News of the award in Green Power and Obton v Spain is sinking in. Initial responses indicate that this is no ordinary decision – but rather a ‘major earthquake’, a ‘landmark decision’ and ‘one for the history books’. It may well be: on 16 June 2022, an SCC arbitral tribunal seated in Stockholm declined jurisdiction over claims brought by Danish investors under the Energy Charter Treaty (ECT) because of the intra-EU character of the dispute, following the approach developed by the Court of Justice of the EU (CJEU) in its Achmea and Komstroy judgments. In our view, this marks an overdue recognition that investment tribunals addressing disputes between EU nationals and EU member States cannot ignore the primacy of EU law. But the award’s reasoning is unlikely to persuade critics: the Tribunal reaches what we think is the correct outcome via a long and winding reasoning that overstretches the principles of treaty interpretation.
In this post, we situate the Green Power award, explain why its reasoning is problematic, and indicate how future tribunals might engage with the so-called ‘clash of Grundnormen’ between EU law and investment law.
- Clashing Grundnormen: Before Green Power
Like so much else, the story of Green Power goes back to Achmea. The CJEU’s judgment of March 2018 had suggested that proceedings based on intra-EU BITs (which are common) could not go ahead, as ISDS clauses included in intra-EU investment treaties violated the EU principles of autonomy and mutual trust. The judgment sent shockwaves through the EU and investment communities, generating very different reactions.
The EU’s response was robust. The Commission’s Communication to the European Parliament and the Council, asserting that EU law must have primacy, urged an end to intra-EU investment arbitration. Member States’ Declarations of 15 and 16 January 2019 also considered ISDS clauses in intra-EU bilateral investment treaties ‘contrary to Union law and thus inapplicable’, affirmed the primacy principle and noted that ‘[t]he same result follows… under general public international law’. Following this, Member States adopted the Treaty for the Termination of Intra-EU BITs (now with 20 parties) while the CJEU, in Komstroy, extended the Achmea logic to the ECT arbitration clause.
Investment tribunals, by contrast, resolutely rejected the ‘Achmea objection’, holding States to the commitments made in the jurisdictional clauses of the applicable BITs or Article 26(3) ECT (eg RREEF v Spain), whose meaning they considered plain. Moreover, while many tribunals addressed the principles governing incompatible successive treaties (Articles 59 and 30(3) of the Vienna Convention on the Law of Treaties (VCLT)), they held that neither of these could accommodate the primacy claimed by EU law. (Professor Kohen’s dissent in Adamakopoulos so far seemed a ‘lone dissent’). So the cases continued, notwithstanding Achmea, and the ‘clash of Grundnormen’ resurfaced in follow-up proceedings (State aid, recognition and enforcement, etc).
- Overcoming ‘Binary Logics‘ through Interpretation: The Green Power Approach
Given these entrenched positions, the Green Power Tribunal’s desire to overcome the ‘binary logic of an either “insider” or “outsider” perspective with respect to EU law’ (§332) is commendable. The Tribunal sought a way out via an interpretation of Article 26(3)(a) ECT, which contains States‘ ‘unconditional consent to the submission of a dispute to international arbitration’. The Tribunal considered this provision to be ‘[o]n its wording … unqualified’ and ‘unconditional’ (§341), but felt that to ‘stop … there‘ would ‘ignore the complexities of this case’ (§343).
What followed was a detailed explanation of why this ‘unqualified’ and unconditional‘ wording could not be given effect. In 140 paragraphs of interpretation based on VCLT Article 31, the Tribunal concluded that ECT Article 26(3)(a) could not be interpreted without reference to EU law, and that the application of EU law entailed the invalidity of Spain’s offer to arbitrate. In support, it relied on the ‘evidence’ informing the EU’s approach to investment arbitration sketched out above, notably Achmea, Komstroy, and the Declarations of 15 and 16 January 2019, which the Tribunal considered to have authentically interpreted Article 26(3)(a).
The Tribunal also added summary remarks on the question of priority in the application of EU law and the ECT, noting ‘that the primacy of EU law … is not a matter of lex specialis or of lex posterior, but one of lex superior’ (§469). This superiority meant that Spain was precluded from making a unilateral offer to arbitrate in Article 26 ECT: there was, in short, ‘no unilateral offer by the Respondent which the claimants could accept’ (§§476-477).
- An Alternative Approach: Recognising and Resolving the Treaty Conflict
There is much to commend in the award. The Tribunal was right to avoid the ‘binary logic’ that had, so far, only succeeded in entrenching the positions summarised above. We submit it was also right in rejecting jurisdiction under Article 26(3)(a). But the Tribunal’s way out – to resolve the conflict through treaty interpretation – is problematic. Two difficulties are obvious. First, the Tribunal’s ‘evidence‘ hardly supports its conclusion (see also here). For instance, the Declarations of January 2019 were not made by all treaty parties to the ECT – so how could they amount to an authentic interpretations of Article 26? Second, the Tribunal ends up holding that Article 26(3)(a) does not mean what it ‘unconditionally‘ says – a plain meaning the Tribunal had itself acknowledged (§341).
At the source of this fundamental problem is the Tribunal’s preferred way out: treaty interpretation. Interpretation can assist where conflicts are merely apparent – where the two apparently conflicting rules can be reconciled (Pauwelyn, at 272). This was not the situation in Green Power, however, where two rules plainly clashed: Article 26(3)(a) required Spain to submit to arbitration; EU law required it not to do so. As much as the Tribunal tried, these two rules were genuinely incompatible – as the Tribunal’s brief comments on lex superior seemed to acknowledge (§469). The plain interpretation of Article 26(3)(a) (unconditional consent to arbitration) may not have reflected the ‘complexities’ of the case, but the point of interpretation is definitional: to elucidate the meaning of treaty clauses. The Tribunal was right that interpretation must not be an abstract process (§344), and that the normative environment of the provision to be interpreted must not be overlooked or ignored. But a systemically integrated interpretation, like the one advanced by the Tribunal, was a way to find, rather than resolve, the complexities of the case: a way to reveal the genuine conflict between ECT Article 26(3)(a) and EU law.
A number of earlier tribunals pronouncing on the ‘Achmea objection’ had correctly acknowledged such genuine conflict – and then sought to resolve it. To do so, they notably relied on the lex posterior rule of VCLT Article 30(3), which stipulated that between successive treaties relating to the same subject matter, ‘the earlier treaty applies only to the extent that its provisions are compatible with those of the later treaty’. But inevitably, tribunals before Green Power thought that Article 30(3) VCLT could not accommodate the primacy of EU law over the investment treaty. And perhaps this is plausible: Article 30(3) is not about primacy, but about posteriority; and for that reason, may not fit the Achmea logic, which is, as noted in Green Power, ‘a matter … of lex superior’ (§469).
But this is not the only way to resolve the (acknowledged) conflict between ISDS clauses and EU law. It seems to us that there would have been a better approach — one that takes arguments about lex superior seriously, but that, puzzlingly, has not been pursued by tribunals yet. This approach proceeds from the assumption that States can determine which of their incompatible treaty commitments enjoys priority. They can do so because, outside cases of peremptory rules, they are free to qualify particular treaty rules as superior or inferior. To us, the existence of such a prerogative seems not in doubt, even though the VCLT does not formulate it as a general principle. Even in the absence of a general ‘priority clause’, the idea that superior rules should be given effect is not alien to Article 30, whose para. (1) spells out one priority clause expressly when referring to Article 103 of the UN Charter (which has primacy over both prior and later incompatible treaty commitments). Article 30(2) also addresses priorities, though its approach is selective:
When a treaty specifies that it is subject to, or that it is not to be considered as incompatible with, an earlier or later treaty, the provisions of that other treaty prevail.
As is clear from its wording, Article 30(2) only covers subordination clauses, not claims to priority – but the possibility of such clauses is implicit therein (see eg here). The ILC’s 1966 Commentary notes that States can regulate the relation between their treaty commitments through clauses in an earlier or later treaty. ‘Whatever the nature of the provision’, it continues, ’the clause has necessarily to be taken into account in appreciating the priority of successive treaties relating to the same subject matter’ (Article 26 Commentary , §2). This includes primacy clauses. That States could choose which treaty commitments ought to prevail over others, was so obvious that the ILC did not think it necessary to expressly mention them in Article 30 (ibid, §8). The specific cases of priority mentioned in Article 30 were included because of their peculiar characteristics: Charter Article 103 was included because of the importance of the Charter in modern international law (ibid, §3); subordination clauses because they could influence the operation of the general rules concerning the effects of conflict clauses and in particular the fact that they may affect the position of third States (ibid, §4). But these were just illustrations of a more general principle.
- Giving Effect to the Intentions of EU Member States
In our assessment, this general priority principle provides a more convincing path to resolving the conflict between intra-EU ISDS clauses and EU law. It can apply because the parties are clear that, between those two incompatible rules, EU law clearly enjoys priority. It does so as a matter of EU law (because of the primacy principle), but also as a matter of general international law. Member States in fact said so expressly in the Declaration of 15 January 2019 where they declared ‘all investor-State arbitration clauses contained in bilateral investment treaties concluded between Member States [to be]… inapplicable’, relying on the primacy principle, but also on ‘the relevant provisions of the Vienna Convention on the Law of the Treaties.’
The Declaration, in our assessment, reflects three clear positions taken by EU member States: (i) There is a conflict between EU law and arbitration clauses. (ii) EU law enjoys priority. And (iii) this priority is recognised (also) as a matter of international law. So far, investment tribunals have failed to give effect to these clearly expressed positions. Tribunals before Green Power felt they could not reach the outcome clearly desired by EU member States. The Green Power Tribunal reached the desired outcome, but felt it could only do so by interpreting away the conflict. It thereby managed (unlike the earlier tribunals) to give effect to the parties’ intentions, but did so via an unconvincing argumentative avenue.
This blog post is an attempt to encourage further thinking about the interaction between EU law and ISDS jurisdictional clauses. The solution sketched out above, based on a general priority principle implicit in Article 30 VCLT, will need to be further developed. But the preceding considerations do suggest, in our view, that there is a way of reflecting EU law primacy within the general law of treaties. Only by exploring this, we submit, will investment tribunals be able to overcome the ‘binary logic of an either “insider” or “outsider” perspective with respect to EU law’ (cf. Green Power, §332).