Some comments on the CJEU’s judgment in Republic of Moldova v Komstroy LLC
On 2 September 2021, the CJEU rendered its latest – but probably not its last – decision on the salient issue of the incompatibility of intra-EU investment arbitration, this time under the 1994 Energy Charter Treaty (ECT), with EU law. Without a doubt, the Court’s judgment brings some certainty – but no clarity – on this question in respect of EU law and the EU legal order. Nevertheless, the Court carefully avoided to address the consequences and the unveiling uncertainties that its decision is likely to create for investors, investment arbitration tribunals, EU Member States, the EU and its partners.
No surprise …
It comes without surprise that the Grand Chamber of the Luxembourg court ruled that ‘Article 26(2)(c) ECT must be interpreted as not being applicable to disputes between a Member State and an investor of another Member State concerning an investment made by the latter in the first Member State.’ (§ 66) Of course, the 2018 Achmea judgment did not concern a multilateral investment instrument comparable to the ECT, but related only to the issue of intra-EU arbitration under a bilateral investment treaty, in that case the Netherlands-Slovakia BIT and its Article 8. Of course, that judgment used language that could be and has been interpreted as distinguishing between such BITs from other investment instruments, in particular those that are not concluded only between Member States, but also by the EU (§ 58); the ECT is such an agreement concluded by the EU, its Member States and a number of third States. Yet, despite some doubts – and, probably, some hope – concerning the scope of the Achmea decision, the position of the Court and the European Commission had become rather obvious over the last couple of years. Therefore, the Court’s judgment was certainly expected by arbitral tribunals among others (see, e.g., Hydro Energy 1 and Hydroxana v Spain, Decision of 9 March 2020, § 493).
One might not agree with the European Commission’s hostile position against the existing system of investment arbitration and its proposals to replace this system by a Multilateral Investment Court. But there is some merit in the Court’s position against arbitration between an EU Member State and an investor of another EU Member State. In a community of law, there is little reason not to submit such disputes before regular courts and tribunals without any fear of partiality or unprofessionalism. It is not, and it should not be shocking to litigate against State authorities in front of State courts within the EU. The decision of the German Constitutional Court in the Vattenfall case confirms this point quite well.
As a matter of fact, the very existence of intra-EU investment arbitration is not the result of a conscious choice, but of mere oversight. This was certainly the case of intra-EU BITs and their arbitration provisions. No BIT was ever concluded between two EU Member States in the first place. They all ‘became’ intra-EU treaties in the realm of EU enlargements and accessions. As surprising as this might sound, they were simply forgotten. As of 15 September 2021, 67 intra-EU BITs have been terminated – slowly but surely – under the 2020 Termination Agreement. Nevertheless, 96 intra-EU BITs are still in force.
This is probably also the case of the ECT. As explained in the LBBW v Spain case, a disconnection clause that could have prevented intra-EU arbitrations was tabled during the negotiations, but did not make it to the final text. The reasons why remain now pure speculation. In any event, back in 1997 when the EU acceded to the ECT, there was not the slightest suspicion that some of its provisions might be incompatible (Opinion of Advocate-General Wathelet, Achmea, 19 September 2017, para 43). The Court has now ruled that everybody was wrong.
… yet surprising …
The Court’s judgment is nevertheless surprising as to its form and as to its reasons.
The judgment is surprising as to its form, because the underlying dispute was not an intra-EU dispute. The annulment proceedings before the Paris Court of Appeal concern arbitral proceedings brought under Article 26 ECT by a Ukrainian investor, Energoalliance Ltd., against Moldova in which an award was rendered in 2013. Indeed, the Paris Court of Appeal never asked the CJEU about a potential incompatibility between the ECT arbitration clause and EU law. In accordance with the Court’s prevailing case law, such a preliminary ruling request would certainly have been dismissed, because ‘a reference for a preliminary ruling made by a national court is to be rejected where it is quite obvious that the interpretation of EU law that is sought is unrelated to the actual facts of the main action or its object, [or] where the problem is hypothetical …’ (see, e.g., Eli Lilly and Company, C-239/19, ECLI:EU:C:2019:687, para 15).
The Paris Court of Appeal only sought guidance from the Court on the interpretation of Article 1(6) ECT and Article 26(1) ECT and the definition of ‘investment’. The operative paragraph of the judgment contains the response to this question of interpretation only. Yet the Grand Chamber’s decision will not be remembered for this, as controversial the interpretation of these provisions might be.
By a rather ‘creative’ reasoning, the Court nevertheless explained that it was indeed ‘necessary’ to address the question as to ‘which disputes … may be brought before an arbitral tribunal pursuant to Article 26 ECT’. This allowed the Court to proceed and to provide a purely abstract interpretation – a legal opinion – on the question whether intra-EU disputes can be submitted to arbitration under the ECT. The fact remains that an interpretation concerning this particular question is not and cannot be necessary for the Paris Court of Appeal to give judgment in the case upon which it is called to adjudicate. The Court’s finding on incompatibility between the ECT arbitration provisions and EU law remains – for the time being – pure obiter dictum. But the preliminary ruling request from Paris was a welcome opportunity for Luxembourg to set out its opinion on an issue that has been haunting Member States for several years, and that tribunals within the EU, e.g. the Svea Court of Appeal, were reluctant to submit to the scrutiny of the CJEU so far. Now, the die is cast.
The judgment is also surprising as to its reasoning and approach. To recall, in Achmea, the Grand Chamber did not purport to interpret Article 8 of the Netherlands-Slovakia BIT, but the relevant provisions and principles of EU law. It concluded that ‘Articles 267 and 344 TFEU must be interpreted as precluding’ an arbitration provision such as Article 8 of that BIT. In the recent Komstroy Judgment, however, the Court ‘interpreted’ Article 26(2)(c) ECT as such.
Of course, as the Court rightly explained in paragraph 23, being a mixed agreement which is binding on the EU and its Member States, the ECT forms an integral part of the EU legal order and EU law. As such, the Court is entitled, under Article 19 TEU, to provide an interpretation of the provisions of the ECT. Yet, the Court cannot simply ignore the nature of the ECT. Being an international agreement, its interpretation is subject to international law, and in particular the rules and principles of treaty interpretation as codified in Articles 31 and 32 of the 1969 Vienna Convention (see, e.g., Brita, C-386/08, EU:C:2010:91, paras 39-43). Despite the well-established case law on this issue, the Court neither mentioned the Vienna Convention rules, nor applied them. Nothing in the judgment suggests that the CJEU even tried to interpret Article 26 ECT ‘in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose’. The Court simply imposed its vision of what Article 26 ECT should have said or should not have said. What must not be, cannot be! This essential flaw in the Court’s reasoning will certainly constitute one of the most precious arguments for investors before international arbitral tribunals.
In addition, the way the CJEU transposed the findings of its Achmea judgment to the ECT are not beyond criticism, in particular in light of the Court’s 2019 CETA opinion. The essential basis of the incompatibility analysis is the following: Arbitral tribunals cannot participate in the preliminary ruling procedure established under Article 267 TFEU. Therefore, if they are called to apply and interpret EU law, their decisions could jeopardize the uniformity and autonomy of EU law. To put it simply, arbitrators could get it wrong and there may be no remedy to get it right. Yet, in respect of the CETA, the Court, rightly, found that the CETA tribunals’ and appellate body’s powers are confined to the provisions of the CETA and not EU law (Opinion 1/17, ECLI:EU:C:2019:341, para 122). There is no reason to distinguish between the CETA and the ECT in this respect. Article 26(1) ECT equally limits the jurisdiction under this provision to disputes concerning ‘an alleged breach of an obligation of [a Contracting Party] under Part III [of the ECT]’. Indeed, the Court itself recognized that an ECT arbitral tribunal has to rule ‘on the issues in dispute in accordance with the ECT and with the applicable rules and principles of international law’ (§ 48). Nevertheless, the Court chose to balk this obstacle declaring that the ‘ECT itself is an act of EU law’ (§ 49). The Court went even so far as to set out the relevant provisions of the ECT as part of EU law in the introduction to its decision (§§ 3-6).
This is bold legal imperialism. The CJEU could have shown some more respect for the other international legal order. In this respect, the rather unnuanced Komstroy decision contrasts with the very fine reasoning in the CETA Opinion. In the latter, the Court provided some useful elements to distinguish intra-EU disputes. The magical concept is, and would have been, ‘mutual trust’ (Opinion 1/17, ECLI:EU:C:2019:341, paras 127-9). Surprisingly, the CJEU chose not to mention either mutual trust or its 2019 Opinion in this respect.
… yet without any (surprising) consequences.
The Komstroy judgment and the rather categorical position of the CJEU will have to be tested by ECT arbitral tribunals in the weeks and months to come. EU Member States will rush to include the judgment in the records of pending intra-EU ECT arbitrations, or try to reopen the discussion of jurisdictional matters in light of this new decision.
Nevertheless, arbitral tribunals remained quite unimpressed by the Achmea judgment and have constantly rejected intra-EU objections in the past, including in arbitrations conducted under the ECT. There is no reason to believe that they will act any differently in the future. Hopefully, they will not feel bound by the interpretation provided by the CJEU, but (continue to) abide by well-established rules and principles of treaty interpretation, taking into account the ordinary meaning of the terms of Article 26 ECT. This is also in line with the position of the CJEU itself: ‘Indeed, with respect to international agreements entered into by the Union, the jurisdiction of the courts and tribunals specified in Article 19 TEU to interpret and apply those agreements does not take precedence over either the jurisdiction of the courts and tribunals of the non-Member States with which those agreements were concluded or that of the international courts or tribunals that are established by such agreements.’ (Opinion 1/17, ECLI:EU:C:2019:341, para 116).
The problem is more complex in respect of annulment and set aside actions, in particular those before tribunals within the EU. As a matter of EU law, these courts and tribunals have no choice other than to respect the CJEU decisions and to implement the primacy of EU law, just like the Bundesgerichtshof did in the Achmea saga. This means that, in principle, they will have to annul or set aside intra-EU awards rendered under the ECT. Therefore, intra-EU arbitration proceedings seated in the EU and enforcement actions in the EU are now extremely risky.
To what extent other jurisdictions, and in particular the United Kingdom, will adopt a similar position is questionable.
In any event, the CJEU’s judgment does not resolve the incompatibility issue. It is not enough to ‘interpret’ incompatible provisions in international treaties away, as long as not all parties to the treaty agree to such an ‘interpretation’ (see Vienna Convention on the Law of Treaties, Article 31(2)(a)). The EU and EU Member States can of course modify Article 26 ECT or any other provision of this instrument in their mutual relations, as long as they respect the formal requirements imposed for such inter se amendments (see Vienna Convention on the Law of Treaties, Article 41(2)). Surprisingly, they have not taken any steps in that direction so far, with the notable exception of Italy’s withdrawal from the ECT in 2015. In the long run, the EU and its Member States will have to take some concrete actions. The ECT will have to be modified to give full effect to the Court’s decision, maybe as part of the wider process of its modernisation. Hiding behind questionable court decisions, even decisions from the EU’s top court, is not and cannot bring a suitable solution.
Daniel has been working in international dispute settlement for twenty years, and has represented States as counsel and advocate on numerous occasions before international courts and tribunals, including the International Court of Justice, the International Tribunal for the Law of the Sea, the European Court of Human Rights, the Court of Justice of the European Union, as well as before arbitral tribunals constituted under Chapter VII of the United Nations Convention on the Law of the Sea and ad hoc arbitral tribunals. He has also represented investors in investor-State disputes before ICSID tribunals and before tribunals operating under the UNCITRAL Rules.