On 6 October 2021, Advocate General (AG) Michal Bobek delivered his Opinion in Joined cases C-59/18 and C-182/18 Italy v Council (Siège de l’Agence européenne des médicaments) and Case C‑743/19 Parliament v Council, two challenges of the decisions on the seat of the newly established European Labour Authority (ELA) and the new seat of the European Medicines Agency (EMA). The main conclusion of the AG in his Opinion is that the Court of Justice lacks jurisdiction over the decisions of Member States on the location of the EMA and the ELA. In his view, “as a matter of principle and certainly in view of how the Treaties currently stand, the Court has no jurisdiction under Article 263 TFEU over decisions taken by the representatives of the Member States” (para 82). The AG’s Opinion, however, also reveals an incorrect interpretation of the scope of Article 341 TFEU.
Article 341 TFEU provides that: “The seat of the institutions of the Union shall be determined by common accord of the governments of the Member States.” It is understood that due to their political sensitivity, decisions on the location of EU institutions should be taken unanimously by the Member States. Decisions on the seat of EU agencies can be just as politically sensitive, but are such decisions included in the scope of Article 341?
In his Opinion, the AG suggests that Article 341 TFEU is not appropriate to serve as the basis for decisions on the seat of EU agencies (para 143). In this respect, he proposes for the Court to rule that decisions made by the Representatives of the Member States do not produce any legal effects under EU law. Whilst fully in agreement with the AG’s conclusion on inadmissibility of the action, this post will present a critical view on the suggested inapplicability of Article 341 TFEU for decisions on the seat of EU agencies.
If the CJEU were to follow the AG’s Opinion, decisions on the seat of an EU agency, instead of being made by common agreement of the Member States under Article 341 TFEU, would have to be adopted by the corresponding EU legislative procedure (see para 176). To a large extent, the Opinion therefore meets the demands of the European Parliament (para 84), which has long been interested in co-deciding the seats of EU agencies. For the Member States, this conclusion, if accepted by the Court, results in the loss of Article 341 TFEU as a legal basis for the selection of EU agencies.
The AG’s conclusions on the inapplicability of Article 341 TFEU for the selection of seats of EU agencies are, as I will show below, not substantiated.
It should be noted that the AG’s Opinion is rather brief on the historical interpretation of Article 341 TFEU (para 110). Whilst Article 341 TFEU and its predecessors in previous Treaties have been interpreted by the Court on several occasions (see e.g. Case 230/81 Luxembourg v Parliament), the question of its scope has so far never been addressed. When the scope of a particular Treaty provision is at stake, the Court regularly also examines the historical context of that provision to ascertain its purpose and scope. For example, in Case C-370/12 Pringle (para 135) and Case C-62/14 Gauweiler (para 100), the Court inferred the purpose and scope of Articles 125 and 123 TFEU, respectively, from their historical context. However, in the present Opinion, the AG laconically explains that the substance of Article 341 TFEU and its predecessors has remained the same since 1951 (when a similar provision was included in Article 77 of the Treaty of the European Coal and Steel Community (ECSC)), without trying to find out its purpose (see para 110).
Undoubtedly, finding the documents related to the preparatory work for the Paris and Rome conferences in the 1950s may be quite challenging. However, looking into the historical context may reveal interesting elements of the drafters’ intentions. Whilst at the time of establishment of the European Economic Community (EEC) no agency was created, Article 216 of the EEC Treaty (today Article 341 TFEU) was drafted with the need in mind to establish not only the seat of the Community institutions but also, in the future, that of other bodies. Indeed, the original EEC Treaty already envisaged, aside from the Community institutions, also the European Investment Bank, the Economic and Social Committee, and the Monetary Committee (Articles 129, 193 and 105). It is therefore obvious, that the drafters of the original EEC Treaty envisioned already in 1957 a wider scope for the Article 216 (now Article 341 TFEU) than only to serve as a basis for deciding the seats of Community institutions. The historically wide scope of this provision was most likely the reason why the Lisbon Treaty did not amend it to mention explicitly also bodies, agencies and offices.
The two Edinburgh Decisions
Another problematic part of the Opinion is the qualification of the two 1992 Edinburgh Decisions. The first Edinburgh Decision establishes the seats of the EU institutions, bodies and departments, the second one deals with the Danish arrangements for the Maastricht Treaty. The AG suggests in paragraph 116 of the Opinion that the first Edinburgh Decision was not adopted pursuant to the provisions of EU law and only has legal value under EU law to the extent that its content has been taken over by EU law. This conclusion seems to be in contradiction with other parts of the Opinion. On one hand, in paragraph 116, the AG limits the legal effects of the first Edinburgh Decision under EU law to the extent is has been taken over by EU law, on the other hand, in paragraph 166, he admits that the decisions of the Member States adopted pursuant to the Treaties produce legally binding effects under EU law. In this respect, it is apparent from the wording of the first Edinburgh Decision that it was adopted pursuant to the Treaties. Indeed, in its preamble, the Decision refers to the Article 216 EEC (today Article 341 TFEU). In Article 1, it sets the seats of the EU institutions and some other bodies (thus using a wider scope than institutions only) as envisaged in Article 216 and, in Article 2, it envisages the decision on the location of other bodies and departments at the next European Council. In Case C-345/95 France v Parliament, the Court in fact confirmed that the first Edinburgh Decision represents the decision based on Article 216 EEC (now Article 341 TFEU) (see para 23).
However, that same first Edinburgh Decision is qualified by the AG only as an international agreement without taking into account its basis in the EEC Treaty. Furthermore, the AG also seems to conflate the two Edinburgh Decisions, which were admittedly adopted at the same time but substantially are completely disconnected. The first Edinburgh Decision, which relates to the seats of institutions and certain bodies, was adopted pursuant to Article 216 EEC (Article 341 TFEU), and hence its qualification as a purely international agreement without any binding effects under EU law is, in my view, not appropriate. The second Edinburgh Decision regarding certain Danish arrangements for the Maastricht Treaty is a completely separate one, dealing with issues other than that of the seats of institutions, bodies, offices and agencies. It has no basis in the EU Treaties and thus can be correctly qualified as an international agreement. In analysing the first Edinburgh Decision as an international agreement, the AG refers in paragraph 118 to Case C-135/08 Rottmann, which concerns the second Edinburgh Decision only.
The Opinion contains another contradiction, namely between paragraph 117 and footnote 68. On the one hand, paragraph 117 claims that only Article 1 of the first Edinburgh Decision was incorporated into EU law through the later adopted Protocol No. 6, thus suggesting that it was not part of EU law before 1999 (the entry into force of the Amsterdam Treaty to which Protocol No. 6 was annexed). On the other hand, in footnote 68, the AG confirms that when adopted on the basis of EU law provision, an act like the first Edinburgh Decision becomes part of EU law and refers to Case C-345/95 France v Parliament, in which the Court confirmed that the first Edinburgh Decision was adopted on the basis of what is now Article 341 TFEU and thus was part of EU law.
Depriving Article 341 TFEU of its effectiveness
In the present challenges of the seats of the EMA and the ELA, the Council argued that the exclusion of agencies from the scope of Article 341 TFEU would deprive this provision of any effectiveness, since the seats of the EU institutions are already set out in Protocol No. 6, and any change to those seats would necessitate the use of the Treaty revision procedure enshrined in Article 48 TEU. Whilst the AG admits that Article 341 TFEU has to a large extent exhausted its potential, he does not find the Council’s argument convincing enough to conclude that the scope of Article 341 TFEU also covers agencies (para 138). He insists that Article 341 TFEU can still be used for the seat of a potential new institution or for the change of an existing one (para 138). However, he fails to address the Council’s argument that the seat of a new institution or the change of an existing one would have to follow the revision procedure enshrined in Article 48 TEU, thereby rendering Article 341 TFEU futile. Indeed, in both cases, Protocol No. 6 would have to be amended by using Article 48 TEU, and Article 341 TFEU would not be applicable.
Subsequently, the AG’s Opinion rejects the unanimity voting in the Council due to the political sensitivity of decisions on the location of EU agencies. In this respect, the Opinion argues that, in voting for the seat of EU agencies, the Council has departed from unanimity and used simple majority voting instead (para 142). In this respect, it is true, that simple majority voting is used to narrow down the choice of city candidates for the seat. However, the AG omits the fact that the simple majority voting is regarded only as an indicative expression of preferences and the city taking the most votes subsequently has to be unanimously agreed upon by the Representatives of the Member States. For example, as the document on the ELA selection stipulates, “the outcome of the voting process will be confirmed by common agreement of the Member States’ representatives at the same meeting”.
The decisions of the Representatives of the Member States and their connection to EU law
Finally, the AG suggests that decisions of the Representatives of the Member States, adopted outside the Treaties, do not produce any binding legal effects in EU law (para 167). This absolute statement is in need of further qualification. In principle, I concur with this conclusion to the extent that these decisions themselves do not produce legally binding effects directly under EU law. However, whilst I still believe that, in taking the EMA and ELA seat decisions, the Representatives of the Governments of the Member States acted under Article 341 TFEU, even if they had acted outside the framework of the Treaties (quod non), there exists a certain connection between EU law and the decisions of the Representatives of the Member States adopted outside the Treaties, if such decisions are related to the functioning of the EU.
It is true that the EU Treaties whilst stipulating the legal effects of the acts of EU institutions in Article 288 TFEU, are silent on the legal effect of decisions made by the Representatives of the Member States. However, the accession treaties bind new Member States to these decisions, without specifying whether their binding nature under EU law is contingent on them having been adopted on the basis of the Treaties or outside the Treaty framework. In precise terms and by way of example, according to Article 5 of the 2003 Act of Accession:
The new Member States accede by this Act to the decisions and agreements adopted by the Representatives of the Governments of the Member States meeting within the Council. They undertake to accede from the date of accession to all other agreements concluded by the present Member States relating to the functioning of the Union or connected with the activities thereof.
Following the logic that the main purpose of an accession treaty is to ensure that the EU law and the acquis communitaire as a whole become binding upon the new Member States, it is difficult to see how decisions of the Representatives of the Member States meeting within the Council lack any legal connection with EU law and the acquis. One could argue that the above-mentioned Article 5 covers only the decisions adopted on the basis of the Treaties. However, nothing in Article 5 indicates such a limitation. There is also no need for “all other agreements”, as specified in the last sentence of this provision, to be based on the Treaties.
In my view, Article 5 of the 2003 Act of Accession means, in practice, that the ten new Member States became bound, for example, by both Edinburgh Decisions at the date of their accession in 2004 through the accession treaty, which represents an instrument of both EU and international law. Indeed, stemming from this logic, the decisions made by the Representatives of the Member States, as well as any other agreements related to functioning of the EU or connected with the activities thereof, whilst not perceived as part of EU law, should be perceived at least as international obligations of the Member States binding upon them either directly through international law or indirectly through EU law, i.e. the accession treaty. Indeed, as the Court of Justice confirmed in Case 44/84 Hurd, the purpose of the type of provision, like Article 5 of the 2003 Act of Accession, in the accession treaty is to extend the acquis communitaire, which the Member States are bound to accept, to all other acts adopted by the common agreement of the Member States (para 29). The view of the Court, together with the wording of accession treaty, mean that decisions of the Representatives of the Member States related to the functioning of the EU may, depending on their nature, create legal effects for the Member States.
In addition, as will be demonstrated below, decisions of the Representatives of the Member States related to the functioning of the EU create a special connection with EU law. This connection means that, these decisions facilitate the proper implementation of EU law, and consequently without implementing the decisions, the proper application of EU law might be hindered.
This was the case for the Decision of the Representatives of the Governments of the Member States, meeting within the Council, of 27 February 2002 on the financial consequences of the expiry of the ECSC Treaty and on the research fund for coal and steel (2002/234/ECSC). By this decision, the Member States entrusted the European Commission with the management of the ECSC assets and liabilities after expiry of the ECSC Treaty, an act certainly creating legal effects under international law. However, without the application of this decision ensuring the transition of assets from the ECSC to the Community, the effective management of former ECSC assets by the Community would be precluded. The same is true for another decision, namely the decision of the representatives of the Member States of 28 April 2004 concerning privileges and immunities granted to Athena (2004/582/EC). Without the recognition of privileges and immunities for Athena by the decision of the representatives of the Member States, the mechanism established by Council Decision 2004/197/CFSP would be prevented from efficient application.
The above conclusions can also be applied to the decisions establishing the seats of EU agencies. Without these decisions on their seats, the agencies would only function on a provisional basis and not become fully operational, thereby hindering the efficient application of the EU acts governing the establishment and functioning of such agencies. Thus, whilst I maintain that decisions on the seat of EU agencies are correctly based on Article 341 TFEU, if the Court were to rule the contrary, the AG’s conclusion that the ELA’s seat in Bratislava is only matter of fact, and does not produce any legal effects (see para 172) is, in my view, incorrect.
The AG could have easily suggested to the Court, after examining the relevant case law on the judicial review of the decisions of the Representatives of the Member States (Case C‑181/91 and C‑248/91 Parliament v Council and Commission, C-684/20 P Sharpston v Council and Conference of the Representatives of Governments of the Member States, etc.) to declare the actions inadmissible, without elaborating on substantive issues, including certain legal aspects of the seats of EU agencies. However, he decided to dwell on the scope of Article 341 TFEU and the legal nature of decisions of the Representatives of the Member States. Whilst this is certainly his right, as outlined above, his arguments are open to serious doubts, which should be resolved by the Court. In any case, as Shakespeare said, sometimes, less is more.