In the award on Green Power v Spain, an SCC arbitral tribunal decided that, while operating under the Energy Charter Treaty (ECT), it had no jurisdiction over claims of investors from Denmark, because the dispute in question had an intra-EU character. The dispute was therefore, pursuant to the approach taken in decisions of the EU judiciary in Achmea and Komstroy, excluded from ECT arbitration. Even though EU and ECT legal orders are based on separate treaty frameworks, the Tribunal chose to present the whole issue as one of interpretation of a treaty, namely ECT.
As a starting point of its interpretative exercise, the Tribunal professed the adherence to Article 31 of the 1969 Vienna Convention on the Law of Treaties. It suggested that Article 26(5)(a) ECT indicates that norms other than “may be applicable to determine the validity of the agreement to arbitrate” (para. 342 of the Award). However, that provision has more to do with the effect of the otherwise valid consent to arbitration, than with (some external) criteria that determine whether the consent to ECT arbitration is valid in the first place. It was clear anyway that Article 26 ECT did not accommodate any restraining effect of EU legal order and, as though reacting to this conclusion, the Tribunal has stated that “The ordinary meaning of the terms must be clear not only on paper but as applied to the relevant facts of the case.” The Tribunal then alluded to the passage in ICJ’s 1997 decision on Gabcikovo-Nagymaros that “it is the purpose of the Treaty, and the intentions of the parties in concluding it, which should prevail over its literal application” (paras 344-345).
It is clear that the Tribunal has understood that ECT’s plain and ordinary meaning was not delivering the outcome it was keen to endorse, which perhaps explains why the Tribunal felt it had to make the above far-reaching statement about the overall relevance to interpretation. That statement involves a confusion between fact and law. Interpretation is inherently, and exclusively, a matter of law. An interpretative exercise involves the use of legal standards whose relevance cannot depend on facts which the treaty in question purports to regulate. An assertion that a treaty’s meaning could be influenced by facts it regulates would amount to an absurdity. Moreover, the ICJ’s above statement to which the Tribunal referred was not made in the interpretation context, but in a rather different context which related to adaptation of a violated treaty to evolving circumstances – a phenomenon which simply did not arise in Green Power.
The Tribunal then proceeded to identify the meaning of ECT’s “Contracting Party” by reference to context as an interpretative element (Article 31(2) VCLT). Here too, the Tribunal has drawn a far-reaching inference from the fact that Article 1(10) ECT refers to the possibility that regional economic international organisations (REIO) could be contracting parties to ECT. From that, the Tribunal has inferred the conclusion that legal order of the EU (which is such an international organisation) is exempt from the dispute settlement jurisdiction available under ECT (paras 350ff.). However, ECT speaks here of the possibility of REIO being a party of ECT, not of a REIO’s internal legal order being exempt from the effect of ECT. It is not clear why ECT refers to REIOs for the sake of exempting them from its own legal framework, as opposed to integrating them into that legal framework. The Tribunal’s “interpretation” exercise has, here again, amounted to drawing far-reaching inferences from words and phrases that are not conclusive enough to sustain those inferences. The same applies to the Tribunal’s reference to Article 25 ECT which emphasises that parties to economic integration agreements (EIAs) are not obliged to extend the latter’s benefits to its non-parties (para. 357). Article 25 speaks here of the calculus of substantive rights and obligations of ECT parties and their nationals, and has nothing to do with the validity of consent to arbitration. Nor does Article 25 require any modification of ECT standards. Another point is that, ECT’s conclusion has long predated the Achmea decision and ECT’s contracting parties were not expected to have a faintest idea about the likelihood that, one day, ECJ would endorse the approach that investment agreements are incompatible with the EU legal order. In addition, when the ECJ endorsed that approach, it did so not in terms of treaty interpretation, but about the incompatibility between EU and certain non-EU legal orders (explained here, among others).
The Tribunal also addressed the issue of subsequent agreement and subsequent practice as an interpretation method according to Article 31(3), sections (a) and (b), VCLT. As an instance of what it saw as subsequent practice, the Tribunal has referred to declarations made by a number of EU States in 2019, stating that investment treaties such as ECT were not applicable on the intra-EU plane and in relations between EU member-States. (paras 365ff.) The truth of the matter is that none of those declarations amounted to a “subsequent agreement between the parties [to a treaty such as ECT] regarding the interpretation of the treaty or the application of its provisions” or to “any subsequent practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation”. Authors of 2019 declarations did not focus on the establishment of ECT’s meaning or its reinterpretation, but merely addressed the issue of ECT’s applicability on the intra-EU plane. All they were concerned with was to exclude the ECT’s effect on the intra-EU plane; hence, they clearly reasoned on terms of normative conflict between two treaty regimes, not on terms of treaty interpretation. The Tribunal’s comparison with the NAFTA Free Trade Commission’s 2001 interpretation of NAFTA (para. 381) is also flawed, because in that case the NAFTA Commission clearly reasoned on the meaning of “fair and equitable treatment” under the NAFTA Agreement and, unlike the case at hand, it stated the position of all parties to NAFTA. Even if NAFTA interpretation was an authentic one, examples the Tribunal placed reliance on by far lack the requisite qualities.
The Tribunal then moved to the area covered by VCLT’s Article 31(3)(c) and placed an emphasis on the so-called “systemic integration” thesis even though, as shown by the body of evidence, “systemic integration” is not part of the treaty interpretation framework and is, moreover, inimical to any interpretative exercise (as explained here, here, here, here, and, specifically with regard to the EU legal order, here). That is the case even though the “systemic integration” thesis is popular among academic writers. What is more concerning, though, is that the Tribunal essentially evades the analysis of practical evidence on this matter and forms its view by reference to the work of one single academic writer (paras 390ff). The Tribunal does not pay attention to what other tribunals may have done in this area, especially to the fact that the International Court of Justice has more recently come to reject the relevance of “systemic integration” in cases such as Djibouti v France, Costa-Rica v Nicaragua, or Jadhav.
The real reason why ECT arbitration should not operate on the intra-EU plane is that EU judiciary and EU member-States have declared that such arbitration is incompatible with the EU legal order. This has nothing to do with the meaning and interpretation of ECT as a treaty. One might still suggest that as the outcome is correct anyway, the damage done by the Tribunal’s pretence to be engaged in treaty interpretation is not that great. However, the problem addressed here cannot be contained in a single situation. The Tribunal has, above all, openly endorsed judicial policies to downplay and misapply the VCLT principles and methods of treaty interpretation. It may be that, from a practising lawyer’s viewpoint, a pragmatic resolution of a dispute is more important than a coherent application of a legal framework (in this case of Articles 31 and 32 VCLT). Nevertheless, in a State-centric legal system of international law, interpretation principles are also arranged in the order of hierarchy, with the effect that the interpretative method that most vividly and obviously demonstrates what the agreed will of States is ought to be given priority over other interpretative methods. In this sense, a tribunal tasked with the application of international law to facts (and collecting half million euros in fees for doing just that), does not simply engage in legal practice, but it also operates as an agency of the international legal system which functions on the basis of the consent of States. Should the Tribunal’s loose handling of interpretative principles, its disregard for the basic distinction between interpretation and normative conflict, or its reliance on one single academic writer shaping its views on important aspects of interpretation be found attractive by other similar tribunals, such persistence in delivering counter-consensual judicial outcomes would do little to secure the overall standing of investment arbitration that has, by now, witnessed repeated backlashes from States.