While the terrorist attacks in Brussels understandably attracted most of the attention during the last weeks, migration issues, and particularly the situation in Greece and Turkey, remain high on the EU’s agenda. The EU and Turkey recently stepped-up their cooperation in migration matters by adopting a joint statement, that soon appeared questionable in terms of international and European asylum law (Chetail 2016; Labayle and de Bruycker 2016; Mandal 2016; Peers 2016; Roman 2016). Through this contribution, I intend to demonstrate that the EU-Turkey statement is problematic also because of another reason: it was adopted in violation of the European Parliament’s powers and of the democratic principle at large. Therefore, the Parliament has the right, and the moral duty, to bring action against the statement before the Court of Justice.
Bad for Refugees
Both Turkey and the EU are concerned by the migratory crisis prompted, inter alia, by the Syrian civil war. They have negotiated a number of instruments during the last few months, including a Joint Action Plan in October 2015 and a statement on 7 March 2016. The cooperation framework was completed on 18 March, through the adoption of a joint statement, which took effect on 20 March.
Through the latter statement, Turkey committed to readmit migrants who have not applied for asylum in Greece or whose application has been found ‘inadmissible’ or unfounded under the EU’s Asylum Procedures Directive. On the other hand, the EU accepted that, for every Syrian being returned to Turkey from Greek islands, a Syrian will be resettled from Turkey to the EU (the so-called ‘1:1 scheme’), with a maximum of 72.000 persons. The EU also committed to accelerate the visa liberalisation for Turkish citizens and to ‘speed up’ the disbursement of 3 billion euros allocated under the Facility for Refugees in Turkey (a fund constituted by the EU and its Members, which provides humanitarian assistance to refugees in Turkey and their host communities).
It has been argued that the 18 March statement is ‘Kafkaesque’ and ‘morally wrong’ (ECRE 2016), because it sets up a ‘trade in human misery’ (Peers 2016). The statement may also run afoul of European and international law, since it might lead to collective expulsions and may not give asylum-seekers an effective opportunity to apply for international protection in the EU. If Greece defined Turkey as a ‘safe third country’, it would be able to consider any application for international protection by asylum-seekers transiting through Turkey as ‘inadmissible’, after a simple interview (see Articles 33, 34 and 38 of the Asylum Procedures Directive). It would seem that Greece indeed considers Turkey as a ‘safe third country’ (see Commission’s 10 February Communication, footnote 38). Hence, Greece may send back to Turkey any asylum-seeker whose application it considers to be ‘inadmissible’ – that is, virtually any asylum-seeker.
It is not clear, however, whether Turkey actually is a ‘safe country’, given its largely dysfunctional asylum system and the existing inequalities in access to protection (Peers and Roman 2016). For instance, Amnesty International recently reported that Turkey forcibly returned around 30 Afghans, after having forced them to sign “voluntary return” papers. The EU-Turkey statement of 18 March does not assuage the concern for the protection of refugees, notably because it does not put in place effective monitoring of Turkish commitments in the asylum field (Peers 2016).
More generally, one may expect that the statement may contribute to worsen the already difficult situation of asylum-seekers. The UN Refugee Agency (UNHCR) has alleged that the asylum-seekers and migrants who arrived in Greece after the entry into force of the EU-Turkey deal (20 March) are being detained, to be subject to the new return policy. A number of international organisations and non-governmental organisations have consequently suspended at least part of their operations in Greek ‘hotspots’, ostensibly to avoid being ‘instrumentalized for a mass expulsion operation’ (Médecins sans Frontières; see also UNHCR, International Rescue Committee, Oxfam, Norwegian Refugee Council and Save the Children). The deportation of migrants began on 4 April 2016.
Bad for Democracy?
While the concerns for the rights of asylum-seekers are certainly of paramount importance, the statement of 18 March raises also another issue: the respect for the democratic principle. This statement was negotiated and approved, on the EU’s side, by ‘the Members of the European Council’ (that is, the Heads of State or Government of the Member States). To the best of my knowledge, no Parliament – either European or national – ever approved, or even read, the statement before it was adopted.
This is a relevant problem, since the European Council does not have impeccable democratic credentials. Article 10(2) of the Treaty on European Union (TEU) affirms that the European Council is composed of Heads of State or Government ‘themselves democratically accountable either to their national Parliaments, or to their citizens’. However, the democratic accountability of the European Council seems more ‘fictitious’ than real (Curtin, Mair and Papadopoulos 2012). The activity of this institution does not form a core part of the public debate in the framework of national electoral competitions. And, even if EU citizens wanted to closely scrutinise the European Council, they would hardly be capable of doing so in practice. The activity of this institution is not transparent; for instance, there is no public record of the debate on the 18 March statement. Hence, neither European citizens nor national parliaments have a clear picture of the positions expressed by their own governments. Each member of the European Council may thus easily claim credit, or blame others, for collective decisions.
It is remarkable that an agreement on a hotly debated topic, such as international cooperation on migration matters, was concluded by a body whose democratic legitimacy is not beyond doubt. This situation appears problematic, not only as a matter of principle, but also in practical terms. Some democratic debate might have allowed EU institutions to forge a better deal – one that has less negative effects for asylum-seekers and migrants. Some discussion may also have given more legitimacy to the statement, and might have allowed for a clearer allocation of political responsibilities. In a democratic system, it should be possible to identify with precision those who have supported such a crucial arrangement – either to praise them for having facilitated a ‘step towards a sustainable solution’ (Thym 2016), or to criticise them for ‘treating refugees like cattle’ (Chetail 2016).
These problems could, and probably should, have been avoided, by involving the European Parliament in the procedure for the adoption of the 18 March statement. Article 218 of the Treaty on the Functioning of the European Union (TFEU) normally requires the participation of the European Parliament in the conclusion of ‘international agremeents’. Yet, apparently, it was ignored.
An ‘International Agreement’
To demonstrate that the European Parliament should have been involved in the adoption of the 18 March statement, one should preliminarily show that this instrument falls within the scope of Article 218 TFEU. This provision regulates the conclusion of ‘international agreements’, i.e. international undertakings which have binding force (see Opinion 1/75). EU primary law, on the other hand, is silent in respect of the procedures for the conclusion of non-binding international arrangements.
At first sight, the 18 March Statement may seem to be a non-binding international instrument. Peers, in particular, argued that ‘since the agreement will take the form of a “statement” […] it will not as such be legally binding. Therefore there will be no procedure to approve it at either EU or national level, besides its endorsement by the summit meeting.’ This argument has merit, since the formal designation of the 18 March ‘statement’ is typical of non-binding instruments. The statement also seems to employ the terminology of non-binding arrangements: instead of using the modal ‘shall’ (typical of international agreements), it consistently uses the word ‘will’ (that is typical of non-binding arrangements).
Nonetheless, certain elements of the statement suggest that it is a fully-fledged international agreement. As is well known, international agreement means a treaty concluded between States or international organisations in written form and governed by international law, whether embodied in a single instrument or in two or more related instruments and whatever its particular designation (Vienna Convention 1969, Art. 2(1); Vienna Convention 1986, Art. 2(1)). As noted by the International Court of Justice (ICJ), international agreements ‘may take a number of forms and be given a diversity of names’ (Qatar v Bahrain, para 23). Similarly, the EU’s Court of Justice held that an international agreement is ‘any undertaking entered into by entities subject to international law which has binding force, whatever its formal designation’ (Opinion 1/75).
What matters is not the form, but the ‘actual terms’ of the agreement and the ‘particular circumstances in which it was drawn up’. Any international instrument that states what has been ‘“agreed” between the Parties’ and that enumerates ‘the commitments to which the Parties have consented’ creates ‘rights and obligations in international law’. Hence, it constitutes an international agreement. The case-law of the ICJ demonstrates that atypical instruments, such as the minutes of a meeting or a ‘joint communiqué’ (i.e. a statement), may actually be international agreements (see ICJ, Qatar v Bahrein, paras 23-25; Aegean Sea Continental Shelf, paras 96-97).
The 18 March statement constitutes, in my opinion, another instance of atypical international agreement. Like the instrument that formed the object of Qatar v Bahrein, the 18 March statement unequivocally affirms that the parties ‘agreed’ on several actions points. Then, it describes the content of these action points, thereby enumerating the commitments to which the parties have consented. It states, in particular, that ‘for every Syrian being returned to Turkey from Greek islands, another Syrian will be resettled from Turkey to the EU’ (the aforementioned ‘1:1 scheme’). I tend to believe that, by expressly agreeing to specific commitments, the EU and Turkey intended to enter into an international agreement. Interestingly, the very European Commission labels the statement as EU-Turkey ‘agreement’ in its website.
One may wonder whether the 18 March statement merely restates previous engagements. The new arrangements are indeed set to function ‘within the framework of the existing commitments’, i.e. the Greece-Turkey readmission agreement and, from 1 June, the EU-Turkey readmission agreement. Nonetheless, I would argue that the 18 March statement contains original obligations: the ‘1:1 scheme’, in particular, cannot be found in either readmission agreement. The scheme originates from other sources: the joint EU-Turkey action plan of October 2015 and a previous statement (of 7 March). The action plan and the 7 March statement, in any event, do not seem to provide for binding commitments. The joint action plan simply describes what each party ‘intends’ to do, while the 7 March statement merely acknowledges that the parties ‘agreed to work’ on a number of issues. The 18 March statement was arguably adopted to transform generic political compromises, contained in previous non-binding intruments, into legally binding commitments. This circumstance confirms that the 18 March statement is indeed an ‘international agreement’.
If the statement is binding, why is it formulated in such an ambiguous way? The use of misleading language might be intentional. The European Council possibly intended to ‘mask’ the agreement with Turkey as a non-binding instrument, to avoid the cumbersome procedures for the negotiation of international agreements (imposed by Article 218 TFEU). In particular, it may have wished to negotiate the agreement in the intergovernmental framework, instead of entrusting the negotiation to the Commission (cf. Gatti and Manzini 2012). And, above all, it might have sought to avoid consulting the European Parliament – whose approval would have been necessary under Article 218 TFEU, as the next section will show.
It is also possible that the leaders of EU States intended to ‘hide’ the binding nature of the statement to avoid consulting national parliaments. The statement may indeed be binding on both the EU and its Members under international law; in other words, it may constitute a ‘mixed agreement’. The statement is certainly binding on the Union, since it explicitly requests the EU to take certain actions (for instance, covering the costs for readmission procedures). It may also be binding on the Member States, since it contains elements that seem to fall beyond the scope of EU competences: the statement commits the ‘EU’, in particular, to ‘speed up’ the disbursement of the Facility for Refugees in Turkey, which is mostly made up of resources provided by the Member States and which is ‘guided’ by a Committee that includes representatives of the Member States. Had the statement been adopted as an international agreement, some national parliaments may have had the power to authorise its ratification under national constitutional rules – hence, another incentive to conceal the real nature of the statement.