Abstract: In case C-741/19 Republic of Moldova v Komstroy LLC ECLI:EU:C:2021:655, the Grand Chamber of the Court of Justice of the European Union found that the acquisition of a claim arising from a contract for the supply of electricity does not constitute an “investment” within the meaning of the Energy Charter Treaty (ECT). Yet the impact of the case goes far beyond this finding. In coming to this conclusion, the Court found that i) it has jurisdiction to give a preliminary ruling in a dispute that has little or no connection to the EU legal order and ii) the intra-EU application of the ECT’s investor-state dispute settlement mechanisms is incompatible with EU law. The Court thus answered the question, debated in academia and before arbitral tribunals, whether the reasoning in its 2018 Achmea judgment applied in relation to the ECT’s dispute settlement provisions. Whereas arbitral tribunals have approached the issue through the lens of public international law, in particular the law of treaties, the EU Court approaches the question as one about the autonomy of the EU legal order. Like Achmea, the effects of Komstroy outside the EU legal order are likely to be limited.
Keywords: art. 267 TFEU – international agreements – Achmea – jurisdiction – Energy Charter Treaty – autonomy.
Art. 26(6) of the Energy Charter Treaty (ECT) sets out that a tribunal established under the ECT shall decide disputes “in accordance with this treaty and with applicable rules and principles of international law”. The Tribunal in Electrabel SA v Hungary found that EU law forms part of the relevant rules of international law that can be taken into account when interpreting a treaty pursuant to art. 31(3)(c) of the 1969 Vienna Convention on the Law of Treaties (VCLT). It reasoned “EU law is international law because it is rooted in international treaties; and both Parties accepted, of course, that the EU Treaties are legal instruments under public international law… Moreover, the Tribunal considers that EU law as a whole is part of the international legal order…”. Is EU law international? This debate is arguably as old as the European Community. The question again became legally relevant in the case of Republic of Moldova v Komstroy LLC (Komstroy), in which the Grand Chamber of the CJEU held that the “ECT itself is an act of EU law”. Therefore, the Court reasoned, an arbitral tribunal “such as” that referred to in art. 26(6) ECT “is required to interpret, and even apply, EU law”. The consequence of this finding is that the dispute settlement provisions in the 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”. In other words, the CJEU found that the intra-EU application of the ECT’s investor-state dispute settlement mechanisms is incompatible with EU law.
In 2018, the CJEU found in Slovak Republic v Achmea BV (Achmea) that the arbitration clauses in intra-EU bilateral investment treaties (BITs) “such as” the Agreement on Encouragement and Reciprocal Protection of Investments between the Kingdom of the Netherlands and the Czech and Slovak Federal Republic (1991) (Dutch-Slovak BIT) are incompatible with EU law. Since then, there has been debate, both in academia and before arbitral tribunals, about the effect of the Achmea decision. In particular, there was discussion about whether the CJEU’s reasoning in Achmea would apply also with respect to the intra-EU application of the ECT’s dispute settlement provisions. One of the reasons put forward for limiting the effect of Achmea in the ECT context was that, unlike the Dutch-Slovak BIT, which would allow a tribunal to interpret EU law, and thus violate the autonomy of the EU legal order, an ECT Tribunal would only apply the ECT and “applicable rules and principles of international law”.
Komstroy originated as a request for a preliminary ruling involving a dispute between the Republic of Moldova and Komstroy LLC, a Ukrainian investor. The case dealt with the question of whether a claim arising from a contract for the supply of electricity constitutes an “investment” within the meaning of the ECT. Yet the importance of the case goes much further than this technical question. First, the Court found that it had jurisdiction to issue a preliminary ruling interpreting the ECT in a dispute that had little or no connection to the EU legal order. Second, the Court used the case as an opportunity to state – albeit in obiter dicta – that the ECT’s investor-state dispute settlement mechanisms are incompatible with EU law.
This note addresses these two main issues. First, it discusses why the Court found that it had jurisdiction to provide a preliminary reference. It then addresses questions related to the intra-EU application of the ECT. While Komstroy builds upon its previous case-law, including Achmea and Opinion 2/13, on the application of the principle of “autonomy” in EU law, it fails to shed much light on the meaning of this principle. Whereas the Court has in the past interpreted EU agreements using the principles enshrined in the VCLT, Komstroy shows how the Court can approach these issues entirely through the lens of EU law.
Komstroy originated as a request for a preliminary ruling from the Paris Court of Appeal concerning the interpretation of art. 1(6) (on the definition of “investment”) and art. 26(1) (on dispute settlement) of the Energy Charter Treaty. The ECT is a multilateral treaty designed to establish a framework for cooperation in the field of energy, to promote energy security and to “catalyse economic growth by means of measures to liberalise investment and trade in energy”. One of the goals of the ECT is to protect foreign investments, and to do so it establishes a system of dispute resolution between investors and host states. Art. 26 ECT establishes a system of investor-state dispute settlement, under which an investor may submit a dispute for resolution before an arbitral tribunal. The European Union is a party to the ECT alongside its Member States. Whereas the CJEU has addressed the compatibility of intra-EU BITs with EU law, it had not yet addressed the question of whether ISDS provisions in the Energy Charter Treaty were compatible with EU law.
How did this case come before the EU Court? The underlying dispute was between the Republic of Moldova and Komstroy LLC, a Ukrainian investor. Energoalians, a Ukrainian electricity distributor (later succeeded by Komstroy), initiated the arbitration procedure against the Republic of Moldova under art. 26 ECT. In 2013, the arbitral tribunal declared that it had jurisdiction and found that Moldova had failed to comply with its obligations under the ECT. Under the French Code of Civil Procedure, a party may bring an action for annulment of an arbitral award delivered in France where the tribunal incorrectly declared itself to have jurisdiction. Moldova brought an action for annulment of the 2013 award, challenging jurisdiction. In 2016, the Paris Court of Appeal annulled the award on the grounds that the tribunal had wrongly declared itself to have jurisdiction, since the dispute did not involve an “investment” within the meaning of the ECT. This decision was appealed, and in 2018 the Cour de cassation set aside the 2016 judgment and referred the case back to the Paris Court of Appeal.
This required the Paris Court of Appeal to interpret provisions of the ECT (art. 1(6) and art. 26(1)). The Court of Appeal referred three questions to the CJEU. Each of these questions related to the interpretation of the concepts of “investment” and “investor” in the ECT. One might assume, then, that this would involve the CJEU interpreting the provisions of the ECT, applying principles of treaty interpretation in the VCLT binding on the EU and the prevailing understanding of “investment” in the ECT context. However, as explained below, the CJEU views the ECT as an act of EU law and addresses the questions by applying its familiar narrative of the autonomy of EU law.
At first glance, the case does not have any EU connection beyond the fact that the seat of arbitration was in France, an EU Member State. The case involved a dispute between two parties, a Ukrainian investor and Moldova, both of whom were outside the European Union legal order. Indeed, the Court and the Advocate General addressed this issue, and discussed the question of jurisdiction in some detail. AG Szpunar mentioned that the dispute “appears at first sight to be unconnected with the European Union” and “might be called a ‘purely external’ situation”. The Council of the European Union, Hungary, Finland and Sweden argued that the Court could not provide a preliminary reference as the dispute did not involve the interpretation or application of EU law. The Court thus had to decide whether it had jurisdiction to hear the dispute.
Why did the CJEU have jurisdiction? First, the CJEU recalls that it has jurisdiction to give preliminary rulings concerning the validity and interpretation of acts of the European Union. This includes the jurisdiction to interpret international agreements that were concluded by the Union in accordance with arts 217 and 218 TFEU. This is because, according to the Court’s case law, an international agreement concluded by the Union is an “act of the institutions” and can therefore be subject to the Court’s preliminary rulings. The Court and AG find support for this in Haegeman and jurisprudence regarding the interpretation of international agreements, including Lesoochranárske zoskupenie and Aebtri. In Haegeman, the Court famously decided that international agreements concluded by the Union are acts of the institutions, and that the provisions of such agreements form “an integral part of Community law” from the moment they enter into force.
Does that mean, then, that the CJEU has jurisdiction to interpret the provisions of any international agreement binding on the Union? In Haegeman the Court added that it has the power to interpret an agreement “in so far as [it] concerns the [Union]”. Similarly, in Andersson and Wåkerås-Andersson, a case dealing with the interpretation of the EEA Agreement, the Court held that its “jurisdiction to interpret the EEA Agreement […] applies solely with regard to the [Union]; the Court has no jurisdiction to rule on the interpretation of that agreement as regards its application in the EFTA States”.
In KP, a case involving the interpretation of the Hague Protocol, the Court held that “an agreement concluded by the Council, in accordance with Article 218 TFEU, is, as far as the European Union is concerned, an act of one of its institutions…”. In Bogiatzi, relating to the Warsaw Convention, the CJEU referred to its case law that an agreement concluded by the Union is: “as far as the [Union] is concerned, an act of one of the institutions of the [Union] […] The provisions of such an agreement form an integral part of the [Union] legal order as from its entry into force and, within the framework of that order, the Court has jurisdiction to give preliminary rulings concerning the interpretation of such an agreement”.
According to these cases, the Court has jurisdiction to interpret the provisions of an agreement to which the EU is a party, but only within the framework of the EU legal order.
There were two main arguments against CJEU jurisdiction. The first argument was that the underlying dispute related to a multilateral agreement and involved non-EU parties. The second argument was that the underlying dispute did not involve the interpretation and application of EU law.
III.1. Multilateral agreement argument
The ECT is a multilateral agreement, and the underlying dispute involved non-EU parties. In Andersson and Wåkerås-Andersson and Salzmann mentioned above, the CJEU found, in relation to the EEA Agreement, that it did not have jurisdiction to rule on the interpretation of an agreement regarding its application to non-EU states. In Komstroy, AG Szpunar distinguished those two cases dealing with the EEA. Unlike the EEA, the ECT does not have any central body responsible for giving authoritative interpretations of the treaty. “In the present case, the referring court is asking the Court to interpret provisions of an international agreement which are not interpreted uniformly, and in accordance with the case-law of the Court, in disputes outside the European Union and which could, in principle, also be applied to situations internal to the EU legal order”.
The Court distinguished the two EEA-related cases on the grounds that, “the referring courts had to apply the EEA Agreement to situations that did not fall within the EU legal order”. The Court did not really engage with the argument relating to the multilateral nature of the ECT. The argument that the ECT is different because it is an “ordinary multilateral treaty” was made by AG Wathelet in Achmea. The contracting parties to the ECT did not establish a central court or tribunal to provide an authoritative interpretation of the ECT and they certainly did not intend to place the authoritative interpretation of the ECT in the hands of the EU Court of Justice. Yet the EU Court essentially took that role upon itself, reasoning that this is required to prevent divergences in interpretation of the ECT.
III.2. EU law interpretation argument
The second argument against jurisdiction was that the underlying dispute did not involve any interpretation of EU law, since it only dealt with the ECT. Although the Court views the ECT as an “act of the institutions”, it would not have jurisdiction to provide a preliminary reference unless the dispute involves the interpretation and application of EU law. The Court acknowledged that it does not have jurisdiction to interpret the international agreement in the context of a dispute that is not covered by EU law. Yet the Court found it has jurisdiction to interpret the relevant provisions of the ECT because of the need to prevent the possibility of future divergences in interpretation. Since it is possible that an arbitral tribunal might interpret a provision of the ECT in a way that diverges from the interpretation of the CJEU, the Court reasoned, it has jurisdiction.
The Court gives the impression that this is simply a reading of its existing case law: “the Court has held that, where a provision of an international agreement can apply both to situations falling within the scope of EU law and to situations not covered by that law, it is clearly in the interest of the European Union that, in order to forestall future differences of interpretation, that provision should be interpreted uniformly, whatever the circumstances in which it is to apply”. AG Szpunar similarly advised: “where a provision can apply both to situations falling within the scope of EU law and to situations not falling within that scope, it is clearly in the EU interest that, in order to forestall future differences of interpretation, that provision should be interpreted uniformly”.
The Court and AG base this on a passage in Hermès. Yet they have both modified the language used in that case. Compare the original passage in Hermès: “where a provision can apply both to situations falling within the scope of national law and to situations falling within the scope of Community law, it is clearly in the Community interest that, in order to forestall future differences of interpretation, that provision should be interpreted uniformly”.
Hermès dealt with a provision falling within both national law and EU law. The Court and AG Szpunar have replaced “national law” in Hermès with a broader notion of “a situation not falling within the scope of EU law” without explaining why. As a consequence, the Court significantly expanded its jurisdiction to include the interpretation of international agreements involving non-EU parties in a dispute outside the EU legal order.
The Court justifies this expansion of its jurisdiction on the grounds that there is a risk of future divergences in interpretation: the Court “could find it necessary, in a case falling directly within the scope of EU law, such as an action concerning a dispute between an operator of a third country and a Member State, to rule on the interpretation of those same provisions of the ECT”. Those familiar with the Court’s case-law on autonomy, including Opinion 2/13 and BITS cases, will recall that the Court is highly concerned with preventing possible future divergences. Rather than show trust in the EU legal order to address divergences if and when they arise, the Court requires international law to adjust to EU law in order to prevent the (even remote) possibility that a clash may occur.
Finally, the Court also justifies its jurisdiction on the basis that the seat of the main proceedings was in Paris, situated in an EU Member State. The parties to the dispute had voluntarily chosen to submit the dispute to an ad hoc arbitral tribunal in an EU Member State. Since EU law is part of the law of every Member State, the proceedings necessarily involve the preliminary rulings procedure. Moreover, the Court appears to give some discretion to the referring Court to decide that it required a reading of EU law. When the parties instituted proceedings in Paris, it is unlikely they envisaged that the dispute would come before a court in Luxembourg. Yet for the CJEU, the seat of arbitration was a sufficient jurisdictional trigger to involve the CJEU. Law firms are already advising their clients to avoid the consequences of Komstroy by referring disputes to arbitration with a seat outside of the EU.
IV. Intra-EU application of the ECT
Having found that it had jurisdiction, the Court could then turn to the questions posed by the referring Court involving the concept of “investment” in the ECT. Yet, the Court found that, in order to answer that question, it was first necessary to specify which disputes could be brought before an arbitral tribunal pursuant to art. 26 ECT. At this point, the Court goes off on somewhat of a tangent to discuss: “whether a dispute between a Member State and an investor of another Member State concerning an investment made by the latter in the first Member State may be subject to arbitration proceedings under Article 26(2)(c) ECT…”.
In essence, the Court discusses whether EU law precludes the intra-EU application of the ECT. However, the Court does not fully explain why this is relevant to addressing the questions posed by the referring Court. The operative part of the judgment only refers to the meaning of “investment” and does not refer to the issue of intra-EU dispute settlement. While this is strictly speaking obiter, the finding is nevertheless important, and sends a clear message about the Court’s view on how it would decide a case involving that question. Given the importance of this question, however, the Court should have addressed the issue in a case where that was the main issue in dispute.
This question – whether the dispute settlement mechanism established by the ECT is compatible with EU law – was particularly relevant since the CJEU’s 2018 judgment in Achmea. As mentioned above, the CJEU found that EU law, particularly the principle of autonomy of EU law, precludes “a provision in an international agreement concluded between Member States such as Article 8 of the [Dutch-Slovak BIT]” between two EU Member States. Some commentary argued that the impact of Achmea would be limited and unlikely to affect a multilateral treaty such as the ECT. Dashwood, for instance, argues that in the operative part of the Achmea judgment, the CJEU added “such as Article 8 of the [Dutch-Slovak BIT]” to limit the application of the judgment only to intra-EU agreements that share more or less the same properties as the Dutch-Slovak BIT, and not, as some had argued, to all international arbitration agreements between EU Member States. Conversely, the term ‘such as’ could be read, not as limiting the effects of Achmea, but as signalling that it could be applied to other treaties, even multilateral ones such as the ECT.
Another reason put forward to distinguish the situation in Achmea from the intra-EU application of the ECT was that, under ECT arbitration, a tribunal would not be called upon to “interpret or apply EU law”. Art. 26(6) ECT sets out that a tribunal shall decide disputes “in accordance with this treaty and with applicable rules and principles of international law”. Dashwood contrasts this provision with the applicable law provisions in the Dutch-Slovak BIT, which includes “the law in force of the Contracting Party concerned” and to the provisions of “other relevant agreements between the Contracting Parties” (art. 8(6)). The BIT includes a reference to EU law, whereas the ECT does not. Investment tribunals that heard arguments related to the intra-EU objection also found that they do not lack jurisdiction on this basis of Achmea. They tended to restrict the application of Achmea to bilateral investment treaties that did not include the EU as a party: “it is clear that Achmea pertains only to BITs concluded between EU Member States”. According to this reasoning, since the ECT is a multilateral treaty, to which the EU is a party, the findings in Achmea do not apply. This idea was confirmed in 2021: “the ECT is a multilateral agreement to which the EU is itself a signatory. The EU therefore consented to its dispute resolution provisions. It is difficult to see how the ECT would violate EU principles of mutual trust, sincere cooperation or the autonomy of EU law in such circumstances”.
This could be viewed as the narrow reading of Achmea. It contrasts with the more expansive reading supported by the European Commission and a number of EU Member States. According to this expansive reading, not only is the intra-EU application of the ECT incompatible with EU law, it also has the effect of negating the offer of arbitration, hence leaving the arbitral tribunal with no jurisdiction.
IV.1. Does an ECT tribunal interpret EU law?
In Achmea, the Court looked at whether the arbitral tribunal would be called upon to interpret or apply EU law. “Even if […] that tribunal, despite the very broad wording of Article 8(1) of the BIT, is called on to rule only on possible infringements of the BIT, the fact remains that in order to do so it must, in accordance with Article 8(6) of the BIT, take account in particular of the law in force of the contracting party concerned and other relevant agreements between the contracting parties”. The Court found problems with the fact that an arbitral tribunal established under the BIT could be called upon to interpret EU law. According to AG Szpunar in Komstroy, art. 26 ECT would similarly also allow a tribunal to hear disputes involving the interpretation of EU law. Yet art. 26 ECT sets out the applicable law to include “applicable rules and principles of international law”. According to the AG, EU law has a dual character: “given the nature and characteristics of EU law, that law must be regarded both as forming part of the law in force in every Member State and as deriving from an international agreement between the Member States. On that basis, the arbitral tribunal established under Article 26 of the ECT may, where necessary, be obliged to interpret or even apply EU law”. This echoes the Court’s reasoning in Achmea: “Given the nature and characteristics of EU law […] that law must be regarded both as forming part of the law in force in every Member State and as deriving from an international agreement between the Member States”. The question of whether “applicable rules and principles of international law” includes EU law was addressed by arbitral tribunals, such as in Greentech v Italy: “In the context of the arbitral jurisdiction created by the ECT, reference to “international law” cannot be stretched to include EU law, absent doing violence to the text which would be impermissible under the Vienna Convention on the Law of Treaties […] The Tribunal has not been called upon to apply EU law, since Claimants asserted breaches of the ECT and international law, but not of EU law”.
Is EU law international? This conception of EU law – simultaneously domestic law, international law, and an “independent source of law” – potentially allows the CJEU to interpret the provisions of any agreement to which the EU is a party whenever the Court sees this as necessary to avoid possible future divergences in interpretation. Whereas EU Member States have consented to the CJEU providing the authoritative interpretation of EU law, contracting parties to an international agreement such as the ECT have not given the CJEU the authority to interpret the agreement. The Court finds that “the ECT itself is an act of EU law. It follows that an arbitral tribunal such as that referred to in Article 26(6) ECT is required to interpret, and even apply, EU law”. This reading may be technically accurate from an EU law perspective, but the Court downplays the fact that the ECT is not just an act of EU law, but also a multilateral treaty under public international law that involves other treaty parties, and has its own methods of dispute settlement. Could the EU Court also apply similar reasoning to give interpretations of, say, the UN Convention on the Law of the Sea (UNLCOS) in a non-EU dispute, because the UNCLOS is an act of EU law, and the Court must prevent the possibility of future diverging interpretations?
IV.2. Autonomy of the EU Legal order
Komstroy is another judgment building upon the CJEU’s conception of autonomy. Yet these judgments tend to be devoid of clear legal reasoning to explain how and why the principle is applied. It is understandable that academic commentators are left perplexed: “it is however difficult to predict its consequences and the ECJ is criticized for stretching its limits beyond what is legally necessary”. As with Achmea, time will tell how the Komstroy judgment will be given effect outside the EU legal order. Arbitral tribunals will likely treat it in a similar manner, downplaying its legal effects and limiting it to the EU legal order. While Komstroy is important insofar as the Court finds that intra-EU dispute settlement in the ECT is incompatible with EU law, the judgment does not reveal much new about the principle of autonomy and how it is applied. I have argued against a ”checklist approach” whereby international agreements are analysed only according to a set of criteria that must be satisfied. Rather, given that autonomy is a structural principle of EU law, it will continue to be applied in ways that are specific to the legal and political context of the agreement in question.
Beyond the findings on the ECT, the lasting impact of the judgment is likely to be in the way the Court expanded its jurisdiction to be involved in disputes that have no EU law connection, justified by the need to prevent possible future divergences in interpretation.
The CJEU has a central role to provide authoritative interpretations of EU law. This is justified on the grounds that EU law requires a body to provide such interpretation in order to prevent the law’s fragmentation. The CJEU’s judges are appointed by the EU Member States and they have relevant expertise and experience in EU law. This authority does not extend to the interpretation of multilateral agreements such as the ECT, where the contracting parties decided not to put the authoritative interpretation of the agreement in the hands of a central court.
The CJEU will need to find ways to live with these forms of dispute settlement that exist outside the EU legal order, especially those that allow diverging interpretations and application of the law. When the CJEU has allowed some forms of investor-state dispute settlement to exist, as in Opinion 1/17, this was only possible where the agreement includes safeguards specifically designed to protect EU autonomy. A truly autonomous legal order would not require such safety nets and would trust in its own constitutional order to address the reality of diverging interpretations by other courts and tribunals. Yet the rationale for the Court’s autonomy narrative is often based on arguments related to the fragility of that legal order.
Even if the EU and the Member States were to renegotiate or withdraw from the ECT, the CJEU will still have to find ways to coexist with these other forms of dispute settlement outside the investment law context. Arbitral tribunals tend to approach the issue through the lens of public international law, applying, for example the VCLT and rules of treaty conflict. The CJEU could have similarly approached the issue as one of treaty interpretation, engaging in a dialogue with other international dispute settlement bodies and using reasoning that would resonate with arbitral tribunals deciding on their jurisdiction. Rather, its starting point is the autonomy of EU law. On the other side, arbitral tribunals have not shown willingness to engage in such dialogue either, using similarly blunt language to downplay the effects of CJEU judgments. The Commission and some Member States will use the Komstroy judgment to urge arbitral tribunals to decline jurisdiction in the case of intra-EU disputes. Given the analysis adopted by the CJEU, which is based entirely on autonomy reasoning, rather than engaging with international law arguments, Komstroy may, like Achmea, have little impact outside the EU legal order.
* Lecturer, City Law School, University of London, firstname.lastname@example.org. Thanks to Panos Koutrakos and Benedikt Pirker for their comments on earlier drafts.
 ICSID decision on jurisdiction, applicable law and liability of 30 November 2012 ICSID case n. ARB/07/19 Electrabel SA v The Republic of Hungary.
 Ibid. 4.120.
 In A Roberts, Is International Law International? (Oxford University Press 2017) Roberts challenges the concept of international law as a universal system.
 T Schilling, ‘The Autonomy of the Community Legal Order: An Analysis of Possible Foundations’ (1996) HarvIntlLJ 389, 403–404: “At least at its inception, the European Community was clearly a creature of international law. As there are no indications that a revolution in its legal sense has subsequently occurred […] the European Treaties are still creatures of international law”.
 Case C-741/19 Republic of Moldova v Komstroy LLC ECLI:EU:C:2021:655 para. 49.
 Ibid. para. 50.
 Ibid. para. 66.
 Case C‑284/16 Slowakische Republik (Slovak Republic) v Achmea BV ECLI:EU:C:2018:158.
 Case C‑386/08 Brita GmbH v Hauptzollamt Hamburg-Hafen ECLI:EU:C:2010:91 para. 42. See case C‑266/16 Western Sahara Campaign UK ECLI:EU:C:2018:118 para. 58: ”In order to interpret those provisions [of an agreement], it is necessary to refer to the rules of customary international law reflected by Article 31 of the Vienna Convention, which are binding on the EU institutions and are part of the EU legal order”.
 Preamble, ECT cit.
 Art. 26 ECT, entitled “Settlement of disputes between an investor and a Contracting Party”.
 Council and Commission Decision 98/181/EC, ECSC, Euratom of 23 September. The ECT was approved by the Union in September 1997. On 31 December 2014, Italy notified its withdrawal from the ECT.
 Joined cases C-798/18 and C-799/18 Federazione nazionale delle imprese elettrotecniche ed elettroniche (Anie) and Others ECLI:EU:C:2021:280 did not address the question.
 Art. 1520 Code de procédure civile: “An action for annulment is available only in the following cases: (1) Where the arbitral tribunal wrongly declared itself to have or not to have jurisdiction, or (2) Where the arbitral tribunal was improperly constituted, or (3) Where the arbitral tribunal issued a ruling without fulfilling the mandate entrusted to it, or (4) Where the adversarial principle was not observed or (5) Where the recognition or enforcement of the award is contrary to international public policy”.
 Case C-741/19 Republic of Moldova v Komstroy cit. para. 20
 Case C‑741/19 Republic of Moldova v Komstroy LLC ECLI:EU:C:2021:164, opinion of AG Szpunar, para. 2.
 Ibid. para. 26.
 Republic of Moldova v Komstroy cit. para. 21.
 Art. 267 TFEU: “The Court of Justice of the European Union shall have jurisdiction to give preliminary rulings concerning: […] (b) the validity and interpretation of acts of the institutions, bodies, offices or agencies of the Union…”.
 Case C-180/73 R & V Haegeman v Belgian State ECLI:EU:C:1974:41.
 Case C‑240/09 Lesoochranárske zoskupenie VLK v Ministerstvo životného prostredia Slovenskej republiky ECLI:EU:C:2011:125 para. 30: “Within the framework of that legal order the Court therefore has jurisdiction to give preliminary rulings concerning the interpretation of such an agreement” (regarding the Aarhus Convention). Emphasis added.
 Case C-224/16 Asotsiatsia na balgarskite predpriyatia za mezhdunarodni prevozi i patishtata (Aebtri) v Nachalnik na Mitnitsa Burgas ECLI:EU:C:2017:880 para. 50 (regarding the Customs Convention on the International Transport of Goods under Cover of TIR Carnets (“TIR Convention”)).
 Haegeman cit. para. 5: “The provisions of the Agreement, from the coming into force thereof, form an integral part of Community law” (regarding the Agreement of Association between the European Economic Community and Greece).
 Ibid. para. 4.
 Case C‑321/97 Ulla-Brith Andersson and Susannne Wåkerås-Andersson v Svenska staten ECLI:EU:C:1999:307 para. 26.
 Case C-83/17 KP v LO ECLI:EU:C:2018:408 para. 24.
 Ibid. Emphasis added.
 Case C‑301/08 Irène Bogiatzi v Deutscher Luftpool and Others ECLI:EU:C:2009:649 para. 23. Emphasis added.
 Bogiatzi cit. para. 23. Emphasis added.
 Case C‑300/01 Salzmann ECLI:EU:C:2003:283 para. 66: “jurisdiction to interpret the EEA Agreement […] applies solely with regard to the [Union]”.
 Republic of Moldova v Komstroy, opinion of AG Szpunar, cit. para. 40: “the ECT does not establish any court or tribunal responsible for ensuring the uniform interpretation of its provisions, in a manner consistent with the Court’s interpretation within its legal order. The ECT is intended to be interpreted only in the course of the settlement of disputes by various arbitral or State tribunals in the Contracting Parties, which therefore cannot avoid divergences in interpretation”.
 Ibid. para. 44.
 Republic of Moldova v Komstroy cit. para. 37.
 Case C-284/16 Achmea ECLI:EU:C:2017:699, opinion of AG Wathelet, para. 43: “That multilateral treaty on investment in the field of energy operates even between Member States, since it was concluded not as an agreement between the Union and its Member States, of the one part, and third countries, of the other part, but as an ordinary multilateral treaty in which all the Contracting Parties participate on an equal footing. In that sense, the material provisions for the protection of investments provided for in that Treaty and the ISDS mechanism also operate between Member States. I note that if no EU institution and no Member State sought an opinion from the Court on the compatibility of that treaty with the EU and FEU Treaties, that is because none of them had the slightest suspicion that it might be incompatible”.
 Republic of Moldova v Komstroy cit. para. 23.
 Ibid. para. 28.
 Ibid. para. 29. Emphasis added.
 Republic of Moldova v Komstroy, opinion of AG Szpunar, cit. para. 37. Emphasis added.
 The reference to case C-53/96 Hermès International v FHT Marketing Choice BV ECLI:EU:C:1998:292 and other judgments in footnote 29 of the judgment is qualified “to that effect”.
 Hermès cit. para. 32. Emphasis added.
 Republic of Moldova v Komstroy cit. para. 31. Emphasis added.
 Case C-205/06 Commission v Austria ECLI:EU:C:2009:118; case C-249/06 Commission v Sweden, ECLI:EU:C:2009:119; case C-118/07 Commission v Finland ECLI:EU:C:2009:715.
 “In fact, so crucial does the CJEU view this preservation of uniformity that, in Achmea, the Court does not even concern itself whether the arbitral tribunal has actually made any EU law interpretations—whether conflicting or not—instead, the Court seems to be concerned with the mere possibility of a dispute involving questions of EU law”. See J Lee, ‘The Empire Strikes Back: Case Note on the CJEU Decision in Slovak Republic v Achmea BV, March 6, 2018’ (2018) Contemporary Asia Arbitration Journal 137, 145.
 Republic of Moldova v Komstroy cit. para 40: “In order to answer that question, it is necessary, first of all, as several Member States which have participated in the proceedings have observed, to specify which disputes between one Contracting Party and an investor of another Contracting Party concerning an investment made by the latter in the area of the former may be brought before an arbitral tribunal pursuant to Article 26 ECT”.
 Ibid. para. 47.
 Achmea cit. para. 60.
 A Dashwood, ‘Article 26 ECT and intra-EU Disputes – the Case Against an Expansive Reading of Achmea’ (2021) ELR 415, 423: “The fact that the Court of Justice reformulated the first and second questions so as to include an explicit reference to ‘Article 8 of the BIT’ is a strong indication that the Court intended its ruling to apply specifically to arbitration clauses in intra-EU agreements displaying the particular characteristics of art.8 of the Netherlands/Slovakia BIT”.
 See ICSID decision on the Achmea Issue of 31 August 2018 case ARB/12/12 Vattenfall AB and others v Federal Republic of Germany para. 161: “It remains unclear what alleged rule of international law arising from the ECJ Judgment exists and is of application to the present case. The ECJ’s reasoning was not specifically addressed to investor-State dispute settlement under the ECT. While there is a certain breadth to the Court’s wording, addressing provisions “such as” the dispute resolution provision of the BIT in that case, it is an open question whether the same considerations necessarily apply to the ECT”. See S Hindelang, ‘Conceptualisation and Application of the Principle of Autonomy of EU law – the CJEU's Judgment in Achmea Put in Perspective’ (2019) ELR 383, 384: “Furthermore, the [Achmea] judgment’s reasoning seems not to be limited to bilateral investment treaties between the Member States, but will most likely also apply to the Energy Charter Treaty (ECT) in an intra-EU context”.
 See S Gáspár-Szilágyi and M Usynin, ‘The Uneasy Relationship between Intra-EU Investment Tribunals and the Court of Justice’s Achmea Judgment’ (2019) European Investment Law and Arbitration Review 29, 35.
 ICSID award of 16 May 2018 case ARB/14/1 Masdar Solar & Wind Cooperatief UA v Kingdom of Spain para. 680.
 Ibid. paras 678–83: “The ECT is not such a treaty. Thus, the Achmea Judgment does not take into consideration, and thus it cannot be applied to, multilateral treaties, such as the ECT, to which the EU itself is a party”.
 Stockholm Chamber of Commerce final award of March 2021 case n. 2017/060 FREIF Eurowind Holdings Ltd v Kingdom of Spain.
 See CI Nagy, ‘Intra-EU Bilateral Investment Treaties and EU Law After Achmea: “Know Well What Leads You Forward and What Holds You Back”’ (2018) German Law Journal 982, 996: “while Achmea features an anti-arbitration attitude – which may guide future cases – the ruling’s holding is very narrow”.
 Achmea cit. para. 39.
 Republic of Moldova v Komstroy, opinion of AG Szpunar, cit. para. 74: “In the first place, I would point out that, like the dispute settlement mechanism at issue in the judgment in Achmea, Article 26 of the ECT allows disputes which may involve the interpretation of EU law to be brought before an investment arbitral tribunal”.
 Ibid. para. 75. Emphasis added.
 Achmea cit. para. 41.
 Stockholm Chamber of Commerce final award of 23 December 2018 case n. V 2015/095 Greentech Energy Systems A/S et al v Italian Republic para. 397.
 Case C-6/64 Costa v ENEL ECLI:EU:C:1964:66: “[t]he law stemming from the Treaty, an independent source of law, could not, because of its special and original nature, be overridden by domestic legal provisions, however framed, without being deprived of its character as Community law and without the legal basis of the Community itself being called into question”.
 Republic of Moldova v Komstroy cit. para. 49-50.
 For theoretical discussions of autonomy, see C Contartese, ‘The Autonomy of the EU Legal Order in the ECJ’s External Relations Case Law: From the “Essential” to the “Specific Characteristics” of the Union and Back’ (2017) CMLRev 1627; I Govaere, ‘Interconnecting Legal Systems and the Autonomous EU Legal Order: A Balloon Dynamic’ in I Govaere and S Garben (eds), The Interface Between EU and International Law: Contemporary Reflections (Hart Publishing 2019); C Eckes, EU Powers Under External Pressure: How the EU’s External Actions Alter its Internal Structures (Oxford University Press 2019).
 C Contartese and M Andenas, ‘EU Autonomy and Investor-State Dispute Settlement under inter se Agreements between EU Member States: Achmea’ (2019) CMLRev 157, 158. P Koutrakos, ‘Editorial: What is the Principle of Autonomy About?’ (2018) ELR 1-2; P Koutrakos, ‘Editorial: But Seriously, What is the Principle of Autonomy Really About?’ (2018) ELR 293-294.
 J Odermatt, International Law and the European Union (Cambridge University Press 2021) 180.
 See M Parish, ‘International Courts and the European Legal Order’ (2012) EJIL 141, 142: “A new threat has recently emerged to the consistent application of EU law, namely interpretation of EU law by the ever growing range of international tribunals that sit outside the domestic legal order of any particular state”. See P Koutrakos, ‘Managing Inter-Legality: Conceptualizing the European Union’s Interactions with International Investment Law’ in J Klabbers and G Palombella (eds), The Challenge of Inter-legality (Cambridge University Press 2019).
 E Kassoti and J Odermatt, ‘The Principle of Autonomy and International Investment Arbitration: Reflections on Opinion 1/17’ (2020) QuestIntlL 5.
 See BH Pirker and S Reitemeyer, ‘Between Discursive and Exclusive Autonomy – Opinion 2/13, the Protection of Fundamental Rights and the Autonomy of EU Law’ (2015) CYELS 168, 171 discussing the discursive approach to autonomy: “A discursive approach pursues a more integrationist rationale and requires courts to accept as a matter of principle that there are other courts with their own jurisdiction”.
 M Szpunar, ’Is the Court of Justice Afraid of International Jurisdictions?’ (2017) PolishYIL 125, 140 “The law of the EU is a very unique legal order that cannot exist without the legal orders of the Member States. The very existence of this fragile legal order depends on establishing an adequate balance - on the one hand between EU law and national law; and on the other between EU law and international law. This is the task of the Court of Justice (alone)”. C Eckes, EU Powers Under External Pressure cit. 196: “EU law depends for its constitutional character and its ability to ensure the effectiveness of EU law on its autonomous character, that is, its self-referential nature of not depending on national and international law for its validity and interpretation”.
 P Koutrakos, ‘Editorial: But Seriously, What is the Principle of Autonomy Really About?’ cit. 294: “[...] the principle [or autonomy] may appear to be all about conflict: it builds an antagonistic relationship with international law and does not see any space for EU law to be understood and applied in pragmatic co-existence with international dispute settlement systems”.
 See S Hindelang, ‘Conceptualisation and Application of the Principle of Autonomy of EU law’ cit. 393: “In fact, investment tribunals have not shown any pragmatism on their side either. Furthermore, it is not just the blunt hostility of investment tribunals towards EU law and the CJEU which is worrisome. There are also other problematic issues, relating to their procedural integrity, for example, which might have called into question the meaningfulness of any dialogue”. K Särkänne, ‘EU Law in Investment Arbitration: A View From International Arbitral Tribunals’ (2021) Europe and the World: A Law Review 18: “Whereas the tribunals could be criticised for not showing sufficient recognition of EU law, from the viewpoint of international law the demand from the EU for the jurisdiction to be denied appears to be rooted in the internal structures of the EU legal order”.