Formalising EU “migration deals”
An assessment of the characteristics of a formal agreement
implementing the Regional Disembarkation Arrangements
Subject: LLM thesis
Master: International and European Law Track: European Union Law
Student: Quinty de Nobel (10776435) Supervisor: Prof. Pieter Jan Kuijper Date: 26 July 2019
1
Index
Abstract ... 2
Abbreviations ... 3
Introduction ... 4
1. Cooperation with Third Countries within the Context of the EU’s External Migration Policy ... 7
1.1. The external dimension of EU migration policy: a recent phenomenon? ... 7
1.2. Regional Disembarkation Arrangements: the most recent generation of partnerships in the field of migration ... 9
1.3. The place of RDAs within the existing framework ... 13
1.4. Current state of affairs ... 13
2. Lessons Learned from Past Experiences: the EU-Turkey Statement and the Italy-Libya Memorandum of Understanding ... 15
2.1. The EU-Turkey Statement (2016) ... 15
2.1.1. Content and context of the Statement ... 15
2.1.2. Concerns ... 16
2.2. The Italy – Libya Memorandum of Understanding (2017) ... 19
2.2.1. Content and context of the MoU ... 19
2.2.2. Concerns ... 20
2.3. Lessons learned ... 21
2.3.1. Formal characteristics of the agreement ... 22
2.3.2. Asylum systems in partner countries ... 22
2.3.3. Guarantees of respect for international refugee law and human rights ... 23
3. Relevant international human rights and refugee law ... 25
3.1. Applicable international refugee law and human rights ... 25
3.2. Conditions for a safe third country qualification ... 26
3.3. Procedural safeguards of safe third country arrangements ... 32
4. EU competence division: who concludes the future RDA-agreement? ... 34
4.1. EU exclusive competence to conclude an international agreement ... 34
4.2. Substantive legal basis determination ... 35
4.3. An exclusive EU external competence in the field of readmission and asylum?... 37
Conclusion ... 39
2
Abstract
In this thesis, the option of a formalised “migration deal” is examined, using the example of the European Union’s most recent plans for Regional Disembarkation Arrangements. The most well-known example of such a deal is the EU-Turkey Statement, whereby the European Union lends Turkey various forms of support in return for their assistance in managing migratory flows to Europe. The proposal for Regional Disembarkation Arrangements pursue comparable objectives, drawing on similar tools for cooperation with third countries. The assessment of the characteristics of a formal agreement implementing such arrangements focuses on four aspects: its content, lessons that can be learned from existing migration deals, compatibility with international human rights and refugee law, and finally the legal basis and parties to the agreement.
3
Abbreviations
AU African Union
CEAS Common European Asylum System
CJEU Court of Justice of the European Union
ECHR European Convention on Human Rights
ECtHR European Court of Human Rights
EU European Union
EU RDA-proposal The EU proposals to implement the RDAs
GAMM Global Approach to Migration and Mobility
GC General Court
IOM International Organization for Migration
JHA Justice and Home Affairs
RDAs Regional Disembarkation Arrangements
RDA-agreement Agreement implementing the RDAs
RDA-proposal IOM and UNHCR proposal for RDAs
Refugee Convention Convention Relating to the Status of Refugees
SAR Search and Rescue
TEU Treaty on European Union
TFEU Treaty on the Functioning of the European Union
4
Introduction
The most recent migrant crisis has demonstrated the shortcomings of the European Union’s (EU) migration policy. The high numbers of migrants and refugees that arrived in 2015 and 2016 put pressure on Europe’s asylum systems, making it clear that the EU was not equipped to deal with such migratory flows.1 In response, the European Commission published a package of proposals aimed at the reform of the Common European Asylum System (CEAS).2 The package included a revision of existing instruments, while also foreseeing the exploration of possibilities in the field of external relations. This is referred to as the externalisation or the outsourcing of migration policy, involving greater engagement with third countries in order to solve EU migration policy problems.3
The European Council was quick to endorse these plans and specifically committed itself to the strengthening of the EU’s external migration policy in the Malta Declaration of 2017.4 In the meantime, the Member States had already stepped up their cooperation with Libya and Turkey in order to stem migratory flows passing through those territories and subsequently entering the EU. As will be shown in chapter 2, this cooperation had an informal character, lacking a solid basis in the form of an international agreement. The parties agreed on action plans, common agendas and joint declarations, but no formal agreements were concluded. It is therefore not surprising that the EU-Turkey Statement is referred to as a “migration deal”.
On 28 June 2018, the European Council expressed its wish to continue down this informal road by announcing the most recent tool to be added to the EU’s external migration policy toolbox, the Regional Disembarkation Arrangements (RDAs), thereby implementing informal arrangements on an even larger scale.5 As the EU is struggling with the task of internal reform of its asylum system, the European Council shifts the attention to decreasing the flow of migrants entering Europe by working with third countries in an effort to find an alternative way to end the humanitarian crisis that has been unfolding itself on the Mediterranean.6
1 Commission, ‘Irregular Migration & Return’ (European Commission website, 4 March 2019)
<https://ec.europa.eu/home-affairs/what-we-do/policies/irregular-migration-return-policy_en> accessed 4 March 2019
2 Commission, ‘How the Future Asylum Reform Will Provide Solidarity and Address Secondary Movements’
(factsheet, June 2018)
3 Christina Boswell, ‘The 'external dimension' of EU immigration and asylum policy’ (2003) 79 International
Affairs 619
4 European Council, ‘Malta Declaration by the members of the European Council on the external aspects of
migration: addressing the Central Mediterranean route’ (statement, 3 February 2017)
5 European Council, Conclusions (statement, 28 June 2018) no. 5
6 Francesco Maiani, ‘“Regional Disembarkation Platforms” and “Controlled Centres”: Lifting The Drawbridge,
5 However, apart from outsourcing its own tasks, the EU does not even seem to want to do so on a formal basis. No mention is made of what shape the arrangements with future partner countries should have. In an effort to fill this gap, this thesis explores how the EU can formally implement the RDAs. The research question is as follows:
“What are the characteristics of a formal agreement implementing the Regional Disembarkation Arrangements?”
The underlying assumption of this question is of course that a formalised approach is preferable to an informal one, at least from the perspective of the migrants and refugees that are going to be the subject of such arrangements. On top of the ignorance for democratic control by getting around procedural arrangements, informal arrangements lead to reduced (or even total absence) of judicial control. This goes to the detriment of accountability and responsibility taking, and eventually to the detriment of the rule of law.7
While the current EU strategy involves the engagement in many different informal arrangements, this thesis exclusively takes the RDAs as an example in order to examine what a formal “migration deal” should look like. It must be noted that the RDAs can have any kind of migrant as their subject, as not all migrants risking the journey across the Mediterranean fit the same profile. UNHCR chooses to qualify these mixed groups of persons as “migrants and refugees”, maintaining that this ensures that it is recognised that all persons have human rights while also emphasising that refugees have special needs and additional rights.8 It is indeed important to underline the fact that refugees may find themselves amongst the migrant groups. This is because refugees do not become refugees by successfully completing an asylum procedure, they are simply recognised as such. This is important, as it is the reason for which asylum-seekers enjoy additional protection. 9 After all, they may be refugees. For these reasons, the UNHCR terminology is used throughout this thesis.
September 2018) <http://eumigrationlawblog.eu/regional-disembarkation-platforms-and-controlled-centres-lifting-the-drawbridge-reaching-out-across-the-mediterranean-or-going-nowhere/> accessed 6 March 2019
7 See Ramses A. Wessel, ‘‘Soft’ International Agreements in EU External Relations: Pragmatism over Principles?’
(ECPR SGEU Conference, Paris, June 2018) for a discussion of the characteristics and the use of soft law
8 UNHCR, ‘Refugees And Migrants – Frequently Asked Questions (Faqs)’ (UNHCR, 2018)
<https://www.unhcr.org/news/latest/2016/3/56e95c676/refugees-migrants-frequently-asked-questions-faqs.html> accessed 14 July 2019
9 The definition used in this thesis is the definition in Art. 1, Convention Relating to the Status of Refugees [1951]
189 UNTS 137 (Refugee Convention): “any person who […] owing to well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality and is unable or, owing to such fear, is unwilling to avail himself of the protection of that country; or who, not having a nationality and being outside the country of his former habitual residence as a result of such events, is unable or, owing to such
6 As has become clear from the above, this research is mainly reform-oriented. Even though it is based on a doctrinal approach identifying and analysing existing law, it does so in order to recommend change and it elaborates on what that change should entail. According to Pearce, Campbell and Harding, reform-oriented research is research which “intensively evaluates the adequacy of existing rules and which recommends changes to any rules found wanting”.10 After establishing the existing framework of law in the first chapter, this thesis
indeed evaluates that framework in order to recommend certain changes. All in all, the essence of this thesis amounts to an examination of what future law should look like. Those chapters may also contain some normative elements, as the EU’s humanitarian obligations come into play.
In order to answer the research question, four aspects of a future agreement implementing the RDAs (RDA-agreement) are examined. The first chapter sets out the EU’s external migration policy, which is the framework under which the RDAs are to be developed. It also examines how the RDAs are to fit into that framework. This assessment is largely based on existing instruments and EU policy documents. The second chapter evaluates two examples that largely resemble the plans for RDAs, the EU-Turkey Statement and the Italy-Libya Memorandum of Understanding. Its aim is to determine which improvements can be made to future migration deals with third countries, because it is considered important to look at the past in order to determine a better way forward. Chapter three examines an aspect that cannot be ignored in view of the sensitivity of this policy area, namely how the parties to a future RDA-agreement are to respect international human rights and refugee law. Finally, the fourth chapter considers the classic, formal aspects of a future RDA-agreement. A competence assessment is carried out to determine the substantive legal basis and the parties to the agreement.
10 D. Pearce, E. Campbell & D. Harding, Australian Law Schools: A Discipline Assessment for the Commonwealth Tertiary Education Commission (1987) 17
7
1. Cooperation with Third Countries within the Context of the EU’s External
Migration Policy
The RDAs form part of the EU’s most recent plans to reform its migration policy, more specifically the CEAS. The Arrangements belong to the EU’s external migration policy, which mainly consists of cooperation with third countries. Although the EU’s external migration policy consists of many different instruments, this chapter focuses exclusively on international agreements and similar arrangements concluded in this area as these are the most relevant to the research question.11
Part one of this chapter discusses the formal and informal agreements that the EU has entered into so far within the ambit of its external migration policy. Part two sets out the plans for RDAs. Part three assesses the place of RDAs within the framework of existing agreements. The fourth and final part establishes where the EU stands in the process of concluding the RDAs and which obstacles it has already ran into.
1.1. The external dimension of EU migration policy: a recent phenomenon?
The European Council called for an external dimension to the EU’s migration policy by way of partnerships with countries of origin and transit as early as 1999.12 Boswell distinguishes two types of approach when it comes to the implementation of these partnerships. The first concerns what she calls an externalisation of traditional tools of migration control. This entails working with countries of origin and transit on border controls, illegal entry, migrant smuggling and the readmission of illegally staying migrants. The second approach is preventive, as it aims to prevent people from opting for the EU as a migratory destination. It includes addressing the root causes of migration as well as providing solutions for migrants closer to home.13 The EU has developed its external migration policy along these lines and the agreements that have been concluded so far can be placed in one of these categories. Agreements concluded in the context of the EU’s external migration policy vary from rather informal joint declarations to formal international agreements and treaties.
Under the umbrella of the externalisation approach, accession of future Member States has been made conditional upon the incorporation of the Schengen acquis into their national
11 See Nikolai Atanassov and Anja Radjenovic, ‘EU asylum, borders and external cooperation on migration: recent
developments’ (2018) European Parliamentary Research Service in-depth analysis PE 625.194 for a complete overview of instruments in the EU’s external migration policy.
12 European Council, Presidency Conclusions (Statement, Tampere, 1999) no. 11 13 Boswell (n3) 620
8 legislation.14 The instrument to be used here is an accession treaty, which is a treaty to be concluded and ratified by the acceding country as well as the Member States. It contains the conditions and arrangements regarding accession.15 Article 4, Part I, of Croatia’s accession treaty provides a good example of this kind of obligation: “The provisions of the Schengen acquis […] shall be binding on, and applicable in, Croatia from the date of accession”.16
Next to border control measures, this approach involves efforts to facilitate the return of illegal migrants, who entered or stayed in the EU irregularly, by way of readmission agreements with non-EU countries.17 The aim of these agreements is to ensure the cooperation of third countries when it comes to readmitting their nationals or persons who transited through the territory of that country. So far, 17 readmission agreements have been concluded with partner countries including amongst others Russia, Turkey and the Western Balkan countries. Readmission clauses may also be incorporated into broader partnership and association agreements.18 These agreements are usually concluded within the context of the European
Neighbourhood Policy and cover a broad range of areas with a view to bringing the third country’s policies closer to the EU’s policies.19 Interestingly, the Council also authorised the
Commission to negotiate readmission agreements with Morocco, Algeria, Tunisia and Nigeria. No agreements have been concluded yet. In view of the recent RDA-proposal, there may be some overlap in terms of content.
Regarding the preventive approaches, the EU attempts to promote reception in the region and to address the root causes of migration through development assistance and foreign policy instruments.20 From the 2015 Agenda on Migration it becomes clear that the EU wants to make use of partnerships with countries of origin and transit in the form of bilateral and regional cooperation frameworks.21 These frameworks were first pursued as a part of the Global Approach to Migration and Mobility (GAMM), which was to be considered the overarching framework of the EU external migration policy. In 2016, the Mobility Partnership Facility was
14 Boswell (n3) 622
15 Commission, ‘Treaty Of Accession – European Neighbourhood Policy’ (European Commission website, 2016)
<https://ec.europa.eu/neighbourhood-enlargement/policy/glossary/terms/treaty-accession_en> accessed 8 May 2019
16 Treaty between the Member States of the European Union and the Republic of Croatia concerning the accession
of the Republic of Croatia to the European Union [2012] OJ L112
17 Boswell (n3) 622
18 An inventory of agreements containing a readmission clause can be accessed via the EEAS Treaties Office
Database via <http://ec.europa.eu/world/agreements/ClauseTreatiesPDFGeneratorAction.do?clauseID=27>
19 Pieter Jan Kuijper et al., The Law of EU External Relations : Cases, Materials, and Commentary on the EU as an International Legal Actor. 2nd ed. (Oxford: Oxford University Press, 2015) 568
20 Boswell (n3) 624
9 launched to revitalise GAMM by further operationalising it.22 The programme mainly consists of non-legal instruments such as political declarations, common agendas and development funds, but visa facilitation and readmission agreements may also be deployed. Recent examples include the Rabat and Budapest processes.23
To conclude, it is clear that formal agreements and informal arrangements with third countries traditionally play a role in the EU’s migration policy. Next to non-legal instruments, agreements with third countries form the backbone of the EU’s external migration policy. After a brief description of the EU’s most recent plans for partnerships, part 1.3 attempts to place the RDAs within this existing framework of agreements concluded with third countries.
1.2. Regional Disembarkation Arrangements: the most recent generation of partnerships in the field of migration
In 2013, CEAS was created, covering all legal aspects of the asylum process. Not much later, in 2015, the weaknesses of the same legal framework were exposed. Substantial losses of life on the Mediterranean sea were reported as uncontrolled flows of migrants tried to reach EU territory.24 The European Commission responded by announcing a European Agenda on Migration in the same year. The aim was to “build up a coherent and comprehensive approach” to migration, prioritising the duty to protect the migrants embarking on dangerous journeys across the Mediterranean.25 Plans did not only include internal reform, but also a reinforcement of the external dimension of migration policy. The communication also emphasises the Union’s aspirations to cooperate more intensively with Turkey. This would eventually lead to the EU-Turkey Statement, the EU’s first “migration deal”, to be discussed in Chapter 2.
With numbers of arriving migrants rising to significant heights again, EU heads of government convened during a migration summit in the summer of 2018. In order to prevent a repetition of the situation in 2015, the members of the European Council committed to tackling illegal migration and smuggling, by containing existing as well as emerging migration routes. The assumption was made that tackling the smugglers would equate to the “prevention of tragic loss of life”.26 In order to execute this policy, cooperation was necessary with third countries
finding themselves on those migration routes. The Member States had already committed
22 Commission, ‘Mobility Partnership Facility (MPF) - Migration And Home Affairs’ (European Commission website, 2019)
<https://ec.europa.eu/home-affairs/what-we-do/policies/international-affairs/global-approach-to-migration/mobility-partnership-facility_en> accessed 10 July 2019
23 Commission, ‘The Global Approach to Migration and Mobility’ COM (2011) 743 final 24 “Uncontrolled” is the word the European Council uses in its conclusions of 28 June 2018. 25 Commission, ‘A European Agenda on Migration’ (n21) 2
10 themselves to the reinforcement of the EU’s external migration policy in the Malta Declaration of February 2017.27 This time, a “new approach” was added to the EU’s external migration policy toolbox, namely the RDAs. The Commission received the task of further exploring this approach sooner rather than later.
The concept of RDAs was not invented by the EU, but was instead launched by the United Nations High Commissioner for Refugees (UNHCR) in cooperation with the International Organization on Migration (IOM). Their proposal was published on 27 June 2018, one day before the European Council summit.28 The main aim of the proposal is to delink disembarkation from the subsequent processing with a view to ensure greater respect for human rights while enhancing shared responsibility for migrants and refugees rescued-at-sea in the Mediterranean region. Although the proposal addresses the immediate aftermath of search and rescue (SAR) missions, it also urges the States involved to give renewed attention to the drivers of migration as well as to strive to open “safe and regular pathways” for migrants. Finally, the drafting parties submit that the proposal “relies on operational arrangements which would need to be sought and formalised through a set of understandings among all concerned States”, the shape of which is examined in subsequent chapters of this thesis.
The proposed RDAs consist of five steps. The first consists of disembarkation of persons recovered during SAR events to Maritime Rescue Coordination Centres operated by participating States. The second step involves transportation to reception centres, where basic services would be provided. Finally, the persons’ applications would be assessed and they would be categorised as refugees, persons with specific needs or illegal migrants. Refugees disembarked within the EU are to be processed in accordance with the applicable EU law. For refugees disembarked outside the EU, solutions should include third country resettlement and humanitarian admission. While this thesis explores the characteristics of an RDA-agreement, these solutions would naturally require the conclusion of more international agreements. The middle category would include amongst others victims of trafficking or unaccompanied children, and would require tailor-made responses. The latter category would need to be returned to the country of origin. Although this is not made explicit in the proposal, use would then need to be made of readmission agreements as described in section one of this chapter.
27 European Council, ‘Malta Declaration by the members of the European Council on the external aspects of
migration: addressing the Central Mediterranean route’ (n4)
28 UNHCR and IOM, ‘Proposal for a regional cooperative arrangement ensuring predictable disembarkation and
11 Initially, the Commission indeed responded swiftly by publishing a non-paper accompanied by a factsheet as early as July, further exploring the concept of RDAs. Simultaneously, a non-paper and a factsheet were disclosed regarding “controlled centres” in the EU to complement the implementation of the RDAs on the European side.29 According to the Commission, the concepts of controlled centres and RDAs are “fully interlinked” and to be developed simultaneously.30
First, it needs to be established what the EU wants by joining UNHCR and IOM in engaging in RDAs. The EU’s objectives can be divided into two categories. The first category concerns the quick disembarkation of people rescued at sea in line with international law. This includes respect for their human rights, especially the principle of non-refoulement prohibiting the return of refugees to territory where their life or freedom is endangered.31 The second category concerns the post-disembarkation process. Here, the EU prioritises shared responsibility as well as regional solidarity between the EU and third countries surrounding the Mediterranean. Strikingly, the strengthening of intra-EU solidarity is not mentioned in this respect even though IOM and UNHCR stressed in their proposal that this is to be a crucial aspect of RDAs. This category also involves avoiding the creation of pull factors by clearly communicating that disembarked persons in need of international protection will not automatically be resettled to the EU. It wants to encourage resettlement in the region. In short, the EU’s aims with regard to RDAs seem to mainly entail a burden shift to third countries and to discourage migrants from opting to travel to the EU.32 Independently of the desirability of this kind of policy, which may at least be considered questionable, 33 this thesis looks at its legal feasibility instead.
Second, the EU’s preliminary thoughts on implementation need to be examined. Several ideas have been circulated relating to different parts of the process. When it comes to the rescues at sea, the EU plans to support third countries (politically, operationally and financially) to
29 Persons disembarked in the EU are brought to these centres in order to have their application processed. The
aim is to quickly distinguish between persons in need of international protection and those who are not, so that the latter can be speedily returned. From the non-paper it appears that the controlled centres would be the EU-version of reception centres, although it is unclear why the operate under another name. Commission, ‘Non-paper on “controlled centres” in the EU’ (non-paper, July 2018)
30 Commission, ‘Non-paper on regional disembarkation arrangements’ (non-paper, July 2018) 31 Refugee Convention, Article 33
32 See also Commission, ‘Migration: Regional Disembarkation Arrangements’ (factsheet, July 2018). The EU
primarily envisages disembarkation in third countries, rather than in Member States.
33 See a.o. European Council on Refugees and Exiles, ‘Asylum at the European Council 2018: Outsourcing or
Reform?’ (2018) Policy Paper no. 4 < https://www.ecre.org/ecre-policy-paper-asylum-at-the-european-council-2018-outsourcing-or-reform/> accessed 10 April 2019
12 enable them to play a bigger role in SAR events. In terms of the actual Arrangements, the EU wants to work with interested countries on the basis of existing partnerships as well as offering them tailor-made support.34 The Commission’s non-paper even speaks of the option of using the International Maritime Organisation as a platform for the bilateral or multilateral development of procedures and arrangements. Regarding the reception centres, the EU plans to develop the previously mentioned concept of controlled centres. The notion would largely be based on the voluntary EU resettlement scheme. This is an internal EU issue, which does not require the conclusion of an international agreement. However, the transfer of persons in need of international protection from third countries to the EU would require an agreement to be concluded. This would be a part of the Arrangements. Lastly, as to the processing of the disembarked persons, the EU internally adheres to its own rules. The institutions do not seem to envisage a role for themselves with regard to the processing in third countries. In fact, the Commission encourages UNHCR and IOM to support those countries in finding a quick solution for each migrant in full respect of international law. Therefore, an agreement should arguably include provisions specifying which definitions, standards and conditions are to be observed.
Below a short recapitulation of the RDAs and the necessary action to be undertaken.
Parties involved EU and Member States, UNCHR, IOM, third countries (of origin as well as transit) Intended subjects /
beneficiaries
Persons rescued at sea, to be categorised into refugees, persons with specific needs or illegal migrants
Areas of law involved International law (adherence to existing (humanitarian) commitments), EU law (applicable to persons disembarked in the EU), national law of all states involved (either to give effect to EU law provisions on the processing of migrants or to give effect to RDAs in third countries; applicable to persons disembarked in third countries) Topics to be included Provisions on disembarkation of migrants and refugees after SAR-missions, provisions on the resettlement of persons in need of international protection, provisions on the rules to be respected when receiving and processing migrants and refugees
Agreements to be concluded
Existing agreements containing relevant commitments and human rights need to be respected, readmission agreements in order to be able to return illegal migrants need to be concluded, agreements between the EU or its Member States and the third countries surrounding the Mediterranean in order to implement the RDAs need to be concluded (especially responsibility allocation), agreements on resettlement to the EU as well as the other way around (may be included in the RDA-agreement itself) Figure 1: recapitulation of the plans for RDAs
13
1.3. The place of RDAs within the existing framework
In order to situate the RDAs in the existing framework of agreements with third countries, it first has to be established whether they fall under the classic externalisation approach or the preventive approach. The RDAs primarily appear to externalise border control instruments to third countries, aiming to strengthen their capacities in order to tackle illegal migration. Moreover, readmissions play a significant role in the policy plans.
Simultaneously, solutions are pursued in the preventive sphere. RDAs offer alternatives for migrants and refugees that are closer to home and that would terminate the need to embark on dangerous journeys across the Mediterranean. However, the main characteristic of the preventive approach, namely the focus on addressing the root causes of migration, does not play a material role. Commentators have pointed out that this is also the weakness of the plans for RDAs. Due to its lack of an intention to help build up solid asylum systems and a short-term focus on security, the opportunity to establish a real, long-term partnership that genuinely takes into account third countries’ interests is missed.35 Consequently, the RDAs principally
implement the externalisation approach.
Agreements discussed above include accession treaties, readmission agreements and association agreements. The accession treaty can be ruled out immediately, as the third countries involved are not EU candidate countries. Accordingly, the remaining options are the conclusion of readmission agreements or association agreements, adding a protocol on the RDAs and including a readmission clause. Given the fact that the EU wants to offer third countries “tailor-made incentive packages” in order to stimulate cooperation, the agreement would have to be concluded bilaterally with each third country involved. Analyses in the next chapters set out further the possibilities as well as advantages and disadvantages with regard to both options for agreements.
1.4. Current state of affairs
To conclude this chapter, some words need to be dedicated to the current state of affairs, thereby emphasising again the need to dedicate research to the legal options that exist in relation to creative ideas for partnership agreements. According to the timeline set out in the aforementioned Commission factsheet, the EU should now be in the stage of engaging with third countries based on a joint EU approach. A leaked working paper issued by the Council
35 Caritas Europa, ‘Analysis of EU proposals for disembarkation mechanisms’ (position paper, 17 October 2018)
14 presidency at the time confirms that Member States generally supported the proposals made by the Commission in its non-paper. Additionally, they agreed that the best way forward was to undertake an “initial phase of outreach towards potential partner countries”, offering them “tailor-made support and incentive packages”. The choice of partners is to be based on their human rights situations, existing legal frameworks and their administrative capacity.36
The elaborated common approach unfortunately cannot be accessed by the public as that could negatively impact the relationship between the negotiating parties.37 Regrettably, according to public sources, this already seems to be the case. The African Union (AU) was not charmed by the EU’s strategy of approaching individual African countries with big bags of money, nor by the plans themselves. Concerns exist with regard to the collection of biometric data by international organisations, the disembarkation centres in reality resembling detention centres and the danger of them turning into modern slavery markets. Moreover, the AU is not convinced that the EU has a clear view of what the concept of RDAs should entail.38
Dutch parliamentary papers dating from March of this year confirm the impression of an impasse in negotiations. 39 The State Secretary for Justice and Security Mark Harbers briefed
the Parliament following the publication of the agenda of the Justice and Home Affairs (JHA) Council of that month. He confirmed that up to March, there had not been taken any real initiatives to engage with North African countries. In fact, that JHA Council would be the first to discuss RDAs since they had been proposed in June 2018. Priority would be given to discussions with Tunisia, Libya and Morocco. More importantly, Harbers encourages the future Commission, to be installed after the European Parliament elections of May 2019, to put an end to the bilateral forms of cooperation and to step up its efforts to make sure that the EU operates as a whole when it comes to readmission agreements. Consequently, in terms of timing, a new summer will have arrived in which migratory flows may rise again and at least a year will have passed since the publication of the proposal without any concrete steps having been taken.
36 Council Presidency, ‘Presidency Non-Paper on Regional Disembarkation Arrangements’ (non-paper, 17
September 2018)
37 Council General Secretariat, Response to a request for access to documents dated 8 April 2019 (letter, 3 May
2019)
38 The Guardian wrote an article on the matter after obtaining a leaked position paper from the AU. Daniel Boffey,
‘African Union Seeks To Kill EU Plan To Process Migrants In Africa’ The Guardian (2019) <https://www.theguardian.com/world/2019/feb/24/african-union-seeks-to-kill-eu-plan-to-process-migrants-in-africa> accessed 5 May 2019
15
2. Lessons Learned from Past Experiences: the EU-Turkey Statement and
the Italy-Libya Memorandum of Understanding
This chapter discusses agreements strongly resembling the RDAs that have been concluded so far. The first concerns a bilateral agreement between all Member States and a third country, namely the EU-Turkey Statement. The second concerns a bilateral agreement between an individual Member State and a third country, namely the Memorandum of Understanding between Italy and Libya (MoU).
2.1. The EU-Turkey Statement (2016) 2.1.1. Content and context of the Statement
On 18 March 2016, the EU heads of state and the Turkish government reached an agreement, often referred to as the “EU-Turkey Deal” but officially called the “EU-Turkey Statement”.40
The agreement was reached after several meetings held in response to the sharp increase in migratory flows to Europe at the time. The objectives of the Statement resemble those of the RDAs to be concluded, namely to stem illegal migratory flows and to break smugglers’ business models on the one hand and on the other hand to strengthen relations between the EU and the partner country in question.41
The Statement consists of several action points, the most important of which is the return to Turkey of persons who travelled via that country to the Greek islands as of 20 March 2016. Asylum applications are assessed individually in hotspots in accordance with applicable EU law, so that only migrants whose application has been found inadmissible are returned to Turkey. In exchange for this “service”, the EU takes back one Syrian from Turkey for every Syrian returned to Turkey within the framework of a resettlement scheme. Additionally, Turkey receives 3 billion euro on top of previously granted EU funds in order to improve conditions for refugees.42
The EU-Turkey Statement largely resembles the planned RDAs, so that its weaknesses can easily be transposed to future agreements that implement RDAs. Similarities include first of all the processing of asylum applications in both the EU and the partner country with the
40 The legal nature of the EU-Turkey Statement remains unclear. In the case of NF v European Council (n44), the
CJEU had the opportunity to provide to clarify this. Unfortunately, it did not make use of this opportunity and instead referred to the Statement as “the measure”. The ECtHR referred to the Statement as an agreement in J.R.
and Others (n52, para 7), without specifying its legal effects. The ECtHR terminology is used in this thesis. 41 European Council, ‘Meeting Page – European Council, 17-18/03/16’ (European Council website, 11 December
2018) <https://www.consilium.europa.eu/en/meetings/european-council/2016/03/17-18/> accessed 12 May 2019
16 option of resettlements, although the 1:1 ratio is not copied. Furthermore, the processing takes place in controlled centres, which appears to be no more than a new name for the old hotspots. Finally, the partner country receives money to improve reception conditions for refugees.
2.1.2. Concerns
First of all, a distinction should be made between practical or policy concerns on the one hand, and legal concerns on the other. While the former relates to the effectiveness and implementation of the Statement, the latter relates to its legality (in view of e.g. asylum rules) and other legal aspects (such as legal status). A recent report published by the University of Utrecht extensively discusses the policy side of the EU-Turkey Statement, questioning its effectiveness based on the results it generated in 2017.43 As could have been expected, the Statement is not found to be a suitable policy solution to the problem it aims to address. Although numbers had dropped, migrants were still arriving on the Greek islands assisted by smugglers, the percentage of deaths at sea increased and the conditions for migrants and refugees in Turkey had not improved. Nevertheless, this clearly has not discouraged European politicians from pursuing similar ambitions with other partner countries. Therefore, the policy side will be left as is and the legal aspects will be explored further.
One of the most important points of concern, if not the most important, is the lack of legal reviewability by the Courts of the European Union caused by the EU-Turkey Statement’s unclear legal status under EU law. This was demonstrated by the recent case of NF v European
Council.44 Asylum-seekers who had travelled to Greece via Turkey and were therefore subject to the arrangements under the EU-Turkey Statement sought to have that agreement annulled. Being of Pakistani nationality, they feared expulsion to Pakistan by the Turkish authorities.45
Unfortunately for the applicants, the General Court (GC) concluded that it lacked jurisdiction to review the EU-Turkey Statement. This was recently confirmed by the Court of Justice of the EU (CJEU).46 The European Council had argued that the EU-Turkey Statement did not constitute an international agreement nor had it been concluded by the institution. Instead it was held that the EU heads of government, not in their capacity of members of the European Council, had participated in the meetings leading up to the publication of the EU-Turkey Statement. Since the Statement therefore did not constitute a measure adopted by an EU
43 Ilse van Liempt et al., ‘Evidence-based assessment of migration deals: the case of the EU-Turkey Statement’
(2017) University of Utrecht 20171221-Final Report-WOTRO
44 T-192/16 NF v European Council [2017] ECLI:EU:T:2017:128 45 Not an unfounded fear according to Ilse van Liempt (n43) 23-24 46 C-208/17 P NF v European Council [2018] ECLI:EU:C:2018:705
17 institution, judicial review by the GC or CJEU was not possible.47 The issue regarding the legal status of the agreement under EU law was never addressed, as it was not considered relevant anymore. This is unfortunate, but regardless of the EU-Turkey Statement’s legal nature (an agreement that produces legal effects or merely a press release?), human rights would have to be complied with.48
This case also raises the question of whether a situation in which the European Council concludes an international agreement is even possible. Art. 15 TEU clearly states that this institution does not exercise legislative functions, while Art. 218 TFEU indicates that international agreements are to be concluded in accordance with a legislative procedure. Therefore, the Treaties do not seem to allow the European Council to conclude international agreement. Interestingly, Wessel appears to suggest that the European Council does not envisage the conclusion of international agreements at all. Instead, it uses its weight for the conclusion of what he calls ‘soft’ agreements, thereby circumventing (democratic) legislative procedure and legal review by the CJEU.49 With regard to the EU-Turkey Statement, legal
review was avoided in two ways: by opting for a soft instrument and by choosing not to operate as an EU institution, so that the conditions laid down in Art. 263 TFEU could not be met.50
Moreover, this judgement raises the question of whether European leaders should be allowed to escape accountability and responsibility in this way. The EU is involved in monitoring, funding and implementing the agreement, should that not automatically trigger its responsibility regarding action taken under the auspices of the agreement?51 There is of course another European court that is relevant with regard to legal review of the agreement. One year ago, the European Court of Human Rights (ECtHR) was called upon to review the agreement and it assumed jurisdiction. However, it did not show itself very strict on the drafters of the agreement in this case, which involved detention conditions in Greek hotspots where migrants were detained before their transfer to Turkey.52 Anyhow, given the EU’s great involvement in
47 The conditions for review of legality by the CJEU in accordance with Art. 263 TFEU were not met.
48 Although the Charter of Fundamental Rights of the EU is only applicable when implementing EU law (Art.
51(1) of the Charter), jus cogens and ECHR rights are applicable.
49 Wessel (n7)
50 Ibid, Wessel elaborates further the problem of using soft instruments in external relations. See 12-13 specifically
for immigration.
51 For example: the Commission publishes regular Progress Reports concerning the implementation of the
EU-Turkey Statement and refugee facilities are funded by billions of euros coming from the EU through the Facility for Refugees in Turkey.
52 J.R. and Others v Greece App no 22696/16 (ECHR, 25 January 2018). For a detailed discussion of the case, see
Annick Pijnenburg, ‘JR and Others v Greece: what does the Court (not) say about the EU-Turkey Statement?’ (Strasbourg Observers, 2018) <https://strasbourgobservers.com/2018/02/21/jr-and-others-v-greece-what-does-the-court-not-say-about-the-eu-turkey-statement/> accessed 12 May 2019. The author suspects that the court may be stricter in more recently dated cases, as the agreement will not be “new” anymore and asylum-seekers will have
18 the EU-Turkey Statement, it may be held that its own court should be able to assess the legality of what in practice amounts to an EU measure. Furthermore, it does not seem fair for Greece to be held liable on its own for an EU policy. Until the accession of the EU to the ECHR, the CJEU remains the only court capable of doing that.
On top of that, the GC decided that it was not in the public interest to disclose the opinion of the Commission’s own legal service regarding the EU-Turkey Statement, thereby reinforcing the idea that the EU is allowed to avoid accountability as well as the assumption of responsibility for whatever happens under this agreement.53 Given the fact that the TEU requires the EU to respect human rights both in its internal and external action, it may be qualified as strange that EU citizens cannot check whether this happened in this instance.54 Of course, technically, this was not an EU measure. Naturally, the NGO suspected the legal advice to be critical of the agreement in relation to human rights, more specifically Turkey’s asylum system. It is to be considered regrettable that the Commission did not further clarify or comment on this issue.
Regarding this humanitarian issue, there are many comments that have already been made by other authors, the most notable of which being the University of Utrecht’s evidence-based assessment of the agreement. Before turning to Turkey’s asylum system, the EU-side of this arrangement needs to be addressed. In order to implement the EU-Turkey Statement, migrants and refugees arriving in hotspots are processed in accordance with a “fast-track procedure”, whereby the entire process including the appeal should take place within 15 days. The report is critical of the procedure itself as well as the hotspots. Capacity problems cause big numbers of migrants and refugees being stuck on the Greek islands in bad conditions without effective access to legal protection. These conditions force some people to agree to voluntary return.55
Other weaknesses the report identifies relate to the partner country, Turkey. Under the Statement, Turkey is to be considered a “safe third country” so that Greece can in principle declare inadmissible asylum applications by migrants and refugees who travelled through Turkey. The same concept will be used for the purpose of implementing the RDAs and is further explored in Chapter 3. Although Turkey already came a long way in updating its asylum system,
been detained in Greek hotspots for a longer time. A case on whether migrants may be transferred back to Turkey and whether Turkey is to be considered a safe third country is still pending. (J.B. v Greece App no 54796/16 (ECHR, communicated 18 May 2017))
53 T-851/16 Access Info Europe v Commission [2018] ECLI:EU:T:2018:69 54 Arts. 2 and 21 TEU
19 the country applies a geographical limitation to the Refugee Convention whereby non-European refugees are not covered. A new temporary protection status was created for Syrians, but they ran into many obstacles in obtaining access to basic services such as health care. Finally, the resettlement under the one-for-one scheme fell entirely under Turkish discretion.56
2.2. The Italy – Libya Memorandum of Understanding (2017) 2.2.1. Content and context of the MoU
Contrary to the Statement discussed above, the MoU is an agreement concluded between only one Member State and a third country. Unsurprisingly, this Member State finds itself at the EU’s external border. Just before the publication of the Malta Declaration on the externalisation of migration, the Italian prime minister and the Libyan head of government recognised by the UN signed a MoU with a duration of three years on 2 February 2017.57 Funding is provided by Italy as well as the EU and a mixed committee is established to oversee implementation and finances. Although the MoU itself is classified as a non-legally binding agreement,58
commitments are made to implement previously concluded bilateral agreements with the Gaddafi regime between 2007 and 2009.59 The purpose of those agreements was to combat
“clandestine” immigration via Libya, for which Italy would provide operational and technical assistance. This ambition was complemented by a commitment to repatriate the supposedly clandestine migrants to their countries of origin with whom readmission agreements were to be concluded.
Italy suspended the agreements in 2011 in response to the Arab Spring. One year later, in the case of Hirsi v Italy, Italy was condemned for its cooperation with Libya from its beginning.60 Despite this judgement, and with encouragement from the EU seeking to increase cooperation with the Libyan coast guard,61 Italy considered it appropriate to engage with Libya again in 2017.
56 Ilse van Liempt (n43) 20-21 and 25-27
57 Memorandum of understanding on cooperation in the fields of development, the fight against illegal
immigration, human trafficking and fuel smuggling and on reinforcing the security of borders between the State of Libya and the Italian Republic [2017] unofficial translation by the Odysseus Network available at http://eumigrationlawblog.eu/wpcontent/uploads/2017/10/MEMORANDUM_translation_finalversion.doc.pdf accessed 17 May 2019 (MoU between Libya and Italy)
58 Pieter Jan Kuijper (n19) 95
59 Treaty on Friendship, Partnership and Cooperation between Italy and Libya [2007]; Additional protocol [2009] 60 Hirsi v Italy App no 27765/09 (ECHR, 23 February 2012)
61 European Council, ‘Malta declaration by the members of the European Council on the external aspects of
20
2.2.2. Concerns
The most pressing concerns relate to the general human rights situation and the lack of recognition of international refugee law in Libya, which are voiced in the case of Hirsi v Italy. Although these considerations relate to the old agreements, they are still applicable to the more recent MoU as the situation in Libya is still worrisome.62 In this case, a group of asylum-seekers from Eritrea and Somalia alleged that Italy had breached several ECHR rights by returning them to Libya, the country they had just left behind, after their “rescue” on the Mediterranean. The ECtHR ruled that Italy, by doing so, had violated a number of rights. The Court held unanimously that there had been a violation of the prohibition on inhumane and degrading treatment, the prohibition of collective expulsion and the right to an effective remedy.
The materials put forth in this case allow for several conclusions in relation to international human rights and refugee law, the applicable rules and concepts of which are further explored in the next chapter. The first relates to the core of the MoU, namely to stop what the parties refer to as “clandestine” immigration thereby implying that all migrants are illegal. This assumption cannot be made without an individual assessment following a fair asylum procedure.63 Until such an assessment has been made, any migrant should be treated as a potential person in need of international protection. Whilst the migrants may be using illegal channels to reach Europe, the migrants themselves are not automatically illegal. Appropriate opportunities to seek international protection should have been offered. The fact that this did not happen also led the Court to conclude that Italy’s practices under the agreement equalled collective expulsion.64
Second, the partner country in question was unreliable in many ways. Besides the general instable situation in Libya and the bad reception conditions, it was also clear beforehand that it had no functioning asylum system and that it did not even recognise the right to ask for asylum. This lead to the repatriation of migrants and refugees to their countries of origin regardless of the risks they would face upon return. According to the ECtHR, upon cooperation with Libya, it was Italy’s own responsibility to ensure that such practices would not take place. The same would apply to the EU when it concludes similar arrangements with other partner countries. The EU would have to seek reassurances and guarantees from those countries.
62 Human Rights Watch, ‘World Report 2019’ (report, 2019) 357-365; UNHCR ‘Position on Returns to Libya -
Update II’ (2018) <https://www.refworld.org/docid/5b8d02314.html> accessed 20 May 2019
63 ECtHR Judge Pinto de Albuquerque sets out the criteria for an individual, fair and effective refugee-status
determination procedure in his separate opinion in Hirsi v Italy (n60).
21 Although the EU-Turkey Statement knows many shortcomings, Palm argues that the MoU amounts to its poor replication. The MoU does not allow for resettlement to the EU, which was the only innovation of the EU-Turkey Statement. This is mainly caused by the assumption that all migrants are illegal anyway. Furthermore, contrary to the EU Facility for Refugees that sets out to whom which amount of money is allocated to, it is unclear where the funding for Libya comes from and what it is exactly used for. Finally, the political and human rights situation in Turkey is not comparable to the situation in Libya. 65 The Turkish asylum system may not be perfect, but at least it exists.
A final concern relates to Italy unilaterally taking action. An express goal of the RDAs is to act on the basis of a joint EU approach, discouraging Member States from taking unilateral action. This appears logical in view of Art. 4(3) TEU, which establishes the general principle of sincere cooperation and is applicable across all EU policy areas. When it comes to interaction with third countries, this principle is even more important. This becomes all the more clear considering Art. 351 TFEU on the termination or modification of prior Member State agreements. In view of this Article, Italy would have to renegotiate or terminate the MoU if it is considered incompatible with the Treaties. In order to establish that, a competence assessment such as the one in chapter 4 would need to be performed to determine whether the EU is competent instead.66 The MoU covers “cooperation in the fields of development, the fight against illegal immigration, human trafficking and fuel smuggling and on reinforcing the security of borders between the State of Libya and the Italian Republic”. Many of these domain at least constitute a shared EU competence, while the latter area touches upon CFSP matters. In this instance, it should also be remembered that the conclusion of the MoU coincided with the Malta Declaration, which endorsed cooperation with the Libyan coast guard.
2.3. Lessons learned
From the above it may be concluded that concerns mainly revolve around three aspects, namely the formal characteristics of the agreement itself, the asylum systems in the partner countries and finally the guarantee of respect for international refugee law and human rights.
65 Anja Palm, 'The Italy-Libya Memorandum Of Understanding: The Baseline Of A Policy Approach Aimed At
Closing All Doors To Europe? – EU Immigration And Asylum Law And Policy' (Eumigrationlawblog.eu, 2017)
<https://eumigrationlawblog.eu/the-italy-libya-memorandum-of-understanding-the-baseline-of-a-policy-approach-aimed-at-closing-all-doors-to-europe/> accessed 24 May 2019
22
2.3.1. Formal characteristics of the agreement
As has been noted before, nor the EU nor IOM and UNHCR speak of the conclusion of a legally binding international agreement. However, the underlying assumption of this thesis is that a formal agreement is desirable for the purpose of amongst others transparency, legal certainty and clearly defined responsibilities for all parties involved. Political declarations create confusion and do not encourage parties to take their responsibilities.
Highlighted by the case of NF v European Council, it is clear that in order for the EU’s own court to be competent to assess the validity of the agreement in question, it must constitute an EU measure and be intended to produce legal effects vis-à-vis third parties in accordance with Art. 263 TFEU. Contrary to the EU-Turkey Statement, the EU should be among the contracting parties as its great involvement is undeniable. The desirability of legal review by the CJEU is therefore self-explanatory. Moreover, the legal status of the agreement should be made clear.
2.3.2. Asylum systems in partner countries
At the core of this category is the principle of non-refoulement, which dictates that persons in need of international protection may not be returned to a place where their life would be in danger.67 Accordingly, priority should be given to ensure that this does not happen, which so far has not been done adequately. Additionally, the examined agreements seem to neglect that to ask for asylum is a right that people have both under international human rights and refugee law.68 In order to safeguard these rights, the following points should be carefully considered in a future agreement.
Firstly, the EU and its Member States may only return migrants to safe third countries.69 Therefore, they should ensure that the partner countries qualify as such. In the EU, the definition in the Asylum Procedures Directive applies.70 If the conditions in that definition have been fulfilled, the relevant partner country may be considered a safe third country. The next chapter examines these criteria more closely.
67 Refugee Convention, Art. 33(1). See also ECtHR Judge Pinto de Albuquerque’s separate opinion in Hirsi v Italy
(n60) for a discussion on the broadening of the scope of the principle of non-refoulement.
68 Universal Declaration of Human Rights (adopted 10 December 1948 UNGA Res 217 A(III) (UDHR), Art. 14(1);
Refugee Convention
69 Regulation 604/2013 establishing the criteria and mechanisms for determining the Member State responsible
for examining an application for international protection lodged in one of the Member States by a third-country national or a stateless person (Dublin Regulation) [2013] OJ L180/31, Art. 3(3)
70 Directive 2013/32/EU on common procedures for granting and withdrawing international protection (Asylum
23 Secondly, the resettlement mechanism for recognised persons in need of international protection is a desirable innovation. Even though it has many shortcomings, such as the lack of influence over the process for recognised persons in need of international protection, the mechanism at least maintains some degree of accessibility to the EU. The concept could be further explored and improved in order to finally work on the promised legal channels for migrants and refugees. In the EU RDA-proposals, only a very restricted version of the resettlement programme has been incorporated and certainly not on a one-for-one basis. On the contrary, the EU emphasised its preference for local solutions.
Finally, distinctions between migrants should be maintained. It is short-sighted and in complete ignorance of refugee law to operate under the presumption that all migrants are illegal. IOM and UNHCR propose to delink the rescue at sea from the post-disembarkation process. On the one hand, this would prevent situations from the one as in Hirsi v Italy from arising as the case processing would take place in a recognised reception facility, but on the other hand, asylum-seekers probably end up having to ask for asylum somewhere else than they had initially foreseen. For example, migrants and refugees rescued by the Italian authorities in international waters presumably would not be enabled by those authorities to ask for asylum in Italy, but rather in the North African partner country they would be taken to. This kind of situation is discussed in 3.3. The RDA-proposals do foresee a process in which a distinction is made between persons in need of international protection, persons with special needs and illegal migrants.
2.3.3. Guarantees of respect for international refugee law and human rights
Over the course of the asylum procedure and in reception centres, refugee law and human rights need to be respected. These rights are set out in the next chapter. However, as was made clear by the ECtHR in Hirsi v Italy “the existence of domestic laws and the ratification of international treaties guaranteeing respect for fundamental rights are not in themselves sufficient to ensure adequate protection against the risk of ill-treatment”.71 The Court adds that this is the responsibility of the partner country to which the asylum management is outsourced
as well as the European state that returns migrants to that partner country.72 Therefore, guarantees are needed.
The MoU provided for a mixed committee “identifying the priorities in action and the instruments to finance, implement and monitor the commitments undertaken”.73 The idea in
71 Hirsi v Italy (n60) para 128 72 Ibid, para 129
24 itself is not bad, but the creation of a similar committee for the supervision of RDAs would need a better defined mandate. In this regard, the EU-Turkey Association Council can be taken as an example. The Association Council was established within the framework of the Association agreement between Turkey and the EU.74 The Council received powers to decide on cases provided for in the agreement and to be implemented by the contracting parties, to make appropriate recommendations, to review progress made under the agreement, to set up committees for assistance and continued cooperation, and to settle disputes. It operates on a ministerial level.
As has been mentioned in chapter 1, the most suitable options for the conclusion of RDA-agreements are either a readmission agreement or an association agreement, with the inclusion of a protocol on the RDAs. The latter agreement becomes more attractive, given that it includes the establishment of an Association Council. Naturally, this would involve a long-term commitment, which would make the RDA-proposal loose its character of a quick fix to migratory problems.
It may be argued that this actually constitutes a desirable development, as a sustainable solution to migration via the Mediterranean is warranted. An association council facilitates structural dialogue on a high level, so that responsibilities can be fulfilled. More specifically, the EU would be able to better oversee and check the developments in the partner country, while that country would be guaranteed ample opportunity to discuss arising problems with the EU and to seek further assistance. This would also ensure compliance with the judgment in
Hirsi v Italy. Furthermore, such a body could safeguard the uniform interpretation of applicable
international instruments.75 Finally, this would also secure transparency in relation to the allocation of funds. While readmission agreements include the establishment of joint readmission committees, they do not enjoy the same powers nor do they operate on the same level as association councils.
The EU already concluded association agreements including association councils with several North African countries that have been approached within the framework of the RDAs, although migration does not play a significant role in those agreements yet. These countries are Morocco, Tunisia and Egypt. Readmission agreements have not been concluded with any of these countries.
74 Agreement establishing an Association between the European Economic Community and Turkey [1963] OJ
L217 (Ankara Agreement)
75 This concern was raised by the Parliamentary Assembly of the Council of Europe, Resolution 1821 (2011) para
25
3. Relevant international human rights and refugee law
The RDA-proposals are to operate in full respect of international human rights and refugee law. All agreements examined so far contain such a commitment, without enumerating relevant instruments nor how they would apply to the agreement in question. It is therefore not surprising that the parties to those agreements have not been able to honour these commitments. This chapter seeks to address that gap by first of all outlining all relevant international human rights and refugee law that ought to be considered in arrangements such as the RDAs. It then turns to the safe third country concept laying at the core of similar outsourcing policies. It is discussed which rules ought to be respected (de jure as well as de facto) by a partner country in order to be able to qualify as a safe third country. The chapter is concluded by an examination of rules that ought to be applicable to the agreement itself, boiling down to the question of whether it is legal to transfer migrants straight from EU-controlled ships to partner countries in North Africa.
3.1. Applicable international refugee law and human rights
Besides domestic legislation, several legal regimes apply to migration: international refugee law, international human rights law and regional regimes.76 While international human rights law applies to all persons, international refugee law applies to a specific group of persons in need of protection. Regional regimes may relate to human rights law as well as refugee law applicable within a limited jurisdiction, such as the EU or the AU. In 2017, the UNHCR published a comprehensive handbook for parliamentarians on international refugee protection and building asylum systems, including very practical information such as suitable admissibility criteria or how to conduct interviews with asylum-seekers.77 It contains an overview of the international legal framework protecting refugees, including all applicable international and regional human rights and refugee law. It sets out the most relevant provisions of each instrument and how the instruments are to be applied. An overview is included below.
76 Pieter Boeles, European Migration Law. 2nd, Rev. Ed. (Mortsel: Intersentia Uitgevers NV, 2014) Chapter 1 77 UNHCR, ‘A guide to international protection and building state asylum systems’ (2017) Handbook for
Parliamentarians No. 27 <https://www.unhcr.org/publications/legal/3d4aba564/refugee-protection-guide-international-refugee-law-handbook-parliamentarians.html> accessed 5 June 2019
26
Instrument Area of law
1951 Convention relating to the Status of Refugees International refugee law 1967 Protocol relating to the Status of Refugees International refugee law 1969 Organisation of African Unity Convention governing the Specific
Aspects of Refugee Problems in Africa
Regional refugee law 1984 Cartagena Declaration on Refugees Regional refugee law 1999 Common European Asylum System Regional refugee law
1948 Universal Declaration of Human Rights International human rights law 1965 International Convention on the Elimination of All Forms of Racial
Discrimination
International human rights law 1966 International Covenant on Civil and Political Rights International human rights law 1966 International Covenant on Economic, Social and Cultural Rights International human rights law 1979 Convention on the Elimination of All Forms of Discrimination
against Women
International human rights law 1984 Convention against Torture and Other Cruel, Inhuman or Degrading
Treatment or Punishment
International human rights law 1989 Convention on the Rights of the Child International human rights law 2006 International Convention for the Protection of All Persons from
Enforced Disappearance
International human rights law 2006 Convention on the Rights of Persons with Disabilities International human rights law 1981 African Charter on Human and Peoples’ Rights Regional human rights law 1990 African Charter on the Rights and Welfare of the Child Regional human rights law 2003 Protocol to the African Charter on Human and Peoples' Rights on
the Rights of Women in Africa
Regional human rights law 1950 European Convention for the Protection of Human Rights and
Fundamental Freedoms and Protocols
Regional human rights law
2008 Charter of Fundamental Rights of the European Union Regional human rights law Figure 2: overview of all applicable international refugee law, human rights law and regional law
It is regrettable that UNHCR did not include a reference to this handbook in its RDA-proposal, as it contains a lot of useful information. Other chapters include guidelines on the determination of persons in need of international protection as well as guidance on how to respect the rights and dignity of refugees. Parties to a future RDA-agreement could commit to adhering to the best practices included in the handbook. This leads to the following question needing to be asked: should all listed instruments be ratified and in practice be adhered to before cooperation can take place?
3.2. Conditions for a safe third country qualification
The concept of a safe third country makes arrangements such as the RDAs possible. The safe third country concept normally arises during the admissibility procedure, because it is used as a ground for inadmissibility of asylum claims of persons who could have made their claim elsewhere. Elsewhere would be a country “where the asylum-seeker could and should have requested asylum in a country that is safe for him or her and is en route to the country where
