’In view of the exceptional and unique character’ of the EU-UK Trade and Cooperation Agreement – an Exception to Separation of Powers within the EU?


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 (EU-UK TCA). At the stage of opening the negotiations, as is usual, the option of concluding it as mixed agreement was left open. However, the TCA was signed as an EU-only agreement, based on the argument that the EU had exclusive competence for part of the agreement (Common Commercial Policy (CCP)) and ‘potential competence’ (shared competence that the EU exercises for the first time) for the rest. At present, the EU-UK TCA is provisionally applied and the European Parliament is planned to give its consent in the plenary session starting on 26 April 2021.

The decision to sign the agreement as an EU-only agreement was taken in a hasty written procedure between Christmas and New Year. The Council relied on a confidential but leaked opinion by the Council Legal Service (CLS), which ‘does not provide an in-depth examination of all of its aspects, nor does it provide a comprehensive and detailed competence analysis.’ The Opinion is indeed strategically selective. At the same time, the Member States, who were not involved in the negotiations, are unlikely to have had the necessary time to analyse the competence question in any depth. To what extent their parliaments – whose prerogatives are fundamentally affected by the choice for an EU-only agreement – were involved in approving the Council decision depends on national constitutional provisions and the limited time window. 

The EU-only nature of the agreement, or if you like, the choice against mixity strikes us as an unusual choice. Agreements negotiated by the EU that include provisions outside its exclusive competences are generally concluded as mixed agreements. 

The EU-UK TCA departs from this practice by referring to the exercise of a ‘potential non-exclusive competence’ by the Union. The CLS explains this as a ‘political choice’ that ‘does not prevent Member States from continuing to exercise their national competences vis-à-vis other third countries’. We doubt whether the matter really is this simple. 


The CLS opinion states that ‘the conclusion of mixed agreements presents procedural and political complexity as the process relating to the conclusion of recent mixed agreements testifies’. It is evident that there was no appetite for a new CETA, which has been provisionally applied since 2017 but has not yet entered into force.

However, this argument is problematic. Under the Court’s settled case law, it is not a matter of simpler procedure, but one of substance whether an agreement should be concluded as EU-only or mixed agreement. 

Mixed agreements are a conscious means of leaving the division of competences undetermined. Adopting an agreement as ‘EU-only’, by contrast, explicates the competence division in favour of the EU.

While the matter is presented as a ‘political choice’, it has constitutional implications – for the dynamic competence division between the EU and its Member States and for separation of powers between the branches within the EU.

The European Parliament has the power of consent both for EU-only and mixed agreements, but, in the context of the latter, formally only for the part covered by EU competences. Yet, in the context of EU-only agreements, the EP gains visibility and becomes the focus of lobbying and political struggle. 

The real losers in this game are, however, national parliaments. They lose public engagement and their constitutional prerogative to approve the parts that fall under shared competences and that up to the point of conclusion of the agreement were governed by national law. This cannot be compensated by their indirect control of national representatives in the Council. This is also why national parliaments should have been given time to enable an informed debate about the solution and its direct and indirect consequences. 

Exceptional Circumstances Justify Unusual Choices?

The agreement builds on the idea that it is of an ‘exceptional and unique character’. The Commission emphasises that the agreement ‘goes beyond traditional free trade agreements and provides a solid basis for preserving our longstanding friendship and cooperation’. The Council agrees. But how exceptional and unique is the EU-UK agreement? 

Certainly, the UK is the only state that has (so far) withdrawn from the EU. However, the EU has concluded numerous association agreements with the intention of ‘establishing association involving reciprocal rights and obligations, common action and special procedure’ (Article 217 TFEU). 

The choice to conclude the TCA as an EU-only agreement is unusual considering that large parts of the agreement fall under shared competences. It has been justified by reference to the Court opinion relating to the EU-Singapore free trade agreement (FTA). 

However, the two agreements are very different. The EU-UK TCA is more comprehensive and envisages a deeper form of cooperation, reaching far beyond the Court’s very wide interpretation of CCP post-Lisbon. This is confirmed by Article 217 TFEU as its legal basis. For the Commission, ‘[t]his legal basis is the most appropriate given the broad scope of the envisaged partnership’. Yet, association agreements are almost always concluded as mixed agreements.

Compared to the EU-Singapore FTA, for example, the association agreements with Ukraine, Moldova and Georgia include a wider range of provisions, which go beyond the scope of CCP. These agreements, just as the EU-UK TCA, cover several areas that fall outside of the EU’s exclusive competences, such as security and police cooperation, and have thus been concluded as mixed agreements. The EU-UK TCA excludes areas such as migration and energy, that are routinely covered by association agreements. 

For air travel, the agreement does not follow the usual pattern of concluding such provisions as part of a mixed agreements. In line with existing case law, the CLS points out that the EU has not yet exercised the shared competence internally with regard to traffic rights granted to third countries, but can now choose to exercise this competence externally. The CLS does not expand on the motivation or implications of changing the earlier pattern.

The EU-UK TCA has a long list of provisions on law enforcement and judicial cooperation in criminal matters (part three). In this area, relying on the ‘former EU state’ argument is particularly odd, because the UK had a comprehensive opt out and it has participated in selected instruments only. 

This part of the agreement is also particularly significant since it concerns constitutional issues, such as surrendering people, the (automated) transfer of certain data related to law enforcement (DNA profiles, vehicle registrations), and the use of operative powers by authorities of a third state (the UK), which is no longer bound by the relevant EU legal acts or the Charter of Fundamental Rights. Law enforcement is an example of an area that falls squarely within the competences of the Member States, illustrated by the many TCA references to the competent law enforcement authorities of the Member States. The CLS opinion does not offer an analysis of these provisions or their relationship to EU law or competence. 

EU Competence, Exclusivity and Preemption

The CLS simply stated that ‘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.’ This approach differs significantly from the Court’s approach, for example in Opinion 1/13. Here, the ECJ first established the existence of a Union competence, by asserting that the Union had internal competence and that it had exercised this competence. The Court then established the exclusive nature of the EU competence by carrying out a detailed analysis of The Hague Convention and internal regulation to determine whether the former could be said to ‘affect’ or ‘alter the scope’ of the latter.

Article 3(2) TFEU sets out that 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’. The Court interprets this risk quite generously, including to cover situations when the commitments threaten effectiveness of common EU rules. 

We agree that in many areas covered by the TCA, EU internal rules exist and hence exclusivity under Article 3(2) TFEU could potentially be argued. However, any such risk can logically only be established by reference to common EU rules. The CLS does not bother with the second step of analysis (exclusivity), but is satisfied with existence of any category of competence in a given policy area.

Conspicuous absence of references to EU legislation

It is true that, as a former Member State, the UK is at the moment of signing subject to the applicable EU legislation. It would logically follow that in such a case existing EU law and the current competence division are particularly significant. 

The peculiarity with the EU-UK TCA is, however, that in many areas it does not include substantive provisions or follow EU legislation. Instead, it is anchored into various international instruments. 

The provisions on the environment and climate change illustrate how the TCA does not follow applicable EU legislation. Instead, the TCA relies on international arrangements, such as the Paris Agreement in the context of climate change. However, these international references expose the oddity of the EU-only choice for the EU-UK TCA from a different angle: the Paris Agreement, as other environmental conventions mentioned, are mixed agreements. 

In other areas, the agreement relies on instruments of the Council of Europe including on Mutual Assistance in Criminal Matters (1959), Suppression of Terrorism (1977), and Extradition (1957). They are ratified by Member States, often with reservations, and fall under national competence and have been given effect through national legislation. 

The EU has also approved extensive legislation governing money laundering, based on its internal market competence and thus until withdrawal binding on the UK. However, instead of this legislation, the agreement relies on two Council of Europe conventions (1990, 2005). The more recent one has not even been ratified by all Member States. Yet, the provisions of the TCA are also to be applied ‘in place of’ certain provisions and replace some of the Chapters in these Conventions.

The reason for quoting international conventions instead of EU law seems clear – if the agreement relied on EU law, it would need to protect the autonomy of EU law, including the position of the ECJ. Since the UK was reluctant to accept the ECJ’s jurisdiction, dispute settlement takes place in Partnership Council or consultations in Specialist Committees. If no solution is found, parties can take the matter to arbitration. This has been the incentive to distance the agreement from EU law and create a ‘parallel universe’ of international law. However, what does this mean for the legal status of the mentioned Conventions? How is the competence division affected by the arrangement? 

Consequences of EU-only agreement for the division of powers between the EU and its Member States 

Two immediate and significant consequences follow from the EU-only nature of the TCA: First, the EU is represented in the TCA institutions. Member States have no direct role. Second, Member States are seriously limited in their capacity to make bilateral arrangements with the UK in matters falling under the scope of the agreement. 

As all EU-only and mixed agreements, the EU-UK TCA is binding on the Member States under Article 216(2) TFEU and has, together with the duty of sincere cooperation in Article 4(3) TEU, far reaching restraining effects on the Member States. Member States have a duty to refrain from any action which could jeopardise the attainment of the EU objectives and to exercise their powers, including in areas of reserved national competences, in compliance with the international obligations flowing from international agreements of the Union. 

The Council decision on signature illustrates these constraints on Member States for the EU-UK TCA. This internal notification mechanism requires Member States to inform the Commission of bilateral agreements (or arrangements) that Member States intend to conclude with the UK. The Commission is then the position to give or withhold authorization for the conclusion of these agreements. Member States cannot take any action that would jeopardize the cooperation under the EU-UK TCA.

As an EU-only agreement, the EU-UK TCA is based on the assumption that all areas of the agreement fall either under exclusive competences or shared competences that are now exercised by the Union. The TCA specifically allows the Member States to enter into bilateral arrangements or agreements with the UK concerning specific matters (air transport, administrative cooperation in the field of customs and VAT and social security). Logically, this seems to imply that bilateral arrangements and agreements are excluded in all other areas, for which such parallel Member State action is not explicitly allowed. 

Were the EU-UK TCA concluded as a mixed agreement the provisions on law enforcement and judicial cooperation in criminal matters would most likely fall largely under national competence. This would offer Member State authorities more space to make the relevant arrangements with UK authorities, since mixed agreements do not, as a matter of principle, stand in way of Member States exercising competences on aspects of shared competence covered by mixed agreement. 

Also, under an EU-only agreement, if problems emerge with the UK authorities or their way of exercising the powers that the TCA creates , individual states cannot react to them because they are not parties in their own right and not represented in the agreement bodies. Instead, they need to alert the Commission to the issue, and rely on the Council to form a position on actions that involve the application of their own national legislation. 

Indirect Effects 

The decision on signing explicitly states that ‘[t]he exercise of Union competence through the [EU-UK TCA] shall be without prejudice to the respective competences of the Union and of the Member States in any ongoing or future negotiations […].’ The CLS stresses that the Member States remain free to exercise the same shared competence in other contexts and vis-à-vis other third countries.  Again, we are not convinced by this argumentation. 

The EU has tried to cement the power division with similar formulations in the past. An example from the international context is the EU accession to the ECHR. While the Treaties impose an obligation of the EU to accede to the ECHR, Protocol 8 on accession stipulates that ‘nothing [in the accession agreement] affects the situation of Member States in relation to the European Convention’. But since the ECJ would interpret and enforce the ECHR and the case law of the ECtHR vis-à-vis the Member States. This would necessarily change their relation to the Convention. 

Legal clauses intended to limit effects on the dynamic competence division between the EU and its Member States struggle to have tangible effects. They simply cannot contain the effects flowing from international agreements, which create rights of third parties and effects under EU law. In the context of Opinion 2/15, Member States referred to Article 207(6) TFEU providing that the exercise of the Union’s CCP competences shall not affect the delimitation of competences between the Union and the Member States, and shall not lead to harmonization of legislative or regulatory provisions of the Member States in so far as the Treaties exclude such harmonization (para 164). Yet, ‘the objective of sustainable development’ came to be seen ‘as an integral part of the [CCP]’ (paras 141 and 147), an exclusive competence of the EU to agree common rules.

CCP, in its broad and objective-oriented interpretation, includes labour rights and environmental aspects, as long as they are (sufficiently) trade-related. EU FTAs may hence regulate a very broad scope of policies in a deep, detailed and invasive fashion. 

The EU-UK TCA has the ambition to continue uniquely comprehensive and deep cooperation. As such, it has a vast range of indirect effects that constrain Member States’ ability to determine policy in the areas covered by it, despite the attempt the opposite. 

To our knowledge, the general question of the possibility and extent of preemption flowing from choice for an EU-only agreement covering not-yet-exercised shared competences is one that the ECJ has not addressed. Therefore, the exercise of shared comptences under the EU-UK TCA may or may not have a preemptive effect for the conclusion of international agreements of the Member States with third countries. 

The argument of legally relevant precedent can be countered by the settled case law of the Court of Justice that the practices of the institutions do not affect what is legal. Yet, this does not make institutional practice irrelevant. Practices create a presumption of legality as long as the Court does not rebut this presumption by finding the institutional practice illegal. 


The EU-UK TCA is much broader in scope than FTAs. It covers for example environmental and climate change issues in a way that does not justify categorizing these issues as trade-related, and includes extensive cooperation on law enforcement and criminal matters. The EU-UK TCA is also explicitly aimed to achieve non-trade objectives, confirmed by its envisaged conclusion as an association agreement. 

When establishing and analysing EU competence, such an agreement should look to existing association agreements for comparison and not to the EU-Singapore FTA. In addition, serious consideration should have been taken of atypical provisions and their competence implications, which have so far not been comprehensively analysed. 

Who should be charged with such comprehensive constitutional analysis prior to the conclusion of a comprehensive and innovative international agreement? In Council discussions, this task is often allocated to the Council Legal Service. 

However, the function of the CLS in the EU structure is to serve decision making in the Council and promote what it identifies as the ‘Council interest’. In this case, the task of the CLS was to save the EU from a non-Deal Brexit and enable the conclusion of the EU-UK TCA without additional procedural hurdles. In particular, the task of the CLS was not to think about Member State sovereignty or even how Member State authorities will deal with the agreement in the future. Its job was to enable ministers to tell their parliaments in capitals that the agreement was ‘business as usual’, that nothing extraordinary was taking place, and that, in any event, the case was entirely exceptional. 

We believe that only the fact that the UK is a former EU Member State does not justify making an exceptional choice as concerns competences of the Union and its Member States for the conclusion of the EU-UK TCA. In fact, we are not convinced that ‘exceptional choices’ as regards competence truly exist. This is why questions of competence and their implications should be brought to public debate. 

The rule of law applies, also under time pressure. This should be a particularly urgent consideration considering that the EU is struggling with an existential rule of law crisis. The exception defines the general.  And, it is sad that the EU goes for excuses instead of upholding its own treaties and democratic principles in times of crises.