet reception standards. Reports quickly emerged of insecure conditions for women and families, outbreaks of violence and mental and health problems amongst migrants. In an effort to transfer the population registered in the relocation program out of the camps and into urban centres, the Emergency Support to Integration & Accommodation – ESTIA programme was set up. ESTIA is the largest humanitarian assistance offered to a Member State of the EU. The accommodation scheme is funded by the European Union Civil Protection and Humanitarian Aid (DG ECHO) and realised by UNHCR through its implemented partners (NGOs). Twenty-Four (24) NGOs are assisting in the accommodation scheme, with another five in the cash aid scheme . As the accommodation and cash aid scheme expanded, asylum applicants were also allowed to move from camps to the apartments with a priority once more on vulnerable groups and particularly families. In 2017 a budget of €135.5 million was provided for 22,000 urban accommodation places and pre-defined monthly cash grants for refugees and asylum-seekers in Greece . Those in the mainland fall under the regular asylum procedure, irrespective of nationality, and enjoy freedom of movement, access to relocation (dependent on nationality) and access to accommodation.
Impact of the Statement on asylum processing
The Statement is being hailed as a success in reducing the number of arrivals to Greece and by extension to the EU. There is some truth to that argument, though it is not entirely valid. Numbers have reduced and returned to pre-2015 levels though it is the closure of the Western Balkan route that seems to have had the most impact.
What has undoubtedly changed, is the way asylum and protection is understood and applied on the ground. The usage of expedited asylum determination procedures on the islands, and a tendency to differentiate individuals on the basis of their nationality is a critical step backwards in asylum policy. Asylum is an individualised process irrespective of one’s nationality and yet, the nationality segmentation is embedded in the Statement. From returns to the 1+1, the Statement was designed to address the arrival of the Syrians only. The (in)admissibility procedure assumes that the difference between migrant and refugee is easily discernible and largely based on one’s nationality, as is access to protection.
The Statement is swiftly applied at 1st instance since Turkey is assumed to be a ‘safe third country’ despite the fact that there is no consistent nor adequate information on conditions of reception in Turkey, the legal status of Syrians and how it is implemented in practice, access to work and for non-Syrians, actual access to protection and guarantees against deportation.
The indiscriminate use of detention as a measure of deterrence is also highly problematic not only because it fails to generate a higher ratio of returns (the end goal) but due to the conditions of detention. The RICs on the islands have consistently been over and above capacity lacking basic amenities including hot water, toilets and medical resources. On a weekly basis the Ministry of Migration releases a situation picture of the islands of Eastern Aegean. In the last week of August 2018, with weekly arrivals at 68 persons, the islands of Lesvos and Samos were still over capacity and at breaking point. By September, NGOs like Doctors without Borders described a humanitarian crisis unfolding in the hotspots.
Capacity of RIC’s and actual population on the islands as of 12/09/2018
Source: Ministry Of Digital Information, National Situational Picture regarding the islands of the Eastern Aegean Sea (12/9/2018), compiled by author
*Occ. Occupancy, Cap. Capacity
**The migrants present on the islands include those in R.I.C., UNHCR and NGO facilities, Hellenic police facilities and detention facilities as well as makeshift camps.
NGOs have repeatedly issued calls over the situation in the hotspots. Uncertainty over when the asylum application will be processed, its outcome but also whether returns will take place exacerbate people’s psychological problems and pre-existing trauma. Incidents of self-harm have become common as have suicide attempts. Outbreaks of violence between groups, partly born due to ethnic and/or cultural differences but more commonly due to the very conditions of detention they are placed in, are also common .
Despite numerous criticisms, detention has the strong support of the European Commission. Already in 2016, the Joint Action Plan with Turkey suggested that ‘The authorities [must] ensure that the whereabouts of asylum applicants are known as long as their application is pending (including through possible use of closed centres)’ .
At the time of writing the discussion has shifted into increasing the detention capacity on the islands either by expanding the present RIC’s or by setting up additional ones. The plan has been delayed due to the vehement opposition of locals who feel the islands have been transformed into prisons.
On the other hand, Greece has failed in fulfilling its obligations towards arrivals and much more can be done within the country to address the situation.
There is an immediate need for transferring migrants from the islands to camps on the mainland. The hotspots should return to what they were originally created for: screening centres for short stay. Considering the critical aspect, the Statement is not working and returns are not taking place. There is little justification for maintaining thousands of people in conditions that violate their dignity and safety.
The hotspots require major expansion to be able to address significant volumes of arrivals, even for short term stay. NGOs need to return to operate for medical assistance and vulnerability assessments currently mostly done by the Hellenic Centre for Disease Control & Prevention that is unprepared and unfamiliar with the population at hand.
The First Reception, though created in 2011 under the asylum reform, has failed thus far to deliver as a service and requires strengthening in both funding and mandate but especially personnel that is trained to be present in the field.
Additionally, an Integration Plan is needed that ensures a transition from cash-aid assistance and accommodation assistance to becoming an active member of the society. Language training, vocational training, skill development and support for entrepreneurial schemes of refugees are needed and missing.
More importantly though, a different asylum system is needed in Europe as a whole, one that allows for some element of choice by the individual; that acknowledges ties, history and personal preferences without ignoring the needs and capabilities of the Member States. A system that genuinely shares responsibility. Unfortunately, this is not the system envisaged at present.
The future of asylum
The impact of the hotspots and the Statement on asylum extends well beyond Greece. It is visible in the reforms proposed for the revision of the CEAS and particularly the Dublin IV, a system already revealed as deeply problematic in 2015-2016.
The new proposals reinforce the distribution of responsibility to front line states and expands the time limit in which the first country of arrival must carry responsibility for the applicant (from 12 months to 2 years). The admissibility assessment is introduced for all to accelerate the determination procedure. Secondary movement is also penalised for those who have entered from one Member State and sought to apply in another. They become subject to an accelerated procedure after being returned to the first country of entry. This essentially pre-determines the claim as manifestly unfounded.
The ‘burden sharing’ clause proposed in 2017 is a ‘corrective allocation mechanism’. When a Member State exceeds by 150% its allocation of asylum applications under a ‘reference key’, it will cease to receive asylum applications until the percentage reduces. Interestingly, it will continue to receive Dublin returnees. The proposal in practice, would lead to a situation where a Member State first has to face a ‘crisis in numbers’ and then allow for a temporary suspension of applications, until the percentage drops below the threshold. The Dublin reform has failed to move forward largely due to the mechanism proposed. This is indicative of the unwillingness of Member States to ‘share the burden’, even as a last resort. Some Member States wish for the mechanism to be voluntary, while others would prefer the option to substitute redistribution of asylum applicants with other measures of support (e.g. financial support).
Further changes are incorporated in the proposed Qualification Regulation that includes punitive measures for applicants that fail to cooperate with competent authorities, remain in the country or participate in integration measures, if such is prescribed. The proposed EURODAC will collect facial images and fingerprints not only of asylum applicants but those apprehended for irregular border crossing and/or irregular stay in a Member State.
The complexity of the reforms proposed should not be understated nor the impact on receiving protection. Should these proposals proceed, asylum will be even harder to receive and even less ‘common’ than before. Perhaps more crucially, the past three years have reinforced the idea in the EU that the solution for asylum seeking flows should be found ‘elsewhere’, outside the boundaries of the Union, in third countries or even with the European continent (but not the EU).
Regional disembarkation arrangements
On 11 June, Italy's new Interior Minister Matteo Salvini blocked the Aquarius rescue ship carrying 629 refugees and migrants from docking at Italian ports. There is little doubt that the Italian policy endangered the lives of those on the Aquarius ship while undermining the efforts of NGOs to assist with Search and Rescue operations. On the other hand, the Italian argument is not entirely unfounded. Search and Rescue (SAR) operations at sea in the Mediterranean are linked with Dublin, i.e. the country where disembarkation takes place is also responsible for asylum processing and fingerprinting.
The Commission, following the ‘Mini-Summit’ of 24 June 2018, began assessing different options for disembarkation. Three scenarios were proposed: 1) a regional arrangement for disembarkation in third countries for migrants rescued in the territorial sea of a third country or in international waters; 2) a regional arrangement for disembarkation of migrants in EU Member States; and 3) external processing of asylum applications and/or return procedure to a third country.
The third option is not new. It was first put forth in 2003 by the UK but was rejected at the time for fear it violated the EU legal and normative framework on asylum. Though legally nothing has changed, the fact the idea keeps coming back to the table indicates the desire to outsource asylum management as far beyond the EU as possible.
The second option draws from the relocation-hotspot model undertaken under the EU-Turkey Statement and it is an unlikely scenario to unfold since it would require intra-EU cooperation. In this scenario disembarkation takes place in the frontline countries, as is the case now, for migrants rescued by vessels of EU Member States. However, those eligible for protection will be relocated to other Member States. It is unclear how this relocation would take place. Screening and separation of asylum applicants from economic migrants would likely still take place in hotspots in the countries of arrival, and the timeline between registration-asylum processing and relocation would be critical in the success of the scheme. Member States would need to be convinced and implement relocation, ensure its not as highly selective as was the case in the previous scheme, and the question of what would happen while the initial screening takes place remains unclear. Would arrivals be detained? And if so, for how long? What would happen to those rejected, considering the current low level of returns across the EU?
If these options appear difficult for legal and political reasons, it is the first option that is perhaps the most controversial. Those rescued by third country vessels or by EU Member States in international waters or waters of third countries (i.e. some Partnership agreement has been established), would disembark in the third country and their asylum would be processed by UNHCR, in centres run by IOM-UNHCR. The regional disembarkation platform concept ensures first reception and asylum take place as far away as possible from the territory of the EU. The assumption is that the third countries chosen would be considered ‘safe’ to ensure the non-refoulement obligation is respected. It is more realistic to assume they will be approached because of their dependency on EU financial aid and development assistance. Those whose applications have been approved by UNHCR would then be resettled.
The specifics again are unclear, particularly how this resettlement would work. Would all Member States participate based on quotas or would it be voluntary? Would they deploy liaison officers or EASO also in the centres or leave asylum entirely to UNHCR? How long would the process take and who would make the financial commitment (of significant size) to ensure the centres are up to standards? What would happen to those rejected? How would IOM enforce returns? These are only some of the issues of concern, with the legal framework and its feasibility in murky waters.
Breaking the link between disembarkation and asylum processing is arguably a way forward. It would reduce the responsibilities of front-line states. A system of quick resettlement would also ensure burden sharing and protection to those in need. However, for the latter scenario to work in favour of the asylum applicants, those who do receive protection must then be transferred to an EU Member State. Ideally, the process should factor in the needs and choice of migrants (for example, family ties, language, vulnerability etc). It would require all Member States partake in the scheme with an annual quota revised to meet both needs and demand. The European Commission would have to find a way of ensuring Member States fulfil their obligations in a timeline manner and provide strong monitoring of both UNHCR and IOM activities as well as the third country. Even if all these criteria are met, myriad of difficulties exist and pitfalls. As a theoretical construct, a new framework for disembarkation and asylum screening is not a terrible idea. The problem is that rather than looking for solutions within the Union, Member States are increasingly turning to neighbouring third countries and beyond. Three years on since the refugee “crisis” reached Europe, the case of Greece and the hotspots highlight that the cost for the country that functions as a ‘processing site’, but more importantly for the migrants, is too high to ignore. Rather than focusing on ways to externalise responsibility, it is indeed imperative to look inwards and create a functioning asylum policy that perceives migration not only as a challenge but also an opportunity, offers proactive support for Member States under pressure, a strong resettlement policy and a functioning disembarkation system structured around internal burden sharing.
 All data available through UN Portal: Refugee Situation, available at https://data2.unhcr.org/en/situations/mediterranean/location/5179 . 856,723 persons landed on the islands of Northern Aegean in 2015, with arrivals peaking from August to November Five islands were on the receiving end of arrivals: Lesvos, Chios, Samos, Kos and Leros. These would eventually be also the five islands to implement the hotspot approach.
 This essentially meant that only Syrians, Eritreans and Iraqis could participate initially, with Iraqis eventually also removed. The Afghans, making up almost 40% of arrivals to Greece were left out of the relocation scheme.
 Triandafyllidou,A, & Dimitriadi,A. (2014). Deterrence and Protection in the EU’s Migration Policy, The International Spectator, 49, 4, pp. 146-162
 Initially, EASO’s presence in Greece was not included the legal framework applied in Greece. The Agency has been criticised for a preference towards inadmissibility decisions.
 United Nations Human Rights Council, Report of the Special Rapporteur on the human rights of migrants on
his mission to Greece, A/HRC/35/25/Add.2, Par 82, 24 April 2017, available at: www.refworld.org/docid/593a8b8e4.html
 From April 2016 when returns began until 30 June 2018, a total of 1650 persons have been returned to Turkey. 40% of those are Pakistani nationals, with Syrians constituting 18%. Returnees in general had either opted out of the asylum application or withdrew their asylum claim
 The other way of leaving the island, is for one’s case to be examined on merit and deemed eligible for protection
 To ensure this is the case Turkey has sent liaison officers to the Greek islands who validate the lists of returnees.
 Cash assistance to individuals and/or families in the form of a monthly stipend to cover personal and food expenses. Rent is covered by the scheme, as are running costs.
 ESTIA Program available at: http://estia.unhcr.gr/en/home/ ; cash aid is also available since 2017 to the migrants on the islands and hotspots.
 ECRE (28 April 2017), Conditions in the hotspots are chilling. Available at https://www.ecre.org/conditions-in-hotspots-is-chilling/
 Joint action plan on the implementation of the EU-Turkey statement, December 2016.