1. Introduction
1. For nearly two decades, the
European Union (EU) has elaborated legislation and policies aimed
at creating a Common European Asylum System. The stated objective
is to achieve “high standards and stronger co-operation to ensure
that asylum seekers are treated equally in an open and fair system
– wherever they apply”.
Based on the Charter of Fundamental
Rights of the European Union,
the European Convention on Human
Rights (ETS No. 5, “the Convention”)
and the United Nations 1951 Convention
relating to the Status of Refugees (“Geneva Convention”) and its
1967 Protocol,
the EU has thus developed higher
standards through EU legislation.
2. Parallel to the development of internal standards, the EU
has developed an external dimension of EU migration law, with the
main objective of combating irregular migration. With that aim,
the European Union is co-operating with many countries of origin
and transit in Africa, south-east Europe and the Middle East as
well as with Afghanistan and Asia. Such co-operation was initially
primarily based on bilateral agreements, but has shifted to an increased
use of the external competence of the EU under Article 216 of the
Treaty on the Functioning of the European Union (TFEU) to conclude
international agreements, in order to increase leverage to achieve
co-operation from those third countries.
3. On the basis of the Global Approach to Migration and Mobility,
different forms of co-operation have been
agreed upon with third countries, such as Mobility Partnerships,
Common Agendas on Migration and Mobility, readmission agreements
and, more recently, tailor-made global compacts. Co-operation includes agreements
on strengthening border controls and readmission of undocumented
migrants in exchange for trade benefits, financial support, development
aid, visa facilitation and resettlement of refugees. Under its Migration
Partnership Framework of 2016,
however, the EU has shifted its emphasis
from positive incentives (“more for more”) to sanctioning non-co-operation
through less funds, trade or development aid (“less for less”).
4. One of the most prominent financing instruments to support
the implementation of the Partnership Framework on Migration is
the Emergency Trust Fund for Africa (EUTF).
Although the EUTF is framed as an emergency
instrument, most of its resources consist of Official Development
Assistance (ODA), which is intended to fund long-term development
programmes.
5. The Parliamentary Assembly has adapted its position on the
externalisation of asylum procedures according to developments and
issues arising. It supported the suggestion made by the Italian
Minister of the Interior that camps be set up to process asylum
requests in North Africa in
Resolution
2000 (2014) on the large-scale arrival of mixed migratory flows
on Italian shores, but raised concerns in
Resolution 2109 (2016) on the situation of refugees and migrants under the
EU–Turkey Agreement of 18 March 2016, “that the EU–Turkey Agreement
raises
several serious human rights issues relating to both its substance
and its implementation now and in the future” and that “detention
of asylum seekers in the ‘hotspots’ on the Aegean islands may be incompatible
with the requirements of the European Convention on Human Rights
(ETS No. 5), due notably to procedural failures undermining the
legal grounds for detention and inadequate detention conditions”.
6. These statements may appear contradictory in substance, but
the Assembly’s positions have always been based on the assumption
that solutions have to be found to improve migration management
and alleviate the pressure at Europe’s borders. Part of these solutions
may be found outside Europe’s borders, thus reducing the numbers
of migrants risking their lives travelling to Europe, but any external
or extraterritorial solutions are only possible if adequate safeguards
are in place to guarantee migrants’ rights to adequate asylum procedures and
other human rights and that they are not solely driven by the aim
of evacuating European member States’ shared responsibilities to
resettle refugees.
7. The human rights situation is also taken into account, therefore,
by the Assembly in countries where the EU proposes the creation
of hotspots outside its borders for dealing with applications for
asylum. This latter aspect is, however, the subject of the report
being prepared in parallel by my colleague Mr Domagoj Hajduković on
the legal and practical requirements for extra-territorial processing
of asylum claims.
The
potentialities of these types of asylum procedures will be developed
in this report. My report, as stated in its title, focuses on the
concrete evidence of shortcomings in situations where the processing
is to a greater or lesser extent “handed over” to the authorities
of countries outside the EU, rather than being managed by member
States in a third country.
8. This report was enhanced by exchanges of views in the Committee
on Migration, Refugees and Displaced Persons held in January and
April 2018 with the office of the Office of the High Representative
of the European Union for Foreign Affairs and Security Policy of
the European Commission, the Head of the International Organization
for Migration (IOM) Mission in Libya and with the Director for Europe
of Amnesty International. I also carried out a fact-finding mission
to Turkey on 31 May and 1 June 2018, which enabled me to gain first-hand
and up-to-date information on the implementation of the EU–Turkey
deal two years after its signature, on which I reported at the time.
I would like to thank the Turkish Parliamentary delegation for its
co-operation, which enabled me to hold meetings with all the relevant
ministries as well as visiting the Kırklareli readmission centre.
During the visit I met with the IOM, UNHCR, as well as Amnesty International
and refugee rights NGOs and lawyers.
2. The European Union framework for asylum
and migration
9. The current legal standards
of the EU setting out the fundamental right to asylum are the Asylum Procedures
Directive 2013/32/EU,
the Reception Conditions Directive
2013/33/EU,
the Qualification Directive 2011/95/EU
(which are currently in the process
of being revised), as well as the EU Charter on Fundamental Rights.
The application of those instruments should in any case be compliant
with the European Convention on Human Rights and the Geneva Convention.
2.1. Extension
of European Union norms
10. Many of the 19 member States
of the Council of Europe which are not members of the EU are engaged in
an application process for membership, or co-operate closely with
the EU in other frameworks. This means that many of these countries
are implementing EU legislation and policies to a greater or lesser
extent. Some countries, without being candidates for accession,
contribute to the funding of EU programmes, such as Norway.
Norway has also recently demonstrated
high levels of human rights protection for refugees, for instance
in a Supreme Court ruling of 28 March 2018 which stated that a “significant
and stable change” in the country of origin was needed to justify
cessation of refugee status.
11. Some Council of Europe member States outside the European
Union, including the Russian Federation and Turkey, are major countries
of transit for migrants. Turkey, currently hosting 3.4 million registered
Syrian refugees,
was allocated
by mid-October 2017 2.9 billion euros under the EU Facility for
Refugees in Turkey.
The non-member
States of the EU within the Council of Europe are bound by the European Convention
on Human Rights and thus also by the prohibition of
refoulement under its Article 3.
Furthermore, the Geneva Convention is applicable as well, albeit
in Turkey still with a geographical limitation, meaning that only
refugees from countries of the Council of Europe are eligible for
recognition as such under this convention.
2.2. Lower
human rights standards in countries outside European Union jurisdiction
12. In the framework of the external
dimension of EU migration policy, the EU has developed specific
multi-annual funds with non-EU States, in particular with countries
in northern Africa, the Middle East and Turkey. As with other EU
programmes, their implementation requires transparency, monitoring
and evaluation, accountability mechanisms, as well as preliminary
and ongoing assessments of their impact on fundamental rights.
13. Standards on asylum in most partner countries are below the
level of the EU asylum acquis.
This implies that refugees prevented from travelling to EU territory
because of the EU’s external co-operation are not able to invoke
the same rights as they would have been able to in the EU. To what
extent and in which cases does that make the EU responsible for
possible human rights violations in the partner countries? An important
issue to be addressed is whether the delegation of asylum management
by the EU implies that human rights obligations are also delegated
to these countries. If so, what standards should apply in partner
countries? In my view, those third countries should at least comply
with the standards of the European Convention on Human Rights and
the Geneva Convention in their legislation, policies and practice.
Only with these standards in place, can member States consider entering
into co-operation on migration without running the risk of violating
basic human rights of migrants and refugees.
14. As a consequence of externalisation, immigration control tasks
are performed more and more by the transit countries instead of
by the member States themselves. Although this co-operation has
developed into comprehensive partnerships with a large number of
transit countries, only very little is known about its actual consequences.
What
is known, is that what may initially have been intended as “transit”
migration has in practice become more extended, even as far as semi-permanent
immigration, as migrants are prevented from travelling onward to
the EU, or are being returned from there.
15. These transit countries may have been persuaded to readmit
migrants or prevent them from travelling further, but we do not
know the extent to which the governments concerned take responsibility
for those transit migrants. What is clear is that these countries
conclude readmission agreements with other transit countries further
away, which indicates that there is a will to expel transit migrants,
implying the risk of indirect
refoulement.
3. Which
rights are threatened by the externalisation of migration management?
16. The externalisation of immigration
policies by the EU thus brings with it essential human rights issues, especially
regarding the right to leave a country, access to asylum and to
the rights under the Geneva Convention, the prohibition of refoulement and of deprivation of
liberty, the right to dignity and access to basic needs, as well
as the right to an effective remedy in the transit countries. In
controlling their borders, member States are bound by the human
rights of migrants protected under international and EU law. This
also implies that any external co-operation with potential human
rights impacts requires clear criteria on the standards to which
those third countries should adhere, but also clear answers to questions
related to the responsibility of member States and the EU for respect
of human rights.
17. The question of responsibility of the EU relates primarily
to migrants who are prevented from accessing EU territory or have
been returned to transit countries. What does this external co-operation
mean for their access to an asylum procedure, their right to human
dignity and effective remedy and for their rights under the Geneva
Convention in these countries? And if their human rights are violated
in the third country, what legal remedies does (or should) a migrant
have towards the EU or EU member States?
3.1. Border
controls – saving lives but putting human dignity and freedom of
movement at risk
18. Since the sudden increase in
arrivals of refugees into the EU in 2015, the prevention of and
combat against irregular migration to the EU is one of the few areas
where member States are able to find common ground, which may have
contributed to the increasing concentration on ways of stopping
refugees and irregular migrants crossing the EU’s external borders.
The EU has been devoting an increasing part of its budget for migration
policies on controlling its frontiers, for instance building up
the capacity and geographical remit of the European Agency for the
Management of Operational Cooperation at the External Borders of
the Member States of the European Union (Frontex).
19. Frontex launched a new joint operation on 1 February 2018,
replacing operation Triton in place since 2014. The agency announced
that the new operation would continue to include search and rescue,
while at the same time focusing more on law enforcement over an
operational area spanning the Central Mediterranean Sea and incoming
flows from Algeria, Tunisia, Libya, Egypt, Turkey and Albania. The
operation will also assist Italy in tracking down criminal activities
such as drug smuggling across the Adriatic, and continue its presence in
the hotspots in Italy to assist the national authorities in registering
migrants. Frontex vessels will continue search and rescue operations
under the co-ordination of the responsible Maritime Rescue Co-ordination Centres.
20. One of the problems with the EU objective to improve the human
rights situation in partner countries is that those programmes are
not conducted with such a systematic and formal structure and that
they aim for long-term improvements, whilst co-operation on border
controls seems to have a more structural basis, with an immediate
effect. Furthermore, the co-operation on border control receives
significantly more funding than co-operation for the improvement
of human rights.
21. The main problem here is that a level of protection is not
a prerequisite for co-operation on border control or readmission
and there is no suspension mechanism for situations in which a transit
country does not afford adequate protection. This raises the question
of which aspects of the co-operation are prioritised, if human rights
and border control come into conflict.
22. The human rights risks involved in the external dimension
are not theoretical. Amnesty International, for instance, has evidenced
that “the demands being placed on third countries to prevent irregular
departures to Europe put refugees, asylum-seekers and migrants in
those countries at risk of prolonged and arbitrary detention,
refoulement, and ill-treatment”.
Many anti-smuggling
measures merely lead to shifts in smuggling routes which thus become
longer, more dangerous and more expensive.
3.2. From
pillar to post: readmission agreements could leave migrants stranded
23. In the New York Declaration
on Refugees and Migrants of September 2016, the EU Member States promised
to contribute to “a more equitable sharing of the burden and responsibility
for hosting and supporting the world’s refugees, while taking account
of existing contributions and the differing capacities and resources among
States”.
24. The most problematic aspect of this type of co-operation is
that with transit countries. Formal or non-formal agreements on
readmission by a transit country go beyond the narrow meaning of
repatriation as set out in Article 63.3.b of
the Treaty of Amsterdam, which refers to the obligation of a State
to readmit its own citizens. There is no legal obligation for a
transit country to readmit a third country national from another
State. But EU Member States have increasingly shifted their focus
to transit countries, especially to those sharing their borders
with EU territory. The EU thus envisages creating a “buffer zone”
around its territory by agreeing that neighbouring countries will
readmit transit migrants who have passed through their territory
on their way to the European Union. This could have a knock-on effect,
by which the transit countries concerned will restrict incoming
and outgoing migration both from their own neighbouring countries
and to the European Union.
25. Readmission agreements reduce the chances for migrants to
invoke human rights in the EU, but do not include any guarantee
that the transit country offers effective protection to asylum seekers.
The EU member States invest gradually more in the asylum systems
of their neighbouring countries, but the EU, as the party “asking
for assistance”, is reluctant to negotiate with third countries
on human rights in the context of readmission agreements. The agreements
also make migration into a transit country from another non-EU neighbouring
country less likely, through increased visa restrictions and/or
border control. This may lead to migrants being stranded in an even
more remote transit country with less protection guarantees than
the partner country itself. This chain effect can also severely
hinder any mobility opportunities, especially for migrants from
less wealthy countries, and therefore also obstruct regional migration
programmes such as those carried out in the framework of the Economic
Community of West African States (ECOWAS).
These
effects may hinder the objectives of the Global Compact for Safe,
Orderly and Regular Migration and the Global Compact on Refugees,
which are still under negotiation.
26. A final disadvantage of readmission agreements is that they
oblige a transit country to readmit undocumented migrants from the
EU but do not make any provision for access to basic needs such
as the right to housing, health care, primary education, work and
social welfare. The potential chain of transit is a threat to the
principle of human dignity as enshrined in international law, in
particular if the migrant is unable to return to his home country.
Bearing in mind the risk of a (legal) limbo situation for the returnee
in a transit country, the European Commission urged member States
to always give priority to returning undocumented migrants to their
country of origin,
but the extent to which
the member States comply with this principle is unclear. They at
least did not follow up the Commission’s advice to be reticent to
include third country nationals in readmission agreements, as the
need for more incentives would complicate the negotiations substantially.
Thus, negotiations
with important transit countries such as Morocco have resulted in
a deadlock due to the EU’s wish to incorporate clauses on the readmission
of third country nationals.
27. The usual type of agreement on migration currently concluded
by the EU is more comprehensive, involving different types of obligations
and measures on different policy areas. The need for a comprehensive approach
led to the adoption of the Global Approach to Migration and Mobility
(GAMM) meant to safeguard a coherent internal and external migration
policy.
The European Agenda on Migration
(EAM) formulated in the light of Europe’s migration crisis in 2015
confirms the strengthening of the outward, external emphasis of
the EU’s migration policy. The European Commission increasingly
focused on the lack of co-operation on the part of partner countries
and made proposals to foster their willingness to co-operate.
28. The New Partnership Framework published in June 2016 aims
to adopt tailor-made “compacts” with priority partner countries,
in which all instruments, tools and leverage are put together, “to
better manage migration in full respect of our humanitarian and
human rights obligations”.
Here, the principle of conditionality
has been put at the centre of the policy, implying that economic
support of third countries depends on their performances on readmission
and border control. The European Commission recommended that not
only positive but also negative incentives be applied, by using
all EU policy areas with the exception of humanitarian aid. With
these tailor-made solutions, the informal nature of co-operation
has increased, with the result that human rights are not always
explicitly addressed and that transparency and democratic control is
lacking. These omissions are clear obstacles to the legitimacy of
the decisions and the establishment of accountability of the EU
and its member States.
3.3. The
internal-external dimension: the “safe third country” concept
29. An important threat to the
sharing of responsibilities reaffirmed in the New York Declaration
is the redefinition and mandatory application of the safe third
country concept, which member States appeared to agree upon during
the negotiations on the draft Regulation on Asylum Procedures, but
which is now hindered by differences in national interpretation
and the mixed competences (national and EU) the current system entails.
30. In cases where return takes place in the context of this safe
third country concept, asylum claims are not examined by an EU member
State and are thus dependent on the effectiveness of the asylum
system in the third country. The EU–Turkey Declaration is based
on the presumption of Turkey being a safe third country for refugees,
and in a number of cases the Greek judges have applied this concept
to asylum claims of refugees entering from Turkey. Many EU member
States aspire to adopt the same approach with regard to North African countries.
These countries, however, are not even bound by the European Convention
on Human Rights, nor do they offer satisfactory asylum and reception
systems. Readmission agreements do not currently include any guarantee
that the transit country has an effective protection regime in place
for asylum seekers.
31. The proposal for a regulation on Asylum Procedures, released
in July 2016, clearly reflects one of the objectives of concluding
migration deals with third countries: the transfer of responsibility
for refugees.
The main changes in the proposed
safe third country concept are that member States are obliged to
apply the safe third country concept, but are allowed to designate
third countries as safe in addition to the common EU list of safe
third countries, unless the European Commission objects to it.
Furthermore,
the Commission proposed to water down the criteria for a safe third
country. According to the proposal, ratification of the Geneva Convention
is not an obligation for third countries, and guaranteeing the “core
rights” of the convention by third countries should be sufficient.
Criteria for the personal connection between the refugee and the
transit country are left to the member States.
32. These changes have increased the risk that refugees are handed
over to a country where they will not enjoy the full rights of the
Geneva Convention and will not be granted a durable solution, as
they lack the possibility to build a new future together with their
family members. As already foreseen in the London resolution, the
concept should be applied before determining the responsible Member
States in the framework of the Dublin Regulation.
This
would also have the effect that family members who have travelled
irregularly and are stranded in another member State are to be sent
back outside the EU territory where they have to apply for family
reunification. This would render family reunification more complicated
and lengthy and make the provisions on family unity in the Dublin
Regulation in many cases meaningless.
33. On 22 and 23 June 2017, the European Council explicitly made
the connection between its ambition to conclude migration deals
with third countries and the safe third country concept. While instructing
the negotiating ministers of Justice and Home Affairs, it agreed
that “in order to enhance co-operation with third countries and
prevent new crises, the 'safe third country' concept should be aligned
with the effective requirements arising from the Geneva Convention
and EU primary law, while respecting the competences of the EU and
the Member States under the Treaties. In this context, the European
Council calls for work on an EU list of safe third countries to
be taken forward … The European Council invites the Council to continue negotiations
on this basis and amend the legislative proposals as necessary,
with the active help of the Commission”.
The
call for amendment rather than adoption of the proposal suggests
that the member States were far from an agreement. The expressed
aim of the safe third country concept, to further co-operation with third
countries, seems to imply a broadening rather than a restriction
of its scope.
3.4. Non-refoulement
obligations should include effective remedies
34. So far, appeals against
refoulement have only been possible
against specific countries, as the European Union itself could only
be brought before the European Court of Human Rights if it finally
decided to ratify the Convention as was negotiated between the two
organisations for several years. At the same time, in February 2017
the European Court of Justice declared its lack of jurisdiction
to judge in the actions brought against the EU–Turkey Agreement
by two Pakistani nationals and one Afghan national, deciding that
the agreement was reached by heads of State and government and not
the institution itself.
35. Several allegations of violation of the principle of
non-refoulement by member States
under Article 4 of Protocol No. 4 to the European Convention on
Human Rights (ETS No. 46) have been processed by the European Court
of Human Rights, the most important so far being the 2012 case of
Hirsi Jamaa v. Italy, found to have
pushed refugees back to Libya. The judgment states that: “Whenever
the State through its agents operating outside its territory exercises
control and authority over an individual, and thus jurisdiction,
the State is under an obligation under Article 1 to secure to that
individual the rights and freedoms under Section I of the Convention
that are relevant to the situation of that individual.” So even
if there is no legal jurisdiction, the Court also recognises responsibility
in case of
de facto jurisdiction
through effective control. Another decision on violation of Protocol
No. 4 was pronounced against Italy and Greece in 2014 in the case
of
Sharifi and Others.
36. Spain has also been convicted of breaching the prohibition
of collective expulsion under Article 4 of Protocol No. 4 and Article
13 of the Convention by its border control at the fences of Melilla.
These
cases concern the pushback policy of Spain to Morocco. A group of
Sub-Saharan migrants managed to climb the three fences, but were
immediately apprehended by the Spanish civil guard and handed over
to the Moroccan authorities, without checking their identities or
enabling them to explain their circumstances. The Spanish authorities
argued that the events occurred outside their jurisdiction: as the
migrants had not passed the border crossing point in Melilla, they
had not entered Spanish territory. The European Court of Human Rights, however,
recalled its jurisdiction as established in the
Hirsi judgment, and made it clear
that member States cannot escape their responsibility while construing
their jurisdiction in a certain way. Member States are hence not
allowed to move their borders inwards in order to prevent asylum
seekers from making an asylum claim.
After the case was referred to the
Grand Chamber, the Council of Europe Commissioner for Human Rights submitted
his written observations, in which he confirmed the violence against
the migrants and the absence of any individual examination or right
to an effective remedy.
37. Spain avoids reception of asylum seekers at these borders
through bilateral co-operation with Morocco, which actively prevents
sub-Saharan migrants from approaching the fences at Melilla and
Ceuta. This co-operation is based on benefits offered by Spain,
including substantial financial support. The fact that the bilateral
agreements are not accompanied by guarantees of the treatment and
rights of refugees on Moroccan territory could lead to situations
where asylum seekers and refugees are not able to invoke their rights
under the European Convention on Human Rights or the Geneva Convention.
This is not an isolated case: more countries have entered into co-operation
on their borders which result in less protection for asylum seekers
and refugees, for instance the co-operation of Poland with Ukraine,
Hungary with Serbia and Bulgaria and Greece with Turkey.
38. The Hirsi judgment made
it clear that member States exercising effective control over migrants
must ensure that their acts do not lead to refoulement under
the European Convention on Human Rights. The tendency of member
States to delegate their action to third countries raises the question
of their responsibility in the case of a causal link between their
bilateral co-operation and a breach of the Convention, even if they have
not exercised direct physical control. Not to assume responsibility
would imply that member States can easily escape their obligations
under the Convention by simply letting third parties “do the job”.
Such an outcome would clearly undermine the effectiveness of the
Convention.
39. Another threat to the effective protection of rights under
the Convention is the shift from legally defined procedures with
formal commitments to the use of informal tools which enable practical
co-operation in migration control. Although these informal arrangements
are often not intended to create legal obligations under international
law, they do have serious implications for the distribution of responsibility
between States and migrants’ right to protection. The rights of
asylum seekers are inherently dependent upon the possibility to have
possible human rights violations assessed by a court, but without
any formal agreements in place, it becomes very difficult to establish
at the outset whether a State or organisation was engaged in extraterritorial jurisdiction
and whether it actually agreed with a third State to delegate certain
functions or acts.
40. A new case filed by the Global Legal Action Network (GLAN)
and the Association for Juridical Studies on Immigration (ASGI)
with the European Court of Human Rights at the beginning of May
2018 has yet to be registered, but will be interesting to follow
and illustrates the direct consequences of EU externalisation policies. It
concerns 17 survivors of a fatal incident in which a boat carrying
migrants found itself in distress off the coast of Libya: the applicants
include the surviving parents of two children who died in the incident.
The submission makes use of evidence compiled by Forensic Oceanography,
part of the Forensic Architecture agency based at Goldsmiths, University
of London, which produced a detailed reconstruction of the incident
and the policies that contributed to it.
3.5. The
use of development assistance for migration management
41. Making development assistance
for countries in need conditional upon the transfer of responsibilities
for asylum and migration management poses multiple problems. The
situation of the countries in question could deteriorate through
lack of sufficient capacity to manage the migrants arriving as part
of an agreement, which pressurises governance structures and creates
tension among national populations who perceive assistance as being
disproportionally allocated to migrants rather than to them. The
human rights of the migrants concerned may also deteriorate through
the lower standards afforded to them in the “external” context,
and they risk being put “out of sight and out of rights” on the
other side of EU borders. Finally, the EU itself could be found
guilty of mismanagement and of violating rights, of which many are
its own creation, if it retains responsibility for the situation
of these migrants.
42. Since the Trust Fund for Africa is established under the European
Development Fund, its contributions must comply with the conditions
set out in Regulation 2015/323 on the financial regulation applicable
to the 11th European Development Fund and the Cotonou Agreements.
Tensions with the ownership and co-management principles of the
Cotonou Agreement and the principle of mutual accountability for
budget support as stipulated in Regulation 2015/323 raise serious
issues as to whether some of the funded activities comply with these
conditions and with the Treaty’s legal basis on development co-operation.
Furthermore, questions can be raised about whether some of the proposed
activities are consistent with the primary objective of EU development
co-operation, namely the reduction and, in the long term, the eradication
of poverty (Article 208.1 TFEU).
43. It is unclear whether substantive obligations arise from the
agreements concluded under the Trust Fund. According to the Cotonou
Agreement, programmes and projects financed by the multi-annual
financial framework of co-operation are as a rule subject to financing
agreements drawn up by the Commission and the ACP State. But for
many projects no financing agreements have been concluded, which
undermines the mutual accountability principle entailed in the Cotonou
Agreement.
44. It is also currently unclear what “emergency” the Fund seeks
to address, and what consequences its emergency status has for long-term
development programmes. Since governance procedures for the Fund deviate
from the ordinary decision-making processes, they should be seen
as truly emergency-led instruments whose added value and consequences
should be very well justified.
4. Implementation
of externalisation policies – the cases of Turkey and Libya
45. Migrants mostly come to the
EU across the Mediterranean Sea, with the largest numbers of applicants for
asylum coming from Eritrea, Sudan and Somalia. Migrants, who transit
across Turkey mainly come from Syria, Afghanistan and Iraq.
Many of the transit countries have
a deficit in human rights standards.
46. The EU proposal for the Asylum Procedures Regulation, released
in July 2016 and still under negotiation as part of the Common European
Asylum System (CEAS) reform package, clearly reflects one of the
objectives of concluding migration deals with third countries: the
transfer of responsibility for refugees.
The EU especially seeks to harmonise,
but also to weaken, the criteria for qualifying a transit country
as a safe third country. Adoption of this regulation will further
pave the way to the adoption of the EU–Turkey model to other countries.
Political leaders have clearly shown an interest in concluding similar
agreements with Tunisia, but also other countries of interest, such
as Egypt, have their attention.
4.1. Turkey
– a model for future externalisation action?
47. In its
Resolution 2109 (2016) on the situation of refugees and migrants under the
EU–Turkey Agreement of 18 March 2016, the Assembly found that returns
of asylum seekers of any nationality to Turkey as a “safe third
country” was contrary to EU and/or international law, and that returns
of Syrian asylum seekers to Turkey as a “first country of asylum”
could be contrary to EU and/or international law. The Assembly referred
to the many obstacles for asylum seekers to have access to the procedure
due to the rules a lack of capacity and appropriate reception conditions,
the limited access for Syrian refugees to housing, primary education,
the labour market and a livelihood, and the allegations of deportation,
unlawful detention and bad detention conditions.
48. With the conclusion of the EU–Turkey Agreement, the debate
on the requirements for applying the safe third country concept
became ever more relevant, as Turkey retains its geographical limitation
to the Geneva Convention and doubts remain as to whether the temporary
protection status given to Syrians and the treatment of non-Syrians
meet the standards of the Geneva Convention.
49. During the implementation of the Agreement, organisations
continually reported about the precarious position of readmitted
refugees and asylum seekers in Turkey. With a view to immediate
deportation, non-Syrian migrants readmitted by Turkey from Greece
are sent to Turkish removal centres where they have very limited
or no access to lawyers, the UNHCR, NGOs, or to the asylum procedure.
The deal thus makes refugees and migrants end up in a vacuum: although
their access to asylum was very limited in Greece, the Turkish authorities
claim that they already had an opportunity to request asylum there.
According to the Turkish Government, Syrian refugees have access
to temporary protection after readmission from Greece. Researchers
however have reported that they are (first) transferred to de facto closed camps where they
are locked in cells and have very limited communication opportunities
and access to the outside world.
50. The expressed concerns not only relate to the readmitted refugees.
Syrian refugees receiving temporary protection in Turkey live in
extreme poverty, due to the combination of limited access to social
welfare systems and to the labour market, where a quota system for
Syrian refugees is applied and employers requesting a working permit
face long and expensive procedures. The total number of official
work permits being issued to Syrian refugees are not more than 15 000
to 30 000. Many of the refugees are exposed to exploitation on the informal
labour market, including a substantial number of Syrian children.
51. The state of emergency that applies since the coup attempt
of summer 2016 also has repercussions for refugees and asylum seekers,
as it has significantly reduced their safeguards against refoulement. The Turkish Executive
Decree 676/2016 has abolished the automatic suspensive effect of
an appeal against removal orders for individuals considered to constitute
a “threat to public order, security and health” or regarded as somehow
associated with “terrorist organisations”. Although the reason for
such labelling is not substantiated, administrative authorities
and courts do not question its validity. In such cases, removal
orders can be issued even when the person concerned is a recognised
refugee or a registered asylum seeker. The only instance able to
stop the deportation is the Constitutional Court.
52. The positive results of the EU–Turkey Agreement are that funding
has supported access to services for refugees in Turkey. Considerably
more children are able to attend school and health services have
been improved and more supporting programmes are offered to the
most vulnerable refugees. Furthermore, as at 30 May 2018, 13 862 Syrian
refugees had been resettled from Turkey and the capacities of Turkey’s
Ministry for Migration had been significantly strengthened.
53. However, the level of poverty of Syrian refugees, exploitation
on the labour market and housing market, as well as the high level
of child labour remain concerning. The economic breakdown is likely
to even worsen their situation and create tensions in a country
already struggling with socio-economic strains and political tensions.
Also gaps in protection remain, both for Syrian refugees and for
non-Syrian asylum seekers. This is partly related to the sheer number
of people hosted in Turkey. The highest numbers of migrants arriving
in the country were from Afghanistan: 175 000 arrived in 2017, and
for only the first five months of 2018 the figures already stood
at over 100 000. Some 14 000 Afghans had been returned to Afghanistan
in the framework of the readmission agreement with the country,
1 626 of whom had arrived from the Aegean islands. The IOM was co-operating
in these operations, and efforts were being made to absorb the backlog,
for example a second processing centre had just opened in Ankara.
54. The EU–Turkey Agreement not only had consequences in Turkey.
The Statement made refugees dependent on the deficient asylum and
reception system on the Greek Aegean islands. The former Commissioner
for Human Rights, Nils Muižnieks, has voiced the opinion that the
EU–Turkey deal was made redundant by the reinforcement of border
controls. According to the Commissioner, migrants and refugees would
no longer seek to cross into Greece from Turkey even if the migrant
deal ceased to exist, as migrants were aware of all the difficulties
they faced in reaching mainland Europe. So despite the fact that
Turkey is doing more than its fair share in hosting 3.6 million
Syrians, the EU is adding to this number through its failure to
take on its full responsibility.
55. My visit to Turkey on 31 May and 1 June 2018 allowed me to
experience this situation first-hand, and to better comprehend the
context in which the EU–Turkey Agreement is being implemented. I
was indeed impressed in many ways by Turkey’s efforts to create
structures to host so many refugees, in a context of relative internal
upheaval since the attempted coup, and of emerging and serious economic
difficulty. But the visit also confirmed the concerns expressed
in previous reports.
56. The Turkish authorities I spoke with were concerned and frustrated
that the Pathway for Syrians programme for resettling 72 000 migrants
was not functioning. The Ministry for Foreign Affairs regretted
the delays in awaited EU action as part of the agreement. Discussions
on visa liberalisation had just resumed after a two-year break.
During my visit, several representatives of the authorities stressed
the need for EU programmes to be more flexible in their implementation,
as sometimes the speed of response was not sufficient and the programmes
not fully adapted to the circumstances. The parliamentarians I met
in Ankara said that only €750 000 million of the pledged assistance
had been paid out so far.
57. My exchanges with the Ministry for EU Affairs reflected the
same frustration, despite the general statement that the agreement
had allowed the Adriatic route to be shut down and more regular
and safe migration to be ensured. With the “1 for 1” scheme and
the bilateral Greece-Turkey readmissions agreement, sea crossings
had been reduced, although irregular migration continued. Turkey’s
coastguard was assuming most of the burden of border control. The
hope was that the second phase of implementation of the agreement would
allow initial shortcomings to be corrected and good practices to
be propagated.
58. The IOM had a relatively optimistic outlook on the situation
and on the effects of the EU–Turkey Agreement, explaining that their
role involved advice and support for Turkey as well as direct implementation of
programmes. Integrated Border Management was becoming more efficient,
and support had been given to the Directorate General of Migration
Management (DGMM) as well as to the Ministries of Justice, of Labour and
of Family Affairs. Administrative and institutional capacities had
been built up, as had regional and international co-operation. Regional
returns centres, formerly operated by police services, were now
managed by the DGMM and reception conditions were much improved.
59. My visit to the Removal Centre was useful and informative.
The living conditions in the Kırklareli centre were very good, the
building large, spacious and clean and the staff informative, and
visibly on friendly terms with the migrants staying there. On a
daily basis, migrants had access to activities and fresh air. My
exchanges with migrants of diverse origin (Afghan, Eritrean, Iraqi,
Sierra Leonean) at the readmission centre in Kırklareli revealed
a general lack of knowledge among migrants about the possibilities
for asylum application or appeal against decisions, or indeed in
some cases an ignorance of the reasons for their detention. Leaflets
in different languages are distributed with information about the
possibility to ask for asylum. If a migrant doesn’t have any means
to pay for legal aid, he is referred to the bar association. However,
migrants in the removal centre and several NGOs informed us that
the expressed will to ask for asylum does not always lead to action
by the authorities and that legal aid is not always granted. So
although the asylum procedure looks appropriate on paper, the rules
do not always correspond to the practical situation. That leaves
non-Syrian protection seekers often in an insecure and precarious
situation.
60. Asylum procedures for non-Syrians still appear to be the most
problematic. These asylum seekers are sent to “satellite cities”
around the country, where they have to stay during the asylum procedure.
However, there are long waiting periods and migrants suffer from
poverty and lack of employment. If an asylum seeker leaves the satellite
city, the asylum claim will not be examined.
61. An exchange of views with Ms Gauri van Gulik, Director for
Europe of Amnesty International, at the Assembly’s April 2018 part-session
confirmed that the situation was still of the utmost concern, with
an increase in the practice of “forced voluntary returns” previously
observed in Greece, and the multiplication of “removal centres”.
My impression in Turkey was that there is an atmosphere of mistrust
between the authorities and the NGOs, which may impede the capacity
of the latter to assist migrants both legally and socially. Lawyers
still manage to provide assistance, with support in particular from
the Turkish Union of Bar Associations, which organised training
and worked on the quality of legal aid and funding. But the field
suffers from a shortage of qualified legal practitioners; this also
applies to immigration officers.
62. The EU–Turkey Agreement has also led to obstacles for the
entry of new refugees into Turkey. Soon after the Agreement was
concluded, the Turkish authorities closed the land border with Syria.
The 911-kilometre wall along Turkey’s borders with Syria was seen
by the authorities as a necessity. According to the DGMM, the “border
gate” was still open to vulnerable Syrians fleeing their country.
Other stakeholders however pointed out that only refugees in a life-threatening
situation are admitted to the Turkish territory since the closing
of the land border. Practically, this makes it very difficult or
even impossible for Syrian refugees to receive the protection to
which they are entitled in the context of Article 3 of the European
Convention on Human Rights and the Geneva Convention.
63. So-called “safe zones” were currently being tested to allow
Syrians to return to areas close to the Turkish border, in the proximity
of Gazientep for instance: 150 000 people had returned to Jerablus,
which appeared to be a viable solution, although how the people
concerned manage to live with little or no means of subsistence is
not clear. According to the Turkish authorities, these returns were
voluntary – other interlocutors stated that there were many forced
returns, not least by coercion to sign a “voluntary” return document,
often drafted in Turkish which migrants do not understand. These
documents were proposed on arrival. As these zones are established
on Syrian territory, it is questionable if this safety is durable
and if it really enables refugees to build a new future, given the
unstable and dangerous situation in Syria.
64. Although the resistance within the Greek asylum system has
reduced in favour of the EU position, the EU–Turkey Statement has
not been effective in numbers of readmission. According to the European Commission,
on 30 May 2018, 2 224 migrants had been returned from Greece to
Turkey since 18 March 2016. On 7 June, the Turkish Government suspended
the bilateral readmission agreement between Turkey and Greece following
Greece’s release of four military servicemen that Ankara wants extradited
in order to prosecute them for taking part in the 2016 coup attempt.
This analysis shows that if the safe third country concept is merely
applied because of political wishful thinking, asylum seekers and
refugees face difficulties in invoking their rights under the EU
Asylum acquis.
65. At the time of presenting this report, serious doubts persist
as to the possibility of assuring migrants sufficient protection
outside EU borders under this type of agreement. The application
in Turkey is already problematic due to the large number of refugees
and migrants it hosts. But still, the Turkey context is so unique that
the EU–Turkey Agreement cannot be exported as a model for externalisation,
even if future implementation involves stronger guarantees of protection
of migrants’ rights. One of the reasons is that Turkey is bound
by the European Convention on Human Rights, which does not apply
to many other partner countries. Furthermore, the accession process
obliges Turkey to adapt its asylum system to EU standards. There
is an Asylum Law and a system to process claims, although it is
in its infancy. Applying similar deals with other countries, in
absence of these circumstances, appears impossible or, at the very
least, will be at the cost of human rights.
4.2. Libya
– as yet uncertain results of EU resource mobilisation
66. The EU is providing assistance
in Libya's political transition towards a stable, functioning country
and is supporting United Nations-led mediation efforts. The importance
of inclusiveness of the political process and Libyan ownership is
emphasised in the EU’s communications on the subject. Through the
IOM and the UNHCR, the EU is providing humanitarian assistance to
respond to migration challenges and is supporting Libyan authorities
through its Common Security and Defence Policy (CSDP) missions and
operations, EUNAVFOR MED Operation Sophia and EUBAM Libya. The EU
is also working closely with the UN Support Mission in Libya (UNSMIL).
Funding currently stands at 120 million euros to support projects
in governance, health, youth and education, migration and protection,
as well as security. Action is also undertaken under the Emergency
Trust Fund for Africa, which includes giving “voluntary access to
humanitarian return and reintegration”.
67. The European Commission press release of 16 May already mentioned
refers to EU support in IOM action which has assisted over 6 185
people in voluntarily returning home from Libya so far in 2018,
and to the work of a joint African Union-EU-UN taskforce working
with the Libyan authorities to stop the systematic detention of
migrants, which has enabled the release from detention of over 1 000
refugees this year. The EU Trust Fund for Africa’s 147 programmes
had received a total amount of €2.59 billion so far for initiatives including
voluntary returns from Libya. But the European Commission itself
declared that the funding gap of around €1.2 billion risked crippling
efforts if not “addressed together by the EU and member States”.
68. Despite these programmes and the untiring work of humanitarian
organisations on the ground, the situation of migrants in Libya
remains extremely worrying, as can only be expected in the case
of a country emerging from a prolonged period of brutal dictatorship
and corruption. The humanitarian situation in Libya is a subject
for grave concern, as attested by the recent United Nations, IOM
and Amnesty International reports,
coupled with doubts as to whether
the perception of financial gain through foreign investment does
not in fact increase criminal activity within the territory and
at the southern borders. In this case it is extremely worrying that
the responsibility for the lives of migrants is being put by the
EU into the hands of representatives of authorities which do not
yet have the capacity, competence or real means or even intention
to provide them with assistance and protection, as well as proper
access to asylum procedures.
69. In February 2018, the UNHCR reported that since November 2017
it had evacuated over 1 000 highly vulnerable refugees from Libya
and was seeking durable solutions for them in third countries. Flights
departed from Tripoli for Niamey in Niger carrying 128 refugees,
and 150 refugees from Tripoli to Rome, bringing to 1 084 the total
number of refugees evacuated in the first three months of the UNHCR’s
operation. The UNHCR’s special Envoy for the Central Mediterranean,
Vincent Cochetel, reported on UNHCR plans to evacuate thousands
more refugees, with the co-operation of partners and the crucial
support of the Government of Niger, where 770 refugees had so far
been evacuated, including single mothers, families and unaccompanied
and separated children. In total, 312 refugees had been evacuated
directly to Italy.
70. During the preparation of this report, evacuations came to
a standstill as EU member States had not respected their pledges
to resettle the refugees (recognised as such at a UNHCR camp in
Niger), as is evident from the pitifully small numbers of resettlements.
As a result, the detention centres in Libya filled up again. A new
press release on 10 May 2018 stated that evacuations had been resumed
to Niamey for 132 people, citing a total of 1 474 evacuees as well
as 477 resettlements directly from Libya to third countries.
71. In this context of concern and of lack of reliable information,
in October 2017, the Chairperson of the Migration Committee wrote
to the Chairperson of the Italian Delegation to the Parliamentary
Assembly, Mr Michele Nicoletti, commending Italy’s humanitarian
action in favour of migrants, but expressing concern about the implications
of the adoption of a Code of Conduct for NGOs involved in search
and rescue operations in the Mediterranean. The Chairperson of the
committee highlighted the refusal to sign the Code by a majority of
NGOs, including Doctors Without Borders, because of the ban on access
to Libyan territorial waters and the requirement to give access
on board rescue ships to armed police officers, amongst others.
This letter voiced similar concerns to those voiced by the former
Commissioner for Human Rights, Niels Muižnieks, in a letter to the
Italian Minister of the Interior dated 28 September 2017.
72. The replies to the letters emphasised the resulting reduction
in the risks of accidents and drowning, and the establishment of
humanitarian corridors for Syrian refugees. This progress had been
achieved by increased co-operation with Libya (as well as with Niger),
including through capacity-building programmes, in particular training
in combating human trafficking, indicated as the “first cause of
the failure to respect human rights for thousands of people fleeing
war or destitution”.
The
Italian Minister stated that no Italian ships or any vessels co-operating
with the Italian coastguards had taken rescued migrants back to
Libya. Although this may be true, if the Libyan Coast Guard is paid
by Italy to take those migrants back to Libya, the Italian authorities
also have an indirect responsibility for what happens to those migrants
at sea and back on Libyan territory.
73. NGOs are increasingly concerned about the new role attributed
to the Libyan Coast Guard, which has resulted in increasing push-backs
and distressing circumstances during operations, for both the people rescued
and the NGO rescue teams. The grave violations of human rights repeatedly
reported against migrants in Libya appear to show that people rescued
cannot be returned and disembarked therein in compliance with international
human rights law, refugee law and maritime law. NGOs see the obligation
to hand over migrants to Libyan coastguards as a daily violation
of human rights.
74. According to them, the Libyan coast guard has intercepted
and pulled back migrants, in some instances causing deaths. NGOs
state that, in several incidents, EU military ships were present
in the vicinity but did not launch rescue operations, and the NGO
Sea-Watch claims that EU ships have assisted the Libyan coastguard in
carrying out some interceptions. Following these claims and for
the purposes of this report, I wrote to the Libyan coastguard, Frontex,
Eunavfor Med and Inmarsat to request detailed accounts of the incidents.
The replies were not conclusive and I intend to pursue the matter
in the context of follow-up to the report.
5. Conclusions
75. Externalising migration management
to non-EU member States is seen as a necessity in order to ease the
pressure on front-line EU countries, and as a solution to avoid
migrants risking their lives on long and dangerous journeys towards
Europe. But it remains a dangerous enterprise, as there is no guarantee
that the rights of migrants will be respected, and in some cases
that even basic human rights and dignity will be upheld. Indeed,
evidence suggests that these rights are too often compromised. In
order for external co-operation on migration to work, respect for
human rights must be guaranteed through impact assessments prior
to entering into co-operation, safeguards embedded in agreements
as well as clear provisions on responsibilities and effective remedies
for migrants affected by this co-operation.
76. The “emergency” nature of the EU’s Emergency Trust Fund for
Africa and the broad objective of “tackling the root causes of migration”
make it difficult to assess what EU policy objectives projects funded
under the EUTF aim to achieve and how these objectives fit into
the legislation covering EU development assistance. Furthermore,
the non-binding and political nature of the approved action in the
framework of the Fund makes the EU’s specific obligations and responsibilities
unclear.
77. The cost of capacity building in countries which need substantial
political, legal, economic and social assistance in attaining sufficient
capacity to manage migration should be carefully measured against
the potential cost of improving and reinforcing capacities in Europe
for reception and integration.
78. Externalisation of migration control and border management
cannot be carried out at the cost of the rights of migrants and
refugees, including their right to leave countries like Libya in
search of a better future. Levels of human rights protection need
to be defined and adequate monitoring mechanisms designed, including
provision for suspending co-operation if necessary. Resettlement
obligations of EU member States also require enforcement in order
to contribute to a fairer sharing of responsibility. In addition,
resources spent on reinforcing border security and capacity building
outside European member States should in no way detract from development
co-operation with these and other countries in need of assistance.
The detailed recommendations related to these issues are set out
in the draft resolution.
79. In my exchanges, it was made clear that the Turkish authorities
did not agree with the concept of “externalisation” of migration
management by the EU in the case of Turkey, stating that the EU–Turkey Agreement
of 18 March 2016 was based on mutual consensus and understanding.
The authorities point to the humanitarian aims of the agreement
– preventing the loss of lives at sea, crushing migrant smuggling
rings and replacing illegal migration with regular migration.
80. The European Union’s policy, which is likely to be adopted
by its third-country partners, does not seem to be well thought
through as regards its consequences for the long term and at the
global level. It may have regional consequences for free movement
arrangements (such as ECOWAS) and flexible practices of border crossings
for temporary protection and thus undermine the EU’s objective to
enhance regional co-operation and mobility. The accumulation of
the EU’s internal and external migration policy as analysed in this
report will inevitably lead to an increasing responsibility for
third countries and may thus affect their willingness to solve protection
needs in a regional co-operation framework. This impact runs counter
to the aim of the future Global Compact on Refugees to strengthen
international co-operation in order to ease pressures on the host
countries involved, create a more equitable sharing of responsibility
and foster sustainable solutions for refugees.