The Conundrum of the Komstroy Declaration

The European Union (EU) appears relieved by the CJEU’s Komstroy decision, viewing it as a means to exit all pending and future intra-EU arbitrations under the Energy Charter Treaty (ECT). However, a closer look at the Komstroy Declaration through the lens of treaty modification rules – along with Hungary’s differing stance – reveals that withdrawal may be more complex than anticipated.

Background of the EU’s Withdrawal from the ECT

On 28 June 2024, the EU notified its withdrawal from the ECT which will become effective within one year. This announcement followed on the heels of a Declaration signed by 26 of the 27 member states along with the EU which expressed the shared understanding of the signatories concerning the legal consequences of Komstroy ruling.

In Komstroy (2021), the Court of Justice of the EU (CJEU) held that the ECT’s arbitration mechanism, specifically Article 26, does not apply to intra-EU disputes due to the autonomy of EU law (para 98). The Declaration thus holds that all ongoing and future intra-EU arbitrations under the ECT lack legal effect.

However, Hungary, the only member state not signing the main Declaration, issued its own statement based on principles from the Vienna Convention on the Law of Treaties (VCLT). Hungary argued that formally amending the ECT through a bilateral or multilateral treaty would be the correct means to discontinue Article 26’s application. This divergence with the main Declaration reflects underlying tensions in treaty modification and interpretation, and I argue in this blogpost that the EU Declaration stands on a shaky legal footing due to its non-alignment with VCLT and general principles concerning the modification of treaties. The EU’s approach raises critical questions about altering treaties without adherence to recognized legal frameworks and suggests potential ramifications for international treaty law. I make this submission with due recognition of the counter-arguments that may be proffered on the anvil of EU’s lex special regime, conflict of laws, and other facets relevant to the debate.

What Exactly Happened in Komstroy?

In Achmea, the CJEU ruled that arbitration clauses in intra-EU investment agreements conflict with EU law. Building on this, Komstroy applied the same principle to the ECT. The Court interpreted Article 26(6) ECT – which allows tribunals to resolve disputes based on the ECT and other relevant international law – to include EU law as part of the applicable rules (para 48). This interpretation means that the ECT tribunal would have to interpret EU law in their decisions (para 50).

However, under Article 267 of the TFEU, only “courts and tribunals of a Member State” can refer cases to the CJEU for preliminary rulings to ensure consistency in EU law interpretation – an option not available to ECT tribunals. This statutory interpretation of the ECT sparked significant criticism, to be elaborated below, challenging the CJEU’s approach as inconsistent with the principles of the VCLT.

Questionable ‘Interpretation’

The Swiss Supreme Court, in upholding an intra-EU arbitration award under the ECT, expressed its stance on the Komstroy and Achmea (para 7.6). It negated the binding effect of these decisions, noting Switzerland’s non-EU status, and interpreted Article 26(6) ECT based on Art. 31 VCLT, finding the language unambiguous and not permitting an intra-EU carve-out. The Swiss Court asserted that ECT signatories had given unconditional consent to its dispute resolution and criticized the CJEU’s non-consideration of general international law and treaty interpretation principles (para 7.6.5). The Court also dismissed the Achmea Declaration, deeming it non-binding and lacking unanimous EU Member State support, thus failing to meet the criteria of Article 31(2) VCLT as a subsequent agreement or practice (para 7.7.5).

What remains then by analogy of the Komstroy Declaration? From the VCLT lens, it could only be upheld as an interpretative agreement under Article 31(3) VCLT or a modification under Article 41 VCLT.

Can the Komstroy Declaration Be Salvaged as a VCLT Interpretative Tool?

The VCLT provides two interpretative tools: subsequent agreements (Article 31(3)(a)) and subsequent practice (Article 31(3)(b)). Both carry equal legal weight, though the line between them is fluid, as both rely on the “agreement of parties.” The number of parties needed for such an agreement is debated, with ILC’s Draft Conclusion 10 suggesting it may vary; silence from some parties may imply consent if a response would typically be expected. However, a consensualist approach would argue for a high threshold for an agreement or practice to qualify under Article 31(3) VCLT.

The Komstroy Declaration, however, may not meet the standards required for an interpretative act under Article 31 VCLT. Although it operates from the a priori assumption that it holds an interpretative nature, a mere ipse dixit labeling should not be sufficient to qualify a legal text as a valid act of interpretation. In the present case, the Declaration’s signatories don’t include all ECT state parties, and the same limitations likely apply to the proposed plurilateral treaty intended among signatories.

Since the Declaration does not constitute a valid interpretation, it is worth examining whether it could be seen as an amendment or modification under VCLT rules.

Amending or Modifying the ECT in Accordance with the VCLT?

An amendment via Article 40 VCLT is legally unviable, as it requires all parties’ involvement. Further, Article 42 ECT provides a distinct procedure for amendments affecting the treaty as a whole, requiring ratification by three-fourths of the parties – making the VCLT mechanism unnecessary for general amendments. However, Article 42 ECT does not allow for modifications between a limited subset of parties, leaving only Article 41 VCLT as a fall-back option.

Article 41(1)(b) VCLT allows modifications between certain parties if there is no treaty bar and if the following conditions are cumulatively satisfied; (i) it does not affect the rights or obligations of other parties and (ii) does not impede the treaty’s core objectives. A mere plurilateral treaty cannot salvage the situation if the twin conditions remain unfulfilled.

Regarding (i), the Declaration restricts itself to intra-EU disputes, claiming no impact on non-EU parties’ rights or obligations. However, selective application risks fragmenting ECT jurisprudence, creating different standards for EU and non-EU parties.

The terrain becomes murkiest when turning to (ii). The Komstroy decision effectively dismantles Article 26’s dispute resolution mechanism, undermining the ECT’s intent to promote energy cooperation without geographic limits. Dispute resolution provisions are essential to the treaty’s purpose (p. 75). The UK Supreme Court in Infra Luxembourg v Spain (para 87) questioned the CJEU’s role in overriding established treaty obligations. In such circumstances, a CJEU decision that interferes with the existence of ECT’s dispute resolution mechanism negatively impacts ECT’s objects and purposes.

Denouement: Modification Through Subsequent Practice 

The legitimacy of modifying treaties through subsequent practice is itself debatable. The VCLT does not explicitly address subsequent practice as a basis for modification; instead, Articles 39-41 focus on subsequent agreements. The boundary between “practice” and “agreement” is fluid, a matter beyond the scope of this blog post. Initially, the ILC’s 1966 Draft Articles included Article 38 to permit treaty modification through consistent practice reflecting party consent. However, this draft article faced numerous objections and was eventually deleted, which some interpret as rejection, though others argue it remains valid under customary international law.

In any event, just as Article 41 VCLT prohibits modifications that undermine a treaty’s core purpose, modifications via subsequent practice must also avoid affecting the treaty’s object and purpose. As previously discussed, Komstroy complicates the ECT’s dispute resolution, an essential element of the treaty’s purpose.

Additionally, it remains uncertain if EU states have established a consistent practice sufficient for modification. “Practice” can encompass a wide range of actions, including legal arguments in disputes, as noted in the Certain Expenses Advisory Opinion (p. 157). Aspects such as the manifestation of a state’s intention, concreteness of the practice, etc. also factor in. Post Achmea and Komstroy, many EU nations have prayed for the awards’ invalidity before the enforcement courts, including in the UK and the US. The European Commission tried to intervene as amicus before domestic courts and arbitral tribunals. In the ECT’s context, the arguments’ legal content and coherence as well as the number of states that raised them will have to be considered. Political expediency behind the actions of states can also militate against the finding of a good faith intention to bring modification to treaties (see Dissenting Opinion of President Winiarski in Certain Expenses, p. 232).

This leads to a challenging query with which I end this piece: How would one characterize the actions of EU member states? As a good faith and legally sound attempt at interpreting or modifying the treaty, or as a political maneuver that runs afoul of international law norms?