Sunday 31 January 2021

EU Migration Agencies: the Operation and Cooperation of Frontex, EASO and Europol



Dr. David Fernández-Rojo, Universidad de Deusto -


The so-called “refugee crisis” revealed the urge to ensure the functioning of the Schengen area and the Common European Asylum System (CEAS), the desire to operationally assist those Member States most affected by the sudden and extraordinary arrival of mixed migratory flows, and the need to implement effectively and uniformly the EU measures adopted in regard to migration, asylum and border management matters. Against this background, the decentralized EU Agencies, Frontex, EASO and Europol, have emerged as key actors, not only in providing emergency operational assistance to the frontline Member States, but also in implementing the hotspot approach. The expansion of the operational role, multilateral cooperation, presence on the ground and institutional significance within the Area of Freedom, Security and Justice (AFSJ) of Frontex, EASO and Europol, is now unquestionable.


Hence, my book entitled “EU Migration Agencies: The Operation and Cooperation of Frontex, EASO and Europol”, published by Edward Elgar Publishing, comparatively analyzes the evolution of the operational tasks and cooperation of Frontex, EASO and Europol. Special attention is paid to the expansion of the legal mandates of these AFSJ agencies, the reinforcement of the activities they undertake in practice on the ground and to what extent a gap exists between these two dimensions.


The evolution of the operational tasks of Frontex, EASO and Europol is analyzed and two trends are highlighted. Firstly, while the Regulations of these AFSJ agencies continue to stress that their operational role is limited to providing the competent national authorities with the technical assistance they may require, the tasks of Frontex, EASO, and to a more limited extent, Europol, have an operational nature on the ground. Secondly, Frontex, EASO and Europol are increasingly involved in guaranteeing the effective and uniform implementation of EU migration, asylum and border management measures, as well as ensuring that the concerned Member States do not jeopardize the functioning of the Schengen area or the CEAS. These two emerging trends are discussed in turn.


In this book I point out that Frontex, EASO and Europol closely accompany the frontline Member States in the implementation of EU migration, border management and asylum policies. These agencies focus on operationally supporting the competent border, asylum and law enforcement national authorities in effectively implementing EU law. The expansion of EU competences in AFSJ matters has gone hand-in-hand with the reinforcement of their administration, which no longer falls exclusively on the Member States, but rather, on a conundrum of diverse actors, among which Frontex, EASO and Europol play a prominent operational role.


The growing integration that the AFSJ is experiencing has led to a Europeanization of its administration. It is necessary to ensure a uniform and effective implementation of EU border management, asylum and migration laws. The long-standing notion of administrative and implementation power in AFSJ matters is therefore progressively shifting. The deepening of the operational powers and cooperation of Frontex, EASO and Europol is eroding the exclusive procedural autonomy that Member States previously enjoyed, when implementing EU law. These AFSJ agencies increasingly steer and shape the effective and uniform implementation of EU migration, asylum and border management laws and policies at the national level.


Furthermore, the extent of the operational functions of Frontex, EASO and Europol may theoretically range from merely coordinating and providing technical assistance to the Member States, to developing full-fledged enforcement and coercive powers. Since Frontex, EASO and Europol do not have independent executive competences, their tasks can no longer be described as merely technical or supportive. Despite the lack of transparency and the vague legal provisions regulating the activities that Frontex, EASO and Europol undertake in practice on the ground, their tasks do have an operational nature. The issue is that the legal frameworks of Frontex, EASO and Europol lag behind the real operational powers that these agencies exercise on the ground, which creates legal uncertainty.


The reinforcement of the legal mandates and inter-agency operational cooperation of Frontex, EASO and Europol thus reveal a trend, under which these AFSJ agencies are mandated to increasingly develop operational and implementation activities. The operational and implementation role of Frontex, EASO and Europol has followed a constant and linear progression since their respective establishment. While Europol, due to its still markedly intergovernmental nature, is starting to operationally assist the national law enforcement authorities in their national investigations about illegal migrant smuggling, Frontex and EASO already conduct significant operational tasks on the ground and ensure the implementation of the adopted European Union measures at the national level. Whereas the current tasks already represent an erosion of the operational powers and implementation prerogatives of the Member States, none of these AFSJ agencies have been bestowed centralized, fully autonomous operational and enforcement powers on the ground.


The reinforcement of the operational tasks and implementation role of Frontex, EASO and Europol is not in itself an issue. What is problematic is the broad formulation of these AFSJ agencies’ legal bases and the lack of transparency surrounding their operational activities and cooperation, rendering the task of determining the degree of discretion they enjoy difficult. The key challenge involves determining the degree of discretion that Frontex, EASO and Europol enjoy and whether the institutional balance in the EU is respected. In this light, and despite the fact that Frontex, EASO and Europol have not been vested with strictly delegated powers, this book followed the CJEU’s non-delegation doctrine as useful guidance to analyze the legality of these AFSJ agencies’ operational functions under EU constitutional law.


The CJEU, in its Short-Selling judgment (discussed here), updated and relaxed its initial Meroni doctrine, by no longer confining delegation to clearly defined executive powers, but rather to powers precisely delineated and amenable to judicial review in the light of the objectives established by the delegating authority.

Unlike in the case of Short-Selling, the operational powers of Frontex, EASO and Europol are neither circumscribed by well-detailed conditions that limit their discretion, nor clearly detailed in a legal framework or their Regulations. These AFSJ agencies’ operational tasks are not restricted to merely providing technical support to the frontline Member States, but rather, they develop expanding cross-agency operational cooperation and activities on the ground. These agencies’ tasks entail the exercise of discretional prerogatives that are not narrowly delineated or clearly conditioned in any national or EU legal instrument. For instance, Frontex and EASO played a strong recommendatory role in the hotspots, which in principle, is compatible with the non-delegation doctrine, since the concerned Member States are not bound by Frontex and EASO’s recommendations.


Nonetheless, the national authorities, subject to extraordinary migratory pressure, may decide to rubber-stamp the recommendations put forward by the agencies. Frontex’s influence over the Greek officials in determining the nationality of the arriving migrants, Europol’s advice and operational support to the national enforcement authorities to dismantle migrant smuggling networks, and EASO’s admissibility assessment of the asylum applications or the detection of vulnerable applicants encompass in practice discretional and political choices. In these cases, the responsibilities of the agencies are blurred, since the national authorities adopt a final decision based on the assessment of the agencies.


Although fully autonomous enforcement and coercive powers are not possible under the current Treaties and would breach the non-delegation doctrine, the ambiguity and lack of transparency surrounding the operational tasks that Frontex, EASO and Europol undertake on the ground challenge the determination of their discretion and whether they actually make policy choices. In the author’s view, the main limitation and control of Frontex, EASO and Europol’s distinctive operational and implementation role comes from the Member States. While it is true that Frontex, EASO and Europol assist the Member States in matters closely linked to their national sovereignty prerogatives, the competent national authorities that vote at the management boards tightly control their recently reinforced operational, implementation and supervisory functions. Only two representatives of the European Commission have voting rights in Frontex and EASO’s management boards and this figure falls to just one representative in the case of Europol. The presence of the European Parliament in Frontex, EASO and Europol’s management boards is non-existent. Member States also exert their influence over the appointment and supervision of the executive directors, who lead the governance, management and daily administration of Frontex, EASO and Europol.


Member States’ reluctance to fully abandon their well-established bilateral practices, share information and operationally cooperate with Frontex, EASO and Europol in core national sovereign matters, like border management, asylum or migration, is especially reflected in these AFSJ agencies’ management boards. The Member States will thus maintain control of the strategic decisions and the daily management of Frontex, EASO and Europol. While centralizing on the executive, decisional and enforcement powers of Frontex, EASO and Europol will ensure a fully effective and harmonized implementation, it is important to bear in mind that these agencies represent an institutional trade-off or a common ground between intergovernmentalism and communitarization in the AFSJ. That is, Member States do not wish to relinquish further sensitive competences to the EU Institutions; but at the same time, they increasingly need supranational operational assistance regarding matters that can only be effectively managed in an integrated manner at the EU level. For this reason, whereas Europol, Frontex and EASO have been conferred upon significant operational tasks, none of these agencies are vested with full decisional, enforcement or coercive powers, which remain as an exclusive competence of the competent national authorities.


Hence, this book makes four main contributions. First, it maps Frontex, EASO and Europol as EU decentralized agencies, which are clearly distinguished by their operational powers and by the possibility to directly assist the competent national authorities on the ground. In particular, the establishment and early operational functions conferred on Frontex, EASO and Europol are studied. Second, it comparatively analyzes the reinforcement of the operational tasks vested on Frontex, EASO and Europol, as well as the extent of their assistance on the ground and influence on the implementation prerogatives of the national authorities in the aftermath of the “refugee crisis”. Third, it explores the bilateral and multilateral inter-agency cooperation between Frontex, EASO and Europol. Specifically, the expanded multilateral and operational cooperation that takes place in the hotspots is studied. Fourth, the limitations to the reinforced operational activities and cooperation of Frontex, EASO and Europol is analyzed. The constitutionality and legal bases of these AFSJ agencies, as well as the degree of discretion that they enjoy according to the Court of Justice of the European Union (CJEU) non-delegation doctrine, is examined. The internal administrative organization and governance of Frontex, EASO and Europol is also studied as to determine the influence and real control that the Member States and civil society may exert over the increasing operational powers these AFSJ agencies have been conferred.


Barnard & Peers: chapter 26

JHA4: chapter II:4

Photo credit: Rock Cohen, via wikimedia commons

Saturday 30 January 2021

Negotiations for EU accession to the ECHR relaunched - overview and analysis

The prospect of the EU accession to the ECHR has been anticipated for many years.  With the amendment of the Treaties to provide for this (in the Lisbon treaty), and the negotiation of an accession treaty by 2013, it looked as if the CJEU ruling on the accession treaty in December 2014 would formalise the celebrations. But in the event, the Court’s surprising ruling (Opinion 2/13) – highly criticised on this blog – curtailed the accession process.

Eventually, however, the process was restarted in 2019 and has gathered pace since. The following two blog posts – by Anita Kovacs and Stian Øby Johansen – update and analyse the latest developments, providing in turn an overview of the process and more detailed comments on the issues. The comments by Stian Øby Johansen are partly reblogged from here and here.


The on and off negotiations on the EU’s accession to the ECHR – it’s complicated

Anita Kovacs, MA in EU Law, King’s College London

The European Union’s (EU) accession to the European Convention of Human Rights (ECHR) has been on the European agenda for decades. This post intends to provide a short review of past events, an insight into the recently revived negotiations as well as an outlook on what to expect next.


In 1996, the Court of Justice of the European Union (CJEU) gave its first opinion on the EU’s accession to the ECHR (Opinion 2/94). At the time, the court ruled that the European Community did not have the competence to accede to the ECHR (paras 34-35).

A legal basis for the accession was established when the Lisbon Treaty entered into force in 2009. Article 6(2) TEU states that the EU shall accede to the ECHR and hence, creates a legal obligation.

The 14th Protocol to the ECHR paved the way for the accession by introduction of Article 59(2) ECHR declaring that the EU may accede to the ECHR.

In 2013, the Accession Agreement was submitted to the CJEU to obtain an opinion on the compatibility with the treaties pursuant to Article 218(11) TFEU.             

In the following year, the court delivered the famous Opinion 2/13 (discussed here) and despite significant arguments in favour of an accession, concluded that the Accession Agreement was incompatible with EU law (para 258). Interestingly, the European Commission, supported by observing Member States (para 109), advocated the opposite position (para 73).

Advocate General Kokott suggested a ‘qualified approval’ of the Accession Agreement meaning that the agreement is, in principle, compatible with EU law but still requires some modifications to it (paras 278-280). She also provided inspirations on possible ways to conciliate EU law requirements and the accession (e.g. para 234).


In late 2019, the EU and the Council of Europe announced their willingness to proceed with the negotiations on the EU’s accession. The negotiations will take place within an ad hoc Group composed of representatives of the 47 Council of Europe members and one EU representative (“47+1 Group”).

Since January 2020, the parties are formally ready to resume negotiations on the accession of the EU to the ECHR. (See the Council of Europe’s website on the accession process).

A first informal meeting took place in June 2020, where the European Commission clarified that it intends to realise the accession by ‘modulations’ to the Accession Agreement. These should preserve the EU’s special characteristics while meeting the requirements set out in Opinion 2/13. The speakers reiterated that the CJEU’s objections require some serious work, but they are not insurmountable. It was further made clear that the talks are not constituting a new beginning for the accession process, the negotiations are rather picked up where they were left off in 2013. This means that the Draft Accession Agreement lays the foundation for the upcoming meetings and provides the parties with a frame to work in.

The European Commission elaborated four main areas of concern for the upcoming negotiations: the EU specific mechanisms of the procedure before the ECtHR, inter-party applications under Article 33 ECHR and references for an advisory opinion from national courts to the European Convention of Human Rights under Protocol No.16, the principle of mutual trust and the EU’s Common foreign and security policy (CFSP).

These aspects have found their way onto the agenda of the first official negotiation meeting after Opinion 2/13. They were further divided into subordinate issues by the Chair of the 47+1 Group in preparation of this meeting. The Chair’s paper to structure the discussions also included various considerations possibly decisive for finding a solution to the ten issues elaborated.

The meeting initially scheduled for March 2020 was postponed due to the Covid-19 pandemic to 29th September to 1st October 2020 (6th negotiation meeting). It is to be welcomed, that in this meeting, the 47+1 Group decided to extend the scope of consultations with NGOs and members of the civil society in terms of frequency and participants. The meeting summarised various positions on the aspects raised by Opinion 2/13. The main discussions, however, were postponed to the next meeting.

The 7th negotiation meeting conducted from 24th to 26th November 2020 mainly provided the stage for discussions on the EU specific mechanisms of the procedure before the ECtHR, inter-party applications under Article 33 ECHR and references for an advisory opinion under Protocol No.16. Several proposals were put forward, most of them requiring further consideration. Surely, the meeting has demonstrated the willingness to find flexible solutions and to advance these constructively.

The next meeting is scheduled for 2nd to 4th February 2021, where the issues of mutual trust and CFSP shall be examined. The 9th negotiation meeting is anticipated in late March 2021.


According to Article 218(6) TFEU, the European Parliament’s consent to the Accession Agreement is required. Only then can the Council adopt a decision concluding the agreement. The Council must do so unanimously, and the Accession Agreement should be concluded after the EU member states have approved it in accordance with their constitutional requirements (Article 218(8) TFEU).

As mentioned above, Opinion 2/13 was delivered following a request by the European Commission according to the procedure provided by Article 218(11) TFEU. This article provided that an opinion on the compatibility of an agreement with the treaties may be obtained. Hence, one could raise the question whether the revised Accession Agreement will again be subject of an opinion by the CJEU. In its request, the Commission underlines what the CJEU has often stressed in its opinions (see e.g. Opinions 2/13 and 2/94). The purpose of such opinions was to “forestall complications which would result from legal disputes concerning the compatibility with the Treaty of international agreements binding upon the Union” (para 60). Another argument in favour of an opinion is that the implementation of internal rules should be supported by the certainty that the foundation of these rules is compatible with EU primary law.

Not only the CJEU may be given a chance to rule on the revised Accession Agreement. As Switzerland proposed during the informal meeting in June 2020, the ECtHR could be asked to give its opinion on the agreement. Article 47 ECHR could potentially provide for this opportunity. The possibility of such an opinion was mentioned again in the September meeting, adding that the revised agreement would first have to be submitted for adoption to the Council of Ministers.

After the CJEU approves, the Council of Ministers must adopt the revised Accession Agreement and then open it for signatures. All 47 High Contracting Parties to the Convention will have to ratify the revised agreement according to Article 59 ECHR.

In the meeting in June 2020, it was emphasised that the risks of a failed accession can be severe. Participants agreed on the necessity and urgency of the matter. The aims of legal certainty and coherence of human rights protection should always be borne in mind.

While it is not clear if this negotiation round will call for a similar time frame of three years like the previous one, we will certainly have to be patient until we see a solid result.

Even though Opinion 2/13 smashed our hopes, the 47+1 Group is not starting from scratch this time. Having said that, it is evident that this project still requires a lot of effort from everyone involved. 

Let us hope that once the marriage contract is honed, the CJEU marries the couple and the EU and ECHR can finally change their relationship status to ‘married’.


EU accession to the ECHR: Details of the relaunched negotiations

Stian Øby Johansen, Associate Professor at the University of Oslo Faculty of Law

The first renegotiation meeting

The report of the first renegotiation meeting – or the sixth negotiating meeting in CoE parlance, continuing the numbering from the previous round – should be read alongside another key document, the “Paper by the Chair to steer the discussion at the 6th meeting of the CDDH ad hoc group (47+1)“, which was drafted to structure the negotiations. Also, the EU Commission’s negotiating mandate – a heavily guarded document in the first round of negotiations – was leaked almost simultaneously with its approval in October 2019.

The key issues facing the negotiators

In her paper, the Chair attempted to structure the negotiations around four “baskets” of issues:

Basket 1: EU-specific mechanisms of the procedure before the ECtHR (co-respondent mechanism, prior involvement of the CJEU, shared responsibility).

Basket 2: Operation of inter-party applications and of references for an advisory opinion by the ECtHR (ECHR article 33 & Protocol 16).

Basket 3: The principle of mutual trust between EU Member States.

Basket 4: EU acts in the area of the Common Foreign and Security Policy.

The Chair’s paper also included a list of “tools” available to the negotiators, spanning from amendments to the ECHR and the Draft Accession Agreement (DAA) to simple clarifications of the Explanatory Report to the DAA and declarations to be made at the signature of the DAA. The Chair’s paper then went on to analyze the issues under each basket, identifying the relevant parts of the DAA and Opinion 2/13 for each issue, as well as possible action by the 47+1 Group using the “tools” at their disposal.

The Chair in other words tried to play an active role, to ensure that the discussions are focused on the key issue. To me, this strategy seems appropriate for the complex task at hand. Indeed, the active role of the Chair was crucial to bringing the negotiations to a close the last time around. Then, as now, the Norwegian delegate Tonje Meinich chaired the negotiations, and in 2019 she published an article detailing her experiences of the first round of negotiations in The International Journal of Human Rights.

The Chair’s paper appears to have been fairly well received by the negotiating parties. The agenda of the negotiation meeting was clearly drafted with the proposed “baskets” in mind. During the negotiation meeting, though, at least two delegations announced their intentions to raise additional issues during these renegotiations (Meeting report, para 41). Moreover, one delegation inquired about the possibility for an opinion by the ECtHR on the renegotiated DAA (Meeting report, para 43).

Progress made at the sixth negotiation meeting

At the sixth negotiation meeting, the baskets were discussed in the order they were listed above. From the Meeting Report, it appears that the two first baskets were most thoroughly discussed, and some progress made, while the last two baskets were discussed in a more superficial manner.

No concrete, written proposals for amendment of the DAA or the other draft accession instruments appear to have been put forward during the sixth negotiation meeting. However, the discussions generally progressed to the point that the Chair invited concrete textual proposals for discussion during the next (seventh) negotiation meeting.

In the following I will point out the key highlights from the sixth negotiations under each basket, and assess the progress made.

EU-specific mechanisms of the procedure before the ECtHR (“Basket 1”)

An array of sub-issues were discussed under this heading, including the so-called co-respondent mechanism, the so-called prior involvement procedure, and the attribution of conduct/apportionment of responsibility between the Union and its Member States. The thread binding them all together is, according to the CJEU, that they all carry with them the risk that the ECtHR may incidentally have to interpret the internal division of competences between the Union and its Member States.

For the purely procedural mechanisms involved – co-respondent mechanism and the prior involvement procedure – the most obvious solution would be to remove any threshold criteria for triggering them. That is because, under the 2013 DAA, considering whether those thresholds were met was a task for the ECtHR. Even though the thresholds were so low that they almost amounted to a review of whether there was abuse of process, the CJEU in Opinion 2/13 nevertheless found them to be incompatible with Union law.

Making both the co-respondent procedure and the prior involvement procedure unconditional rights were discussed at the sixth negotiation meeting. With the exception of one delegation, there appears to have been a generally positive reaction to making the prior involvement of the CJEU an unconditional right (Meeting report, para 17). However, views were more mixed with regard to removing the threshold for triggering the co-respondent mechanism (Meeting report, para 13-14).

In a case involving both the Union and one or more Member States, the ECtHR’s incidental determinations regarding attribution of conduct may obviously entail assessment of the distribution of competences. In the negotiations leading up to the 2013 DAA this issue was extensively discussed. When white smoke finally appeared, Article 3(7) of the 2013 DAA included a nebulous clause providing for the shared responsibility of all co-respondents (the Union and one or more Member States) – unless the ECtHR “decides that only one of them be held responsible”. This provision could be read as anything from a strong presumption of shared responsibility to a carte blanche for the ECtHR to decide for itself who is responsible.

In Opinion 2/13, the CJEU found that article 3(7) of the 2013 DAA was incompatible with Union law merely due to the risk that the ECtHR could avail itself of the exception to shared responsibility (paras 229–234). Indeed, the CJEU appears to assert that even if the provision is interpreted as a nearly insurmountable presumption, its very existence is incompatible with Union law.

The obvious “solution” to this objection by the CJEU is to delete the exception. In the sixth negotiation meeting, there was some support for this option, but it was far from unanimous (Meeting report, para 16). One delegation also objected, “as a point of principle”, to the idea that a participant before the ECtHR could “set the rules for how it will participate”, and thus be in a privileged position (ibid). The Chair invited concrete drafting proposals for the next (seventh) negotiation meeting. If I allow myself to speculate, based on experience from the negotiations leading up to the 2013 DAA, I predict that the differences between the “EU side” and the Non-EU CoE States will become quite pronounced when they finally get to discussing concrete proposals.

The seemingly tricky issue of how to deal with ECHR reservations made by an EU Member States that is co-respondent with the Union was also discussed. Interestingly, though, the 47+1 actually managed to reach an agreement that “such reservation would remain valid” (Meeting report, para 15). Concrete drafting proposals to put that agreement into action were to be discussed at the next negotiation meeting.

Under this basket the negotiators also discussed the need for a procedure for keeping the Union systematically informed of cases before the ECtHR where the co-respondent mechanism and prior involvement procedure may apply (Meeting report, para 18). Unsurprisingly, some of them asked the obvious question: why should the ECtHR be responsible for keeping the Union informed, and not the EU Member States? Indeed, the EU Member States are probably better placed than the ECtHR when it comes to assessing whether a case also raises issues of Union law. No firm answer was provided in the sixth negotiation meeting, however. All options were kept open, and the negotiations will return to this issue at the seventh negotiation meeting.

Operation of inter-party applications and of references for an advisory opinion by the ECtHR (“Basket 2”)

These two seemingly disparate issues have in common that they could, potentially, be used by EU Member States to violate provisions of the TFEU – notably the infamous Article 344 (Meeting report, para 20), which requires EU Member States to use the procedures in EU law if they have disputes about EU law between them.

With regard to the issue of inter-party applications (ECHR article 33), the CJEU held in Opinion 2/13 that “only the express exclusion of the ECtHR’s jurisdiction under Article 33 of the ECHR over disputes between Member States […] within the scope ratione materiae of EU law” would be compatible with TFEU article 344 (para 213). In other words, the mere “existence of such a possibility” – i.e. that a Member State could launch an inter-party case in violation of Article 344 – led the CJEU to find the 2013 DAA incompatible with that provision (Opinion 2/13, para 208).

During the sixth negotiation meeting, there was strong opposition from some Non-EU CoE Members against amending the ECHR to take this CJEU objection into account (Meeting report, para 22). It appears that those states considered this – rightly, in my view – an internal issue for the Union and its Member States. As I have argued before, the CJEU reinterpreted TFEU Article 344 in Opinion 2/13 – establishing new and stricter requirements. Consequently, Opinion 2/13 seemingly makes it almost impossible for the Union and its Member States to enter into international agreements with proper dispute resolution mechanisms. Both the text of TFEU Article 344 and the preexisting CJEU case-law arguably implied that the mere existence in an agreement of an avenue through which Member States may violate TFEU Article 344 was not problematic.

It should therefore be no surprise that I am sympathetic to the suggestion of the representative of the ECtHR Registry during the sixth negotiation meeting, namely that “the very premise underlying the requirement stated by the CJEU in Opinion 2/13 that the use of Article 33 ECHR be expressly ruled out […] would not be compatible with Convention case-law and should perhaps be revisited” (Meeting report, para 24). Leaving some CJEU objections to the side in this manner, with the hope that the CJEU has changed its mind over the last years, is obviously risky. But the risk of a potential second negative Opinion is worth taking, in my view, rather than going too far in the direction of giving the Union special treatment.

Some potential “solutions” were discussed, including the suspension of inter-party ECtHR proceedings until the CJEU had determined whether they violated TFEU Article 344 or not (Meeting report, para 26). However, all such solutions fall short of what the CJEU actually required – and thus also carry with them a risk of a second negative Opinion.

The Chair ended the discussions on TFEU Article 344 by concluding that “no delegation had spoken against the general principle that inter-party-applications brought before the ECtHR in violation of EU law were undesirable and should be avoided” (Meeting report, para 27). She also noted that there as “sufficient appetite” for further exploring solutions based on the suspension of ECtHR proceedings (ibid).

With regard to requests by national courts for advisory opinions by the ECtHR under ECHR Protocol 16, the CJEU objected in Opinion 2/13 that domestic courts could circumvent the preliminary reference procedure as regards EU law under TFEU Article 267 (para 198). Reading this part of Opinion 2/13 closely, one see that what the CJEU seems to fear here is not that it wouldn’t be involved before the ECtHR gives an advisory opinion, but rather that a request for an advisory opinion would trigger the 2013 DAA prior involvement procedure. This means that the case is brought, by a domestic court, before the CJEU, using a different procedure than TFEU Article 267.

This objection was always somewhat puzzling. First, it is difficult to understand why the exact basis of the CJEU’s jurisdiction is – TFEU article 267 or the DAA prior involvement procedure. The CJEU never says clearly what the problem would be. Perhaps it is simply the fact that the prior involvement was supposed to happen using an expedited procedure (2013 DAA article 3(6) and Explanatory Report para 69) – thus potentially creating a “fast track” to a CJEU opinion in cases raising issues of Union law and fundamental rights? Second, the CJEU does not seem to trust that its Member States and their domestic courts will play by the rules. This lack of trust is also evident in the CJEU’s above-discussed objections relating to TFEU Article 344 issue.

At the sixth negotiation meeting, little progress was made. The 47+1 Group only seems to have agreed on two preliminary points. First, that the CJEU’s objection regarding ECHR Protocol 16 were not as strongly formulated than its objection concerning TFEU Article 344. Second, that the current wording of paragraph 66 of the DAA Explanatory Report “state that the application of the prior involvement-procedure would presuppose an application which the co-respondent mechanism applies”. Exactly what this latter point of agreement means or achieves remains somewhat of a mystery, though. But perhaps this agreement implicitly presupposes that the co-respondent mechanism does not apply to requests for advisory opinion under Protocol 16, since the 2013 DAA provisions on the co-respondent mechanism (Article 3) consistently uses the term “application”? However, not applying the co-respondent mechanism in Protocol 16 cases could lead to the “Basket 1” risk that the ECtHR would have to incidentally determine the division of competences between the Union and a Member State.

Is there a way out of these dilemmas? In my view, there are at least two. Firstly, the Protocol 16 issue is an obvious candidate for EU-internal solutions. A drastic, but effective, solution would be for the Union and the EU Member States to agree among themselves not to ratify Protocol 16 (or to withdraw from it, for those who have already ratified). Less drastic options along the same lines may also be possible.

Secondly, there is always the option of not making substantive changes to the 2013 DAA, instead banking on the CJEU to have changed its mind. Given the quite distrustful approach the CJEU had to its Member States concerning the issues under “Basket 2” in Opinion 2/13, this might be a possible, and perhaps even preferable, option. But there is no harm in thoroughly scrutinizing other potential solutions.

The principle of mutual trust between EU Member States (“Basket 3”)

Taken to its extreme, the Union law principle of mutual trust may be at variance with the ECHR system. That is because that principle, according to the CJEU in Opinion 2/13 in the policy Area of Freedom, Security and Justice (AFSJ) requires EU Member States to “presume that fundamental rights have been observed by other Member States” (para 198, emphasis added). An EU member state may thus, save for in exceptional cases, “check whether that other Member State has actually, in a specific case, observed […] fundamental rights” (ibid). In contrast, the ECHR’s non-refoulment obligations did not distinguish between whether the High Contracting Parties involved are members of the same economic integration organization or not.

Since 2014, both courts have climbed down from their horses somewhat.

In Avotiņš (which I discussed here), the ECtHR Grand Chamber recognized the “importance of the mutual recognition mechanisms” for Union law (para 113). That recognition of the principle’s importance was, however, immediately followed up by the ECtHR’s observation that “the aim of effectiveness pursued by some of the methods used [in the AFSJ, such as mutual trust,] results in the review of the observance of fundamental rights being tightly regulated or even limited” (para 114).

For its part, the CJEU has since 2014 significantly broadened the exceptions to mutual trust when it comes to fundamental rights, notably in the cases of Petruhhin (discussed here) and Jawo. It may not imply blind trust anymore, but exceptions to mutual trust still remain, well, exceptional.

During the sixth negotiating meeting the 47+1 group barely scratched the surface of this basket, laconically noting the “increased convergence” between the case-law of the two courts, and tasking the CoE secretariat with preparing a compilation of the case-law for the next (seventh) negotiation meeting. (this compilation was later published on the CoE accession website.)

EU acts in the area of the Common Foreign and Security Policy (“Basket 4”)

The 2013 DAA would have granted the ECtHR jurisdiction over conduct carried out under the auspices of the EU’s Common Foreign and Security Policy (CFSP) – a policy that is at least partially outside the CJEU’s jurisdiction. This followed from the simple fact that the 2013 DAA did not distinguish between policy areas. Indeed, the principle that the EU should accede to the ECHR on equal footing with other High Contracting Parties seems to demand such a result. No party to the ECHR is entitled to exclude entire areas of policy from the ECtHR’s jurisdiction.

In a move Steve Peers aptly termed “judicial politics of the playground”, the CJEU held in Opinion 2/13 that since it did not have jurisdiction over (parts of) the CFSP, neither could any other international court. By allowing the ECtHR to adjudicate cases arising from CFSP-related conduct the CJEU consequently asserted that the 2013 DAA “faile[d] to have regard to the specific characteristics of EU law with regard to […] judicial review […] in CFSP matters” (Opinion 2/13, para 257).

Finding a way forward through this obstacle, without undermining the ECHR system, is obviously challenging. While the CJEU’s case-law on the scope of the CFSP jurisdictional carve-out has evolved since 2014 (see this discussion of the most recent relevant judgment, with further links), it remains that certain CFSP acts – including acts that may violate human rights – are covered by the carve-out (see e.g. Johansen 2020, Ch 4, particularly at 142–144). The DAA thus hasn’t (yet) been saved by an arguably expanding CJEU jurisdiction over CFSP matters.

There was little progress in finding a solution in the sixth negotiating meeting. Familiar suggestions of including a “clear attribution clause” was mooted (Meeting report, paras 37–38). However, it is unclear what an attribution clause would achieve, since the CJEU’s objections in Opinion 2/13 related to its lack of jurisdiction over CFSP acts. Automatically attributing CFSP acts to the Member States – either exclusively or jointly with the Union – for the purposes of the ECHR does not appear to have any effect on this issue. It would not remove the CFSP from the ECtHR’s jurisdiction, and it would obviously not widen the scope of the CJEU’s jurisdiction.

In the end, the Chair reached the lethargic conclusion that there was a “common goal by the delegations […] that, in order to avoid ‘black holes’ in the European human rights protection, the Convention system should be able to accommodate all acts in the CFSP area. The question was ultimately to find that appropriate way how to get there.” (Meeting report, para 39.)

Further issues raised

Towards the end of the sixth negotiation meeting, some delegations brought up additional issues that were not included in the four “baskets”.

First, the Chair and the CoE Secretariat brought up the issue of the relationship between the mimimum standards clauses: ECHR article 53 and the EU Charter of Fundamental Rights article 53 (Meeting report, para 40). To overcome the CJEU’s worries in Opinion 2/13 concerning the coordination of these two provisions, it was preliminarily suggested to clarify their relationship in the Explanatory Report to the DAA.

Second, two delegations pointed out that there had been changes in the CoE since the adoption of the 2013 DAA, and that articles 6, 7, and 8 of the DAA (dealing with the participation of the Union in CoE organs and its budgetary contribution to the CoE) therefore had to be revised to reflect those changes (Meeting report, para 41). These delegations also announced that they would bring up further issues, as already mentioned.

Finally, one delegation inquired about the possibility if asking the ECtHR for an opinion on the renegotiated DAA, as alluded to above (Meeting report, para 43). Asking the ECtHR for such an opinion would in principle be possible under ECHR article 47. In my view, doing so would probably also be a wise move. A positive opinion by the ECtHR may potentially influence the CJEU when it will, inevitably, be asked to give its opinion on the renegotiated DAA. Moreover, waiting for an ECtHR opinion is unlikely to delay the process much. It might even be possible to ask both courts for an opinion at more-or-less the same time. If so, I would expect that the ECtHR will be able to deliver an opinion sooner than the CJEU. Last time the ECtHR was asked to give an opinion under ECHR article 47, it only took about six months to do so.


While the negotiations have no doubt started, it is already apparent that they will take time. Several of the CJEU’s objections in Opinion 2/13 concern issues that were extremely delicate in the negotiations of the 2013 DAA. From the perspective of the non-EU Member States of the CoE, the negotiations are now essentially being reopened to deal with mostly internal affairs between the EU and its Member States. This is a recurrent theme in EU external relations: the externalization of issues that, at least in my view, should be dealt with internally.

Against this background, the negotiations will likely be quite difficult. Hopefully, however, the obstacles can nevertheless be overcome without undermining the ECHR system. If not, there is only one way forward: amending EU primary law to neutralize the effects of Opinion 2/13.


The second renegotiation meeting

Some progress was made at the meeting of 24–26 November 2020 (which I will refer to as the 7th meeting from the start of talks), but from reading the meeting report it is becoming increasingly clear that no immediate breakthroughs are in sight.

The following briefly summarizes and takes stock of the progress made. I begin by listing the issues that were not dealt with at all during the 7th meeting, and then turn to the issues which were the subject of negotiations at the 7th meeting. Finally, I will say a couple of words about other business at the 7th meeting, notably the exchange of views with representatives of civil society.

Issues deferred to later meetings

During the meeting, there was only time to discuss issues relating to the first two of the four baskets defined above. The last two, and arguably the most difficult ones – mutual trust between EU member states (“Basket 3”) and the situation of EU acts in the CFSP area falling outside the CJEU’s jurisdiction (“Basket 4”) – were not discussed at all.

No particular explanation was given for not discussing issues falling under “Baskets” 3 and 4, but I guess there was simply a lack of time. The meeting report states that the 47+1 group will get back to the issue at its next (8th) meeting.

Issues discussed at the 7th meeting

Since the 6th meeting, the EU side had provided concrete proposals concerning both “Baskets” 1 and 2. These two therefore took center stage at the 7th meeting.

The exact proposals themselves are, sadly, not publicly available. Judging from the list of working documents appended to the meeting report, the document in question is probably the “Negotiation Document submitted by the European Union on 2 November 2020”, which is labeled as “restricted”. Due to the lack of insight into the exact proposal, a certain amount of (qualified) guesswork is necessary when analyzing the meeting report.


The 2013 DAA, article 3(2), envisaged that the Union could join a case against a Member State as “co-respondent”, but only if certain conditions were met. Crucial among these conditions was whether it “appears” that the allegations of the complainant calls into question the ECHR compatibility of the Union’s constituent treaties. The ECtHR was to be the judge of whether it was “plausible” that this low threshold criterion (“appears”) was met, see 2013 DAA article 3(5).

In Opinion 2/13, the CJEU found this to be unacceptable. That was because it could allow the ECtHR to decide on – albeit as a mere procedural and preliminary matter, and using an extremely low threshold – the distribution of competences between the Union and its Member States (Opinion 2/13, para 231).

To overcome this issue, the EU proposal discussed at the 7th meeting of the 47+1 group was one foreseen long ago as a “safe” solution: that the Union is given an unilateral right to join proceedings as co-respondent. It was also suggested that the Union be afforded a similar unilateral right to later terminate its participation as co-respondent. That solution was not chosen during the 2013 DAA negotiations, since, if I remember correctly, it was rejected by non-EU CoE Member States.

The same group of states also appear to be skeptical this time around, although their position appears to have been softened somewhat following Opinion 2/13. Some delegations suggested moving the criteria to other places in the draft accession instruments (i.e. away from the DAA itself), but other still preferred to keep then in the DAA. In this connection, the difference between the EU proposal and the 2013 DAA solution was “discussed”, but no particular decisions appear to have been taken.

Some other, less significant proposals relating to the co-respondent mechanism were also discussed, such as how the Union should be informed of upcoming cases and responsibility issues when EU Member States have reservations against ECHR provisions. Some progress was made on these issues, but again not much is concrete.

Under this baskets an EU proposal for amending DAA article 3(6) on the prior involvement procedure was also discussed. The “problem”, as framed by the CJEU in Opinion 2/13, is virtually identical to that which affects the co-respondent mechanism, since under the 2013 DAA the inclusion of the Union as co-respondent was a trigger for the prior involvement procedure. It is unclear from the meeting report what the EU proposal in this connection was, but it likely suggested clarifying that the ECtHR would not be the judge of whether the CJEU had already assessed the ECHR compatibility of the (EU law-related) measure in question. The report does not say much about the discussions, other than that the ECtHR registry was happy to not be the one making such determinations. No opposition against the proposal was noted. Although we don’t know fore sure, perhaps this was the point where most progress was made.

In sum, the 7th negotiation meeting did not result in much progress under “Basket 1”, although one should not underestimate the importance of concrete proposals hitting the table. Concrete proposals tend to frame and structure negotiations, and makes it necessary for the opposing sides to take a stance. This will hopefully lead to more progress down the road.


On the issue of inter-party applications, the EU proposal appears to have been a clearly formulated exclusion of inter-party cases between EU Member States. In other words, just what the CJEU ordered in Opinion 2/13.

The ensuing discussion appears to have been both frank and brief. “Some delegations suggested that this was rather a matter related to internal EU matters”, as the meeting report (para 13) laconically states. As noted above, I completely agree with the “some delegations”, but at the same time Opinion 2/13 does not appear to leave much wiggle room in this connection. Either inter-party cases between EU Member States have to be completely excluded, or one must bet that the CJEU has changed its mind.

The 47+1 group left the discussions of this issue here, and agreed to revert to it at its next meeting.

A related proposal from the EU would amend the DAA to make it possible for the Union to ask the ECtHR to suspend inter-party applications between EU member states and/or requests for advisory opinions by the ECtHR (protocol 16). Some delegations had reservations of various kinds to this proposal as well, and it was left for further discussion at the next meeting.

Finally, the 47+1 group discussed an EU proposal on the coordination of ECHR article 53 and the Charter of Fundamental Rights article 53. The exact contents of the proposal are unclear to the readers of the meeting report. There was some support for it, as long as it was clear that the minimum protection enshrined in ECHR article 53 was maintained. Some delegations objected to a provision in the DAA itself concerning this, and suggested that it was a better fit for the explanatory report.

The chair concluded that there was some common understanding on the substance of this final proposal, and suggested that the group should discuss a refined wording during the next negotiation meeting. Again we see progress in connection with more technical matters requiring clarification, rather than any real amendments to the 2013 DAA.


In addition to the four baskets of issues identified by the chair before the 6th negotiation meeting, a new basket of issues was added to the list: articles 6–8 of the (2013) DAA, dealing with EU participation in CoE organs and EU contributions to the financing of the ECtHR’s work.

There were some preliminary discussions of these issues at the 7th negotiation meeting. While there was no final outcomes on any of the issues raised under this basket, the discussions appear to mostly have concerned fine-tuning and updates to the DAA and explanatory report in light of recent developments within the Council of Europe. It seems unlikely that these issues will derail the negotiations.

Other business at the 7th meeting

Among the other business of the 7th meeting was an exchange of views with civil society. An array of human rights organizations took part, including Amnesty International, the International Commission of Jurists, and the European Network of National Human Rights Institutions.

The summary of the civil society contributions in the meeting report probably does not do justice to all the views expressed by civil society. Still, when reading the summary, I was struck by how lofty the contributions of civil society seems to have been. Instead of offering more concrete guidance – which could have the potential to break potential deadlocks – most of civil society seems to have used their brief time on stage to express general support for the accession project as such. In my view, this is not where the problems lie.

As Opinion 2/13 so well illustrates, the only opponent of ECHR accession is the CJEU. All others – all EU Member States, all the institutions, and the non-EU CoE Member States – unanimously agreed on the 2013 DAA. And their positions have not shifted. That makes this (re-)negotiation process an unique absurdity: all those at the negotiation table are in agreement on a text, which they must now adapt to meet the (still somewhat unclear) preferences of a third party, the CJEU.

Next meetings

Two further negotiation meetings of the 47+1 group have already been penciled in for 2021:

8th negotiation meeting (2–4 February 2012)

9th negotiation meeting (23–25 March 2021)

Do note that these dates are subject to confirmation in light of the ongoing COVID pandemic. 

Barnard & Peers: chapter 8

Photo credit:

Wednesday 27 January 2021

The Brexit deal - Council legal service opinion

Here's the EU Council legal service's opinion on whether the Brexit deal can be concluded by the EU alone, or requires also the participation of its Member States (known as a 'mixed agreement'). It concludes that it can be concluded by the EU alone, because it is a mixture of competences exclusive to the EU and competences shared between the EU and its Member States. Following the case law of the CJEU, it's an option for the EU to conclude a treaty including provisions on shared competence either by the EU alone, or together with its Member States.  (The main points are also summarised in a Twitter thread). 


I.              Introduction

1.             At the meeting of Coreper on 23 November 2020, the representative of the Council Legal Service (CLS) made an oral intervention on the legal nature of the future agreement with the United Kingdom (UK) which was being negotiated, and more particularly on the issue of mixity and on the possible EU-only nature of that future agreement, through the exercise by the EU of its so-called potential competence. In the meantime, the Trade and Cooperation Agreement between the EU and Euratom, of the one part, and the UK, of the other, was signed on 30 December 2020 (the Trade and Cooperation Agreement).

2.             In reply to a request by several delegations, in particular at the meeting of the UK Working Party on 13 January 2021, this opinion confirms and develops in writing the answers already provided orally by the CLS. It focuses more particularly on the issue of the EU exercising externally its potential competence - i.e. the exercise of its competence in areas of shared competence which are not already subject to common rules within the meaning of Article 3(2) TFEU and the related case-law - and the legal consequences of such an exercise of competences.

3.             In view of the short time available in the ongoing conclusion process, this opinion focusses on the Trade and Cooperation Agreement and does not provide an in-depth examination of all its aspects, nor does it provide a comprehensive and detailed competence analysis.

II.           Factual and Legal BACKGROUND

4.             On 25 February 2020, the Council adopted its decision[1] authorising the opening of negotiations for a future agreement with the UK, to which the negotiating directives were annexed[2]. This decision also nominated the Commission as the EU negotiator. In parallel, the Council and the Representatives of the Governments of the Member States recorded a statement[3] to the Council minutes. In this statement, the Member States' representatives authorised the Commission to conduct negotiations in areas of the future relationship that fall within their competences and stated that the question whether the new agreement would be concluded by the EU or by the EU and its Member States remained to be determined at the end of the negotiations[4]

5.             The negotiations, which were completed on 24 December 2020, resulted in three agreements: the Trade and Cooperation Agreement; the Agreement between the EU and the UK concerning security procedures for exchanging and protecting classified information, which supplements the Trade and Cooperation Agreement; and an Agreement between the Government of the UK and Euratom for Cooperation on the Safe and the Peaceful Uses of Nuclear Energy. As indicated above, this opinion focusses on the first Agreement.

6.             Under its first Article, the Trade and Cooperation Agreement "establishes the basis for a broad relationship between the [EU and the UK] within an area of prosperity and good neighbourliness characterised by close and peaceful relations based on cooperation, respectful of the Parties' autonomy and sovereignty"[5]. When deciding to sign the Agreement, the Council characterised it as "[establishing] the basis for a broad relationship between the [EU and the UK] involving reciprocal rights and obligations, common actions and special procedures" which is the wording from Article 217 TFEU on association agreements, on the basis of which the Council chose to sign the Trade and Cooperation Agreement with the UK[6].

7.             The Trade and Cooperation Agreement with the UK sets out arrangements in a vast array of areas such as trade in goods, services, investments, digital trade, capital movements and payments, intellectual property, public procurement, energy, aviation and road transport, fisheries, social security coordination, the absence of visas for short term visits (Part Two of the Agreement), law enforcement and judicial cooperation in criminal matters (Part Three), thematic cooperation (health and cybersecurity) (Part Four) and participation in EU programmes (Part Five).

It is underpinned by provisions ensuring a common institutional framework (Part One of the Agreement), including a dispute settlement mechanism (Part Six), and a level playing field between the Parties. There are 49 annexes and three protocols to the Agreement : a Protocol on administrative cooperation and combating fraud in the field of value added tax and on mutual assistance for the recovery of claims relating to taxes and duties; a Protocol on mutual administrative assistance in customs matters; and a Protocol on social security coordination.

The Agreement provides in its second Article that future bilateral agreements between the EU and the UK "shall constitute supplementing agreements to this Agreement, unless otherwise provided for in those agreements" and that "such supplementing agreements shall be an integral part of the overall bilateral relations as governed by this Agreement and shall form part of the overall framework"[7].

8.             On 29 December 2020, the Council adopted Decision No (EU) 2020/2252, on the signing, on behalf of the Union, and on provisional application of the Trade and Cooperation Agreement and of the Security of Information Agreement ('the decision on signature')[8].

9.             Through this decision, the Council, "in view of the exceptional and unique character of the Trade and Cooperation Agreement, which is a comprehensive agreement with a country that has withdrawn from the Union (…) [decided] to make use of the possibility for the Union to exercise its external competence with regard to the [UK]" (recital 6). The Council hence made the political choice that the Trade and Cooperation Agreement was to be concluded as an EU-only agreement. This is explicitly indicated in the decision on signature.

The reasons for this political choice are explained in recital 6 - "in view of the exceptional and unique character of the [Agreement]" - but also, indirectly, in recital 16 of the decision on signature which explains the reasons for the Council deciding to provisionally apply the Agreement by the fact that that it concerns "a country that has withdrawn from the Union", the UK being therefore "in a different and exceptional situation with regard to the Union compared to other third countries". The level of cooperation between the EU and the UK was indeed going to decrease from a very high level at the end of the transition period - during which the EU acquis still applied to the UK - to a lower level of cooperation as from the end of that period, thus causing disruption the severity of which could be limited through provisional application of the Trade and Cooperation Agreement.

10.         The consequences, and limits, of having made the above political choice are expressly set out in recital 15 and in Article 10 of the decision on signature where it is stated that "the exercise of Union competences through the Trade and Cooperation Agreement shall be without prejudice to the respective competences of the Union and of the Member States in any ongoing or future negotiations for, or signature or conclusion of, international agreements with any other third country, or in relation to any future negotiations for, or signature or conclusion of, any supplementing agreements [to the Trade and Cooperation Agreement]".

11.         On 25 December 2020, the Commission submitted a proposal for a Council decision on the conclusion of the Trade and Cooperation Agreement[9]. The discussions within the Council on this proposal are ongoing.

III.        Legal Analysis

12.         The views of the CLS have been sought in relation to the legal nature of the Trade and Cooperation Agreement, and more specifically, on the issue of the EU exercising its potential competence, on the issue of the EU-only nature of the Agreement and on the issue of the legal consequences for Member States of such an exercise of competences. As indicated in paragraph 3 above, the CLS will not examine each title of the agreement or proceed to a comprehensive and detailed competence analysis.

A.           EU exclusive and shared competences, as interpreted by the Court of Justice

13.         In accordance with Article 3(1) TFEU, the EU has exclusive competence, in particular, in the following areas: customs union, the establishment of competition rules necessary for the functioning of the internal market, the conservation of marine biological resources under the common fisheries policies, and the common commercial policy.

14.         In accordance with Article 3(2) TFEU, the EU "also [has] exclusive competence for the conclusion of an international agreement (…) in so far as its conclusion may affect common rules or alter their scope". As clarified by the Court in its judgment in the Broadcasting Convention Case[10], this last limb of Article 3(2) codifies the so-called AETR, or ERTA, jurisprudence[11].

15.         On shared competences, Article 4(1) and (2) TFEU provides the following:

"1. The Union shall share competence with the Member States where the Treaties confer on it a competence which does not relate to the areas referred to in Articles 3 and 6.

2.  Shared competence between the Union and the Member States applies in the following principal areas:

(a) internal market; (…)

(e) environment;

(f)  consumer protection;

(g) transport; (…)

(i)  energy

(j)  area of freedom, security and justice

(k) common safety concerns in public health matters (…)"

16.         The Court has consistently held that there is a risk that common EU rules may be affected by commitments undertaken by Member States, or that the scope of those rules may be altered, such as to justify an exclusive external competence of the EU under Article 3(2) TFEU, where those commitments fall within the scope of the said common EU rules[12].

17.         A finding that there is such a risk does not presuppose that the area covered by the international commitments (whether actual or envisaged) and that covered by the EU rules coincide fully. In particular, the scope of common EU rules may be affected or altered by international commitments where those commitments fall within an area which is already largely covered by such rules. Furthermore, such a risk of common EU rules being affected may be found to exist where the international commitments at issue, without necessarily conflicting with those rules, may have an effect on their meaning, scope and effectiveness[13].

When analysing the relationship between the international instrument at stake and the relevant EU rules, that analysis must take into account the areas covered, respectively, by the rules of EU law and by the provisions of that instrument, their foreseeable future development and the nature and content of those rules and those provisions, in order to determine whether the said instrument is capable of undermining the uniform and consistent application of the EU rules and the proper functioning of the system which they establish[14].

B.            Consequences of EU exclusive competences on the Member States

18.         The exercise internally, by the EU, of its competence has consequences on the Member States insofar as, if the EU thereby acquires exclusive competence, Member States may no longer enter into international commitments outside the framework of the EU institutions in the areas falling within EU exclusive competence[15].

19.         Conversely, internally, where the EU has not (yet) exercised its shared competences in a given area, the Member States can continue exercising their competences in that area "to the extent that the Union has not exercised its competences" (Article 2(2) TFEU). The same is true, externally, where it would result from the analysis of an (envisaged) international commitment that the conditions for exclusivity set out in Article 3(2) TFEU, as interpreted by the Court in its extensive case-law on external competences (see paragraphs 16 and 17 above), are not met and that, as a consequence, Member States can still exercise their competences externally.

20.         By way of example, the EU has not yet adopted common internal rules with regard to air traffic rights granted to third countries. The competence to conclude agreements with third countries on such matters has therefore not become an exclusive competence of the EU[16] and can be exercised either by the Member States or by the EU[17].

C.           Mixity of international agreements: obligatory or facultative

21.         It is recalled that, in accordance with the principle of conferral (Article 5 TEU), an international agreement is mixed, i.e. it is signed and concluded both by the EU and its Member States, if it concerns competences that belong both to the EU and to its Member States.

22.         There are, in practice, two types of mixity: obligatory or facultative.

Mixity is obligatory where, in addition to areas of EU competence, the envisaged agreement covers one or several areas that fall outside EU competences, i.e. where the Treaties have not conferred competences on the EU in that particular area. In such a case, there is no political choice: the agreement must be concluded both by the EU and its Member States.

Mixity is facultative where the envisaged agreement covers one or several areas where the EU has shared competences which are potential, i.e. not yet exercised or not yet covered by EU common rules regarding which the envisaged agreement would have consequences as mentioned in Article 3(2) and the related case-law. In such a case, the agreement may be concluded either by the EU and its Member States or by the EU alone. Depending on whether the Council decides to exercise all the EU potential competences or not, the agreement will be an EU-only or a mixed agreement. This is a political choice to be made by the Council on the basis of the relevant Treaty provisions which confer competence on the EU.

23.         In its Singapore FTA Opinion,[18] the Court provided clarifications as to the division of competence between the EU and the Member States in the field of trade and investment. The Court concluded that most of the Free Trade Agreement with Singapore fell within the exclusive competence of the EU either because it was covered by the Common Commercial Policy, including on foreign direct investment, as defined in Article 207(1) TFEU, or because it was covered by the Common Transport Policy (Articles 91 and 100(2) TFEU)[19].

24.         In the same Opinion, the Court recalled that foreign direct investment is an exclusive competence of the EU. However, to the extent that the FTA provisions related to indirect investment (i.e. portfolio investment), the competence for that was "shared between the European Union and the Member States pursuant to Article 4(1) and (2)(a) TFEU"[20]. On the possible exercise of such potential shared competence, the Court clarified in its judgment in Weddell, that "(…) the mere fact that international action of the [EU] falls within a competence shared between it and the Member States does not preclude the possibility of the required majority being obtained within the Council for the [EU] to exercise that external competence alone"[21].

25.         In the case of facultative mixity, where the EU has competence for the matters covered by an agreement, of which at least some fall within its potential competence, that potential competence can still be exercised by the Member States if they wish. The Council may however decide, for a particular agreement, to exercise the potential EU competence on the basis of the relevant Treaty legal basis,[22] in accordance with the voting rules provided therein. Exercising or not the EU potential competence externally when concluding an agreement is a matter of political choice for the Council[23].

26.         This is to be distinguished from cases of obligatory mixity where the subject matter of an agreement partially covers matters for which the EU has competence, and partially matters for which the EU has no competence whatsoever. In such a situation, as said above in paragraph 22, mixity is not a political choice but a legal obligation[24].

27.         It is recalled that the conclusion of mixed agreements presents procedural and political complexity as the process relating to the conclusion of recent mixed agreements testifies. Indeed, the entry into force of a mixed agreement requires ratification not only by the EU (conclusion by the Council, usually after obtaining the consent of the European Parliament), but also by all its Member States in accordance with their constitutional requirements. As a consequence, in principle, the EU will wait for ratification by all Member States before ratifying itself. Such a process may take a long time and depends on whether the ratification runs smoothly or not in all Member States. By contrast, in order to enter into force, an 
EU-only agreement needs only to be ratified by the EU, and can in practice be concluded within a shorter period of time than a mixed agreement.

D.           The particular case of the trade and cooperation agreement with the UK

28.         While not entering into a detailed examination of its different Titles and provisions, a rapid examination of the Trade and Cooperation Agreement shows that no situation of obligatory mixity arises: the EU has competence in all the fields covered by it.

29.         The CLS recalls in this context that, by way of example, provisions related to trade or fisheries contained in Heading One of Part Two (Trade) and Heading Five of Part Two (Fisheries) are exclusive competences of the EU by virtue of Article 3(1) TFEU. Other provisions of the Agreement, for instance, Title II of Heading Two (Aviation safety) or of Part Three (Law enforcement and judicial cooperation in criminal matters), cover matters that have become exclusive by exercise or are largely covered by EU acquis that will be or risk being affected by the Agreement[25].

30.         Conversely, there are certain other provisions, for instance traffic rights in the aviation area, which belong to shared EU competences not yet exercised internally, and which are therefore only potential EU competences. In relation to these potential competences, the Council could decide, when adopting the decision on signature, that the EU would exercise this type of non‑exclusive potential competences, thus making the Trade and Cooperation Agreement an EU‑only agreement.

31.         When it adopted the decision on signature on 29 December 2020, the Council made this political choice and decided to exercise the EU competence in the areas of potential EU competence. That decision produces legal effects not least because it has entered into force on the day of its adoption by the Council (29 December 2020) and the Trade and Cooperation Agreement is being provisionally applied since 1 January 2021. It follows that the Agreement must also be concluded as an EU-only agreement.

32.         As indicated above in paragraphs 21 to 25, only where the EU is vested with the competence to do so in the Treaties, can it conclude an international agreement, in accordance with the relevant legal basis. In the case at stake, the Trade and Cooperation Agreement was signed and provisionally applied on the basis of Article 217 TFEU, as the substantive legal basis, in conjunction with the relevant procedural legal bases (Article 218(5) and (8) TFEU).

33.         Article 217 TFEU empowers the EU to conclude with a third country an agreement establishing an association involving reciprocal rights and obligations, common action and special procedure. The areas covered by such an agreement should be within the limits of the powers that the Member States have conferred on the EU in the Treaties to attain the objectives set out therein.[26] Article 217 TFEU can be used whenever there is EU competence - i.e. where the Treaty confers the appropriate competence on the EU in the different areas covered by the agreement - even if this competence has not been exercised fully or is only potential. Article 217 TFEU however is not to be used as a legal basis if there is no underlying competence, i.e. the underlying sectoral competence must exist for Article 217 TFEU to be used as a legal basis.

34.         Article 217 TFEU allows the EU to conclude, by unanimity, a wide ranging agreement on matters of EU competence without the need to identify in detail the areas where the EU has already exercised or not its competence. It can include areas of EU competence where the sectoral legal basis requires unanimity or qualified majority voting, as well as areas of potential EU competence not yet exercised internally.

E.            Exercise vis-à-vis the UK of the EU shared competence in social security coordination and aviation traffic rights

35.         By way of example, the EU has a shared competence in the area of social security coordination (Articles 48 TFEU). The EU has concluded several agreements with third countries that contain rules on the coordination of social security. This is typically the case for association agreements based on Article 217 TFEU. The lack of completion of free movement of persons is not a hindrance to the conclusion of an EU agreement in the field of social security coordination[27]. So far, agreements with third countries covering also the area of social security coordination have been generally concluded as mixed agreements. However, given that the EU has competence in this area, this is a matter of political choice. It is equally possible for the EU to choose to exercise its competence externally and to conclude such an agreement as an EU-only agreement.

36.         Similarly, the EU has shared competence in the area of air transport (Articles 91 and 100(2) TFEU). Once and to the extent that the EU exercises internally such shared competence, it becomes exclusively competent externally for matters affecting those internal rules. As the EU has not yet exercised this shared competence internally with regard to traffic rights granted to third countries, agreements with third countries on such matters are often concluded as mixed agreements (facultative mixity). The Council can choose whether or not to use it externally.

F.            Effect for Member States of the EU exercising its shared (potential) competence

37.         The external exercise of the above EU competences with regard to a given third country does not prevent Member States from exercising their competence on that same matter vis-à-vis other third countries. To take an example, the EU has an aviation agreement with Switzerland that has existed for more than 20 years[28] and which covers traffic rights. The existence of this agreement has not prevented the Member States from concluding agreements on traffic rights with other third countries.

38.         Hence, the fact that the Council has decided to opt, in the specific case of the Trade and Cooperation Agreement with the UK, for an EU-only agreement does not prevent Member States from continuing to exercise their national competences vis-à-vis other third countries in that same area of potential EU competence. Exercise by the EU of its potential competences in the concerned area vis-à-vis the UK does not trigger a situation of exclusivity as regards its relations with other third countries nor a situation of exclusivity for the area of traffic rights concerning other third countries as if that competence had been exercised internally. Therefore, Member States remain free to continue concluding international agreements in these areas of shared competence with third countries other than the UK under the same conditions as before the signature of the Trade and Cooperation Agreement. As explained above in paragraph 10, this possibility is explicitly confirmed in recital 15 and Article 10 of the decision on signature.

G.           Provisions on possible so-called "top-ups" by Member States

39.         Moreover, the Trade and Cooperation Agreement provides for, or does not exclude, the possibility for Member States to enter into bilateral agreements with the UK concerning specific matters covered by the Agreement in the areas of air transport, administrative cooperation in the field of customs and VAT and social security coordination.[29] Member States may do so provided such agreements are compatible with EU law, do not undermine the functioning of the Agreement and are otherwise compatible with the conditions set out in Articles 6 to 8 of the decision on signature, which foresees an internal mechanism of information and cooperation between the Member States and the Commission, culminating with the possibility of authorising bilateral arrangements or agreements that Member States would conclude with the UK in those areas.

40.         This internal mechanism is an expression of the duty of sincere cooperation incumbent on the Member States (Article 4(3) TEU)[30], which is of general application and does not depend on whether the competence concerned is exclusive[31]. On this basis, Member States have a duty to refrain from any action which could jeopardise the attainment of the EU objectives, and to ensure that such arrangements or agreements are compatible and do not undermine the functioning of the Trade and Cooperation Agreement. To the extent that it frames and organises the possibility of certain bilateral agreements supplementing the Trade and Cooperation Agreement (so-called 'top-ups') as allowed or not prohibited by the Agreement itself, this internal mechanism is also an expression of the fact that the Agreement is part of EU law, is binding in accordance with Article 216(2) TFEU, and has therefore primacy.

41.         The existence of the internal mechanism is independent of the nature of the competence at stake. To take the example of traffic rights, the EU is exercising in the Trade and Cooperation Agreement its external competence on certain traffic rights vis-à-vis the UK. Certain bilateral agreements supplementing the Trade and Cooperation Agreement are permitted by the Agreement itself in accordance with the conditions set out therein[32]. The Agreement itself specifically prohibits further top-ups (Article AIRTRAN.23)[33]. The internal empowerment mechanism in Article 6 of the decision on signature regulates how the permitted top-ups are going to be authorised. Therefore, at least as far as top-up agreements concerning air traffic rights are concerned, the authorisation mechanism is not only an expression of the duty of sincere cooperation. It is also necessary because the shared, previously unexercised, external competence on traffic rights vis-à-vis the UK is now governed by the provisions of the Trade and Cooperation Agreement which has primacy, and is a matter of exclusive EU competence vis-à-vis the UK. However, as stated above in paragraphs 37 to 38 and explicitly confirmed in recital 15 and Article 10 of the decision on signature, that EU competence is not exclusive vis-à-vis other third countries.

42.         Lastly, the CLS takes this opportunity to clarify, as it did in Coreper on 22 January 2021, that contrary to the statement made by the Commission to the minutes of the Council on 29 December 2020, in connection with the adoption of the decision on signature,[34] it does not see any legal obstacle for a legal act based on Articles 217 and 218 TFEU, such as the decision on signature, or the future Council decision concluding the agreements, to contain such an internal mechanism for authorisation of bilateral arrangements or agreements between individual Member States and the UK. Such internal authorisation or empowerment mechanisms may be set out either in a legal act adopted on the basis of the relevant sectoral substantial legal basis (i.e. a legislative act) or in a legal act adopted by the Council for signing and concluding international agreements[35].

IV.        Conclusion

In conclusion, the Council Legal Service confirms its view that, as it only covers areas where the EU has competence, whether exclusive or potential, the Trade and Cooperation Agreement may be concluded as an EU-only agreement on the basis of Article 217 TFEU. The Council decided to make this choice when it adopted the decision on signature on 29 December 2020.

[1]           Council Decision (EU, Euratom) 2020/266 of 25 February 2020 authorising the opening of negotiations with the United Kingdom of Great Britain and Northern Ireland for a new partnership agreement (OJ L 58, 27.2.2020, p. 53).

[2]           Doc 5870/20 ADD 1 REV 3, Annex to Council Decision authorising the opening of negotiations with the United Kingdom of Great Britain and Northern Ireland for a new partnership agreement, 25 February 2020.

[3]           Annex B of Doc. 6239/20.

[4]           "[T]he Member States' representative authorise the Commission to conduct negotiations in areas (…) that fall within the competences of the Member States (…)" and "the question of whether the (…) agreement will be concluded by the Union or by the Union and its Member States will be determined at the end of the negotiations." (emphasis added).

[5]              See Article COMPROV.1 (Purpose) of the Agreement. This wording is similar to that in Article 8(1) TEU on the EU's relationships with neighbouring countries.

[6]              See recital 5 of Council Decision (EU) 2020/2252 of 29 December 2020 on the signing, on behalf of the Union, and on provisional application of the Trade and Cooperation Agreement between the European Union and the European Atomic Energy Community, of the one part, and the United Kingdom of Great Britain and Northern Ireland, of the other part, and of the Agreement between the European Union and the United Kingdom of Great Britain and Northern Ireland concerning security procedures for exchanging and protecting classified information (OJ L 444, 31.12.2020, p. 2).

[7]           See Article COMPROV.2 (supplementing agreements).

[8]           See above footnote 7.

[9]           COM (2020) 856 final.

[10]          Case C‑114/12, Commission v. Council ('Broadcasting Convention'), judgment of the Court (Grand Chamber) of 4 September 2014, EU:C:2014:2151, paragraphs 66 and 67. See also Opinion 1/13 of the Court of Justice (Grand Chamber) of 14 October 2014, Child Abduction Convention, EU:C:2014:2303, paragraphs 69 to 74.

[11]          Case 22/70, Commission v Council ('ERTA'), judgment of 31 March 1971, EU:C:1971:32, paragraphs 17 to 19: "17. In particular, each time the Community, with a view to implementing a common policy envisaged by the Treaty, adopts provisions laying down common rules, whatever form these may take, the Member States no longer have the right, acting individually or even collectively, to undertake obligations with third countries which affect those rules. 18. As and when such common rules come into being, the Community alone is in a position to assume and carry out contractual obligations towards third countries affecting the whole sphere of application of the Community legal system. 19. With regard to the implementation of the provisions of the Treaty, the system of internal Community measures may not therefore be separated from that of external relations."

[12]          Opinion 2/91 of the Court of Justice of 19 March 1993, ILO, EU:C:1993:106, paragraph 25; Case C-467/98, Commission v Denmark ('Open Skies'), judgment of the Court of 5 November 2002, EU:C:2002:625, paragraph 82; and Opinion 1/03 of the Court of Justice (Full Court) of 7 February 2006, Lugano Convention, EU:C:2006:81, paragraphs 120 to 126.

[13]          See Opinion 2/91 ILO (op. cit. footnote 13), paragraphs 25 and 26; Judgment in 'Open Skies' (op. cit. footnote 13), paragraph 82; Opinion 1/03 Lugano Convention (op. cit. footnote 13), paragraphs 120 and 126; Judgment in 'Broadcasting Convention' (op. cit. footnote 11), paragraphs 68 to 73; as well as Joined Cases C‑626/15 and C‑659/16, Commission v Council ('Weddell') judgment of the Court (Grand Chamber) of 20 November 2018, EU:C:2018:925, paragraphs 113 and 114 and the case-law cited.

[14]          See Opinion 3/15 of the Court of Justice (Grand Chamber) of 14 February 2017, Access Rights, EU:C:2017:114, paragraph 108 and the case-law cited.

[15]             Opinion 2/91 ILO (op. cit. footnote 13), paragraphs 25 and 26; Judgment in 'Open Skies' (opt. cit. footnote 13), paragraph 82.

[16]             Judgment in 'Open Skies' (opt. cit. footnote 13), paragraphs 90 to 92.

[17]             See CLS opinion in 5990/18 on the Regulation on competition in air transport, paragraphs 23 and 24.

[18]          Opinion 2/15 of the Court of Justice (Full Court) of 16 May 2017, Singapore FTA, EU:C:2017:376.

[19]          In the case of transport services, the Court deduced this from its AETR case law: see Opinion 2/15 Singapore FTA (op. cit. footnote 19), paragraph 170 and further. Note however that air transport services were not covered in the Singapore agreement, see paragraph 63 of that Opinion.

[20]          Opinion 2/15 Singapore FTA (op. cit. footnote 19), paragraph 243.

[21]          Judgment in 'Weddell' (opt. cit. footnote 14), paragraph 126.

[22]          See CLS opinion in 12866/19 on the CETA Investment Court System (ICS), paragraph 6.

[23]          See Case C-600/14, Germany v. Council ('OTIF'), judgment of the Court (Grand Chamber) of 5 December 2017, EU:C:2017:935, paragraph 68; see also Judgment in 'Weddell' (opt. cit. footnote 14), paragraphs 126 and 127.

[24]          See Opinion 2/15 Singapore FTA (op. cit. footnote 19), paragraph 292. See also CLS opinions in 12866/19 (CETA ICS) and 6442/19 (UN Convention on Investor-State Arbitration).

[25]          See, inter alia, Regulation (EU) 2018/1139 of the European Parliament and of the Council of 4 July 2018 on common rules in the field of civil aviation and establishing a European Union Aviation Safety Agency (OJ L 212, 22.8.2018, p. 1) and Directive (EU) 2016/681 of the European Parliament and of the Council of 27 April 2016 on the use of passenger name record (PNR) data for the prevention, detection, investigation and prosecution of terrorist offences and serious crime (OJ L 119, 4.5.2016, p. 132).

[26]          See Case C-81/13, UK v Council ('social security Turkey'), judgments of the Court (Grand Chamber) of 18 December 2014, EU:C:2014:2449, paragraphs 61 and 62.

[27]             As regards Turkey, for example, the Court compensated for the lack of completion of free movement with the addition of Article 217 TFEU to Article 48 TFEU as a substantial legal basis for the adoption of the EU position, to be taken within the Association Council set up by the EU-Turkey Association Agreement, with regard to the adoption of provisions on the coordination of social security systems (see judgment in 'social security Turkey' (op. cit. footnote 27), paragraph 63.

[28]          Agreement between the European Community and the Swiss Confederation on Air Transport, signed on 21 June 1999 (OJ L 114, 30.4.2002, p. 73).

[29]          See Article AIRTRN.3 and Article 41 of the VAT Protocol.

[30]          Article 4(3) TEU: "Pursuant to the principle of sincere cooperation, the Union and the Member States shall, in full mutual respect, assist each other in carrying out the tasks which flow from the Treaties. The Member States shall take any appropriate measure, general or particular, to ensure fulfilment of the obligations arising out of the Treaties or resulting from the acts of the institutions of the Union. The Member States shall facilitate the achievement of the Union's tasks and refrain from any measure which could jeopardise the attainment of the Union's objectives."

[31]          C-246/07, Commission v Sweden, judgment of the Court (Grand Chamber) of 20 April 2010, EU:C:2010:203, paragraph 71.

[32]          Article AIRTRN.3(5) TCA: "the rights mutually granted in accordance with paragraph 4 (i.e. top-ups) shall be governed by the provisions of this Title."

[33]          "The United Kingdom and a Member State may not grant each other any rights in connection with air transport to, from or within their respective territories other than those expressly laid down in this Title, save as provided for in Article AIRTRN.3 (4) and (9) [traffic rights]."

[34]          Doc. 5525/20 ADD 1.

[35]          The CLS recalls that Article 4 of the Council Decision on the conclusion of the Withdrawal Agreement establishes a similar internal mechanism in relation to agreements in areas of EU competence that certain Member States are allowed, under the conditions referred in that provision, to conclude with the UK (see Council Decision (EU) 2020/135 of 30 January 2020 on the conclusion of the Agreement on the withdrawal of the United Kingdom of Great Britain and Northern Ireland from the European Union and the European Atomic Energy Community, OJ L 29, 31.1.2020, p. 1).