Narine Ghazaryan, Assistant Professor in EU and International Law, Radboud University Nijmegen
The judgment in Case C-180/20 (the CEPA case) is the next development in the interinstitutional battle over the boundaries of the Common Foreign and Security Policy (CFSP) with no signs of abating in sight. In this case, the Commission challenged the Council’s departure from the latter’s proposal for a Council Decision taken within the Partnership Council under the EU-Armenian Comprehensive and Enhanced Partnership Agreement (CEPA). Keen to follow the CJEU’s judgment in Commission v Council (Agreement with Kazakhstan) (C‑244/17, EU:C:2018:662) (the Kazakhstan judgment), the Commission removed Art 37 TEU (the legal base for treaties with non-EU countries concerning the CFSP) as a substantive legal basis in its amended proposal for a Council decision – only for the latter to opt for a new approach of adopting two separate decisions, one of which was based on Article 37 TEU, in conjunction with the relevant procedural provisions.
The splitting of a decision authorising a signature of an international agreement has happened in the past. For instance, the Council adopted a separate decision authorising the signing of the CFSP-related provisions of the Association Agreement with Ukraine on a provisional basis. This might have been viewed as necessary for the provisional application of the agreement given the political context surrounding the signature of the Agreement. Unlike the latter case, however, in CEPA it is the decision on the institutional framework of the agreement adopted on behalf of the EU within the CEPA Partnership Council that has been split in an apparent attempt to circumvent the Kazakhstan judgment.
The case solidifies the practice of applying the centre of gravity test in competence delimitation cases involving the CFSP as part of the trend ‘towards a consolidation of EU external action’ (Cremona, p 33). Arguably, it also attempts to clarify the application of the test itself.
Facts and Legal Background
In November 2017, the Council authorised the signing of the CEPA and the provisional application of some parts of the agreement in a decision based on Article 37 TEU and on Article 91 and Article 100(2) as well as Articles 207 and 209 TFEU, in conjunction with Article 218(5) and (7) and the second subparagraph of Article 218(8) TFEU.
Following the CJEU judgment in the Kazakhstan case, in July 2019, the Commission considered it safe to remove Art 37 TEU from the legal basis for the Council Decision on the position to be taken on behalf of the EU within the CEPA Partnership Council as regards the adoption of the Rules of Procedure of the Partnership Council the Partnership Committee, subcommittees and other bodies set up by the Partnership Council, and the establishment of the list of subcommittees. The Decision was to be based on Article 218(9) TFEU, as a procedural legal basis, and on Article 91 and Article 100(2) as well as Articles 207 and 209 TFEU as substantive legal bases.
Subsequently, Coreper decided to split the decision into two with a separate decision based on Art 37 TFEU alongside the procedural provisions in Article 218(8) and Article 218(9) TFEU for CEPA’s Title II on political dialogue and reform, and cooperation in the field of foreign and security policy. The Commission, supported by the Czech Republic, brought an action for annulment under Article 263 TFEU in April 2020. It challenged the exclusion of Title II of the CEPA from the scope of Decision 2020/245 (the main decision), as well as the choice of Art 37 TEU as the substantive legal basis for Decision 2020/246 (Title II decision), objecting in addition to the addition of the second paragraph of Article 218(8) TFEU as a procedural legal basis for Decision 2020/246. By its second plea, the Commission specifically challenged the artificiality of the Council’s approach creating different centres of gravity in order to manipulate the relevant voting rules.
Opinion of Advocate General Pitruzzella
AG Pitruzzella first of all attempted to clarify the centre of gravity test by rejecting the arguments about a hierarchy existing between the elements of ‘purpose’ and ‘content’ of the measure. Instead, the application of the centre of gravity test depends on the circumstances of each case and is ‘highly pragmatic’ (para 28). While addressing the line of case law where the purpose of the measure appeared to impact the outcome of the test at first sight (Portugal v Council, as well as the Philippines judgment), the AG notes the importance of other factors which affected the findings in the relevant cases, including the context of the measure, the variety of purposes or several components (paras 29-33). He then refers to the Kazakhstan judgment as one which brought significant clarifications to the case law through applying the ‘classic’ centre of gravity test to delimiting the CFSP from TFEU competences for the purpose of the application of Art 218 TFEU (para 35-36). In addition to the ‘classic’ test, the AG notes also the centre of gravity test applied by the Court using the ‘quantitative’ and ‘qualitative’ criteria to analyse the content of the relevant international agreement (para 37).
According to the AG, irrespective of whichever test is applied, within the scheme of the CEPA the CFSP-related component is not ‘distinct from the other components’ which would require the reliance on Art 37 TEU (para 55). He then applied the Philippines judgment to the facts of the case. First of all, he considered whether the CEPA provisions related to policies, other than those identified as predominant (transport, trade and development cooperation), may also fall within that policy or whether they go beyond the framework of that policy. Based on a broad conception of development cooperation in line with CJEU case law, in his view, the agreement is first of all aimed at strengthening economic and trade cooperation aimed at promoting sustainable development (paras 57-62). The development cooperation being multidimensional in nature may be linked to Title II of the CEPA removing the need for a separate decision (para 63).
The AG then considers the second part of the two-step test applied in the Philippines judgment to identify whether the CFSP-related provisions of CEPA contain such extensive obligations that they constitute objectives distinct from those of development cooperation. Given the declaratory nature of Title II provisions, he concluded that they do not set obligations requiring a separate legal basis in Art 37 TEU (para 65-68). Based on the approach in the Philippines judgment, he concludes that Art 37 TEU is superfluous as a legal basis.
For the completeness of his analysis, the AG also addresses the possibility of applying the approach taken in the Kazakhstan judgment – the qualitative and quantitative evaluation of the provisions of the agreement – concluding in one paragraph that the answer would be the same as above (para 70). Given the answer to the first plea, the AG considered it unnecessary to answer the second plea.
At the outset, the Court clarifies the issue concerning the procedural legal basis, in particular the inclusion of the second subparagraph of Art 218(8) and the relationship between Art 218 and 219 TFEU. As noted in the Kazakhstan judgment, the voting in the Council in relation to a position to be adopted on behalf of the EU in a body set up an international agreement will depend on whether the situation is covered by the first or the second paragraph of Art 218(8) (para 30). This in turn will require an inquiry into the substantive legal basis (para 31). The Court then turns to the centre of gravity test, confirming the AG’s position that there is no hierarchy between the aim and the content of the measure (32-33). Rather, all objective factors related either to the aim or the content of the measure must be considered in each specific case to determine the field covered in the relevant decision, including the identification of predominant purpose or component if the measure pursues a twofold purpose or has more than one component. The CJEU repeats the established position according to which only if the measure ‘simultaneously pursues a number of objectives or has several components that are inextricably linked, without one being incidental in relation to the other’ can such a measure be based on different legal bases, which is not possible if the relevant procedures are incompatible (para 34). The question is therefore whether the areas covered in the agreement which require qualified majority voting can be considered as predominant in terms of the content or the purpose of the agreement taken as a whole. The three main areas covered by the agreement include transport, common commercial policy and the development cooperation based on Articles 91, 207 and 209 TFEU.
As far as the content of the Agreement is concerned, the Court first of all takes a closer look at the provisions contained in Title II, including on the aims of political dialogue, domestic reform, foreign and security policy. It sides with the AG in confirming that these provisions are few in number compared with 386 articles contained in the agreement which mostly relate to trade and development cooperation. As far as the nature of obligations in Title II is concerned, the Court observes that they are programmatic without setting specific programme of actions for the parties (para 46).
As for the aims of the Agreement, Article 1 on the objectives read alongside with the preamble and the majority of the provisions demonstrate that the CEPA is a framework agreement established to advance bilateral cooperation in the fields of transport, trade and development cooperation (para 47-48). The Court adopts a broad reading of development cooperation in line with its pre and post-Lisbon case law (the ECOWAS and the Philippines judgment) which in the case of CEPA is not negated by the CFSP-related specific objectives in Article 1 on the aims of the agreement. Crucial for the Court is the fact that the enumeration of the CFSP-specific objectives in Article 1 is not ‘accompanied by any programme of action or concrete terms governing cooperation’ in the field of CFSP (para 52).
The CJEU also dismissed France’s argument that the presence of the Nagorno-Karabakh conflict should necessarily be seen as part of the context of the agreement placing its security component at the forefront. According to the Court, the CEPA includes no provisions with specific obligations concerning the Nagorno-Karabakh agreement, and the contested decisions concerning the functioning of the CEPA institutional framework do not entail any concrete measures between the parties taken based on the CEPA that might be relevant for the conflict (para 54-55).
Ultimately, the Court finds that the CFSP-related components of the CEPA are not such as to constitute an ‘autonomous’ component of the agreement, and orders the annulment of both decisions which also results in the dismissal of the Commission’s second plea (para 56-59). In line with the established position, the CJEU maintained the effect of the relevant decisions for the sake of legal certainty until the Council rectifies the position by adopting a new decision in compliance with the judgment.
The main question to ask is whether the judgment merely trails the previous jurisprudence or adds any clarifications to the case law.
Council’s attempt to prevent the ‘absorption’ of the CFSP?
Despite suggestions that splitting the legal basis ‘will ensure the non-affectation between CFSP and non-CFSP procedures and preserve the special status of the CFSP in the EU legal order’ (MIadzvetskaya 2020), with the CEPA judgment the Court has rejected such an approach. Similar to the Kazakhstan judgment, there is no recourse to Article 40 TEU which further indicates that the CFSP does not have an inherently distinct character requiring strict guarding in each and every case where the division of competences alongside TEU and TFEU legal bases is contested. Rather, the CEPA judgment confirms once again that the centre of gravity test applies similar to all other cases of competence delimitation.
The insertion of Article 37 TEU by the Council in this respect comes across as an attempt to bypass the Kazakhstan judgment by artificially splitting the relevant decision. This led to the odd outcome of a joint substantive legal basis for the entire agreement and a split legal basis for a decision within the Partnership Council on the institutional framework of CEPA meant to apply to the entire agreement. The Council’s argument that the lack of objection by the Commission toward the inclusion of Art 37 TEU as a substantive legal basis when authorising the signing of the CEPA justifies its approach was rightly rejected by the AG: on its own, this fact does not justify such an outcome.
The same argument can be said to give away the underlying concern of the Council after the Kazakhstan judgment: if the decision on the adoption of the CEPA institutional framework can exclude Art 37 TFEU as its substantive legal basis alongside the relevant procedural provision requiring unanimity, can that in principle entail that Art 37 TEU can also be omitted as far as the decision on approving the agreement itself is concerned? This concern is not ungrounded since the CJEU’s proclamation in the Kazakhstan judgment that the links between EU-Kazakhstan Partnership Agreement and the CFSP ‘are not sufficient for it to be held that the legal basis of the decision on the signing of that agreement, on behalf of the EU, and its provisional application had to include Article 37 TEU’ (emphasis added, para 43).
As a result, Art 37 TEU as such is unnecessary as a legal basis for decisions authorising the signature of framework agreements focused predominantly on trade and development cooperation. Indeed, there are already such examples, including the Framework Agreement on Comprehensive Partnership and Cooperation with Vietnam. The CEPA’s CFSP provisions could have similarly have been viewed as insufficient to necessitate a separate legal basis in Art 37 TEU for the decision approving the agreement. One can even claim that the Kazakhstan judgment casts a retrospective shadow on the inclusion of Article 37 TEU as a substantive legal basis for decisions signing trade and cooperation agreements. It is not a surprise therefore that in justifying its position in CEPA the Council raised the fact that the Commission did not challenge the inclusion of Art 37 TEU among the legal basis for the decision authorising the relevant agreement. One can therefore argue that the splitting of the decisions on the part of the Council was a reaction to the Kazakhstan judgment: the Council is keen on preventing the ‘absorption’ of the CFSP. It perhaps anticipates a further challenge to the very practice of including Art 37 TEU as a substantive legal basis for framework agreements focusing predominantly on trade and development cooperation.
What next for the ‘centre of gravity’ test?
The CEPA judgment is worthy of attention in terms of the role of the quantitative and qualitative criteria as far as the centre of gravity test is concerned.
The Court’s approach in this regard can be contrasted with that of the AG Pitruzzella. The latter considered the use of the ‘quantitative’ and ‘qualitative’ tests as second to the ‘classic’ centre of gravity test. He clearly has a preference for the two-step test used in the Philippines judgment, dedicating most of his attention to its application. Only after reaching his conclusion based on the application of the Philippines judgment, does he note that the same conclusion would have been reached if the quantitative/qualitative criterion is applied as seen in the Kazakhstan judgment.
The CJEU, on the other hand, does not make a similar distinction between the ‘classic test’ and the quantitative and qualitative criteria used in Kazakhstan. Unlike in the latter case, however, the CJEU bases its findings on a much more substantive and systematic analysis of the aim and the content of the agreement. The CEPA judgment makes an effort to interlace the quantitative and qualitative criteria with the approach used in the Philippines judgment (para 45). For instance, with reference to the content of the agreement, the Court carries out a quantitative and qualitative analysis of the relevant provisions concluding that they should be viewed with regards to the essential object of the agreement rather than in terms of individual clauses ‘provided that those clauses do not impose such extensive obligations … that those obligations in fact constate objectives distinct from those of development cooperation’(para 45).
Similarly, in relation to the aim of the agreement, the quantitative and qualitative overview of the provisions is linked to the tried and tested approach of Portugal v Council and Philippines judgments (paras 47-53). In looking at the objectives, the Court does not attempt to identify traditional CFSP legal objectives (as was suggested by Advocate General Bot in the Mauritius judgment), and merely refers to objectives which ‘may be linked to CFSP’ (para 52). It finally reaches a conclusion that the CFSP-linked elements ‘cannot be regarded as constituting a distinct component of that agreement but it is, on the contrary, incidental to the principal components’ concerning the development cooperation. As mentioned above, the Court takes a broad view of development cooperation in line with its previous case law, including ECOWAS. In fact, the broad interpretation of the development cooperation should be seen as the ‘lasting effect’ ECOWAS had made on EU external relations law (Hillion and Wessel, 582-583).
Furthermore, in addition to the content and the aim of the agreement, its context also featured in the judgment. While the CJEU’s approach has been criticized in the past for giving significant weight to the context of the agreement (Sánchez-Tabernero, p 910), CEPA demonstrates that the context does remain one of the objective factors reviewed by the court, particularly if relevant arguments have been raised by the parties. At the same time, the CJEU is wary of artificial context constructs, such as the attempt by France to use the Nagorno-Karabakh conflict to create a security-related context for the agreement, which was rightly rejected by the Court.
The CEPA judgment should be seen as another stop in the long saga of the inter-institutional battle for competence delimitation in EU external relations. The Council, ever sensitive to guarding the CFSP, attempted to bypass the outcome of the Kazakhstan judgment only for the Court to cast more shadow on the practice of including Article 37 TEU as a legal basis for decision on signing of trade and cooperation agreements. The CEPA judgment also offers a more substantive and thoughtful application of the centre of gravity test in comparison with the Kazakhstan judgment. Judging by the past experience, CEPA will not be the last call on this matter.
M Cremona, ‘The Principle of Conferral and Express and Implied External Competences’ in E Neframi and M Gatti (eds), Constitutional Issues of EU External Relations Law (Nomos 2018)
C Hillion and RA Wessel, ‘Competence Distribution in EU External Relations after ECOWAS: Clarification or Continued Fuzziness?’ (2009) 46 CMLR 551
SR Sánchez-Tabernero, ‘The Choice of Legal Basis and the principle of consistency in the procedure for conclusion of international agreements in CFSP contexts: Parliament v. Council (Pirate-Transfer Agreement with Tanzania)’ (2017) 54 CMLRev 899