The enforcement of intra-EU ICSID awards: shortcomings of the British approach

Luís Heleno Terrinha

Of Counsel, PLMJ Lawyers[1]

Assistant Professor, NOVA School of Law


This post discusses the reasoning behind the judgment of the High Court of Justice, dated 24 May 2023, on the dispute between Infrastructure Services Luxembourg S.À.R.L. and Energia Termosolar B.V. v. Kingdom of Spain (available here and discussed here). A later post is intended to address the judgment of the Bundesgerichtshof, dated 27 July 2023, in the context of an anti-arbitration injunction requested by Federal Republic of Germany (an outlook of the issues of the case can be read here). A contrast between both decisions seems useful as far as neither decision seems to have managed to correctly apply the law (either domestic or international) to all pending issues. In sum, while the British court engaged in a superficial application of article 351 TFEU and provided unsettled grounds as to the erga omnes character of the ICSID Convention, the German court failed to convince as to the applicability of the anti-arbitration injunction mechanism (as per 1032 Abs. 2 via 1025 Abs. 2 ZPO; here) and as to the sacrifice of the competence-competence of ICSID arbitrators according to article 41 ICSID Convention (here).

  1. The problem: EU Law or UK’s ICSID obligations?

On 24 May 2023 the High Court of Justice handed down a judgment in which it ultimately determined the recognition and enforcement in the UK of an ICSID award obtained by Luxembourgish investors against Spain. The background of the case, as well as its overall reasoning, have been previously described here in a detailed post by Laura Rees-Evans. For that reason, I will only focus on the shortcomings of the legal arguments invoked by the High Court to make the ICSID Convention prevail over EU law. In fact, it is well understood that the issue at hand was “whether EU law, as set out in Achmea and Komstroy, could trump the UK’s existing treaty obligations under the ICSID Convention”. Expectedly it turned to the precedent set by the UK Supreme Court in the Micula case.

  1. The UKSC precedent: reasoning

That precedent revolves around two core tenets.

The first tenet is a temporal one and draws on the application of article 351 TFEU. The UK acceded first to the ICSID Convention and only afterwards to the EU (coincidentally, the same was true for Spain, which became a party to ICSID before entering the EU; EWHC §86). In face of such a chronology, the UKSC found that article 351 TFEU “means that the obligations on the United Kingdom arising from the ICSID Convention are ‘not … affected by the provisions of the Treaties’” (UKSC §85, EWHC §74). This shows that, from the UK point of view, the issue was one of obligations arising out of pre-acession treaties (EWHC §80). That explains the British judges’ wariness in giving “preference to granting the decisions of the CJEU complete primacy over those pre-existing treaty obligations of all states” (EWHC §87, 123). Curiously enough, the UKSC notes the contrasting situation in which Belgium finds itself, since the latter was an original signatory of the Treaty of Rome (UKSC §115).

The second tenet is internationalist and relates with the multilateral character of the ICSID Convention. This was only fleetingly touched upon by the High Court, when it underlined that the UK’s treaty obligations under the ICSID Convention are owed to “all signatories” of the latter (EWHC §80). The High Court needed not to go much further since the UKSC had already teased out this point in full when it addressed the “scope of obligations under pre-accession agreements” (UKSC §98 ff.). In essence, it concluded then that “all States which are parties to the ICSID Convention have an interest in the effective operation of the Convention scheme for the enforcement of arbitral wards (UKSC §101). This meant interpreting the UK’s specific enforcement obligations under the ICSID Convention (in particular articles 54 and 69) as having a collective, erga omnes partes or at least interdependent nature. Indeed, for the UKSC “it is clear” that “[t]he Convention scheme is one of mutual trust and confidence which depends on the participation and compliance of every Contracting State” (UKSC §104) (for a similar view, from an academic standpoint, see the post by Christian Tietje). This view thus posits that enforcement obligations are owed not only to the parties of the award, but to “all other States party to the Convention”.

  1. Shortcomings of the British approach

Each of theses tenets faces its own set of problems.

At the outset, the first tenet was used by the UKSC in a rather truncated fashion. In fact, the UKSC read from the TFUE what it wished to read, essentially picking and choosing from article 351. The UKSC focused exclusively on paragraph 1 of that provision (the conflict rule) but avoided any mention of paragraph 2 (the duty to eliminate the incompatibilities between a pre-accession agreement and the European Treaties). What this makes clear is that, under European Union law, the UKSC cannot – as it did – unrestrictedly invoke the exception to the supremacy of EU Treaties (pursuant to article 351(1) TFEU) without dealing with the fact that such exception ought to be temporary and is coextensive with an autonomous obligation impending upon the Member State to resolve any potential normative contradiction (pursuant to article 351(2) TFEU). Therefore, it would have been interesting – at the time (of the post-Brexit transition) – to watch the UKSC deal with the implications of such a duty of loyal cooperation and distill the latter’s consequences concerning the (non-)enforcement of intra-EU ICSID awards. The point would consist in determining whether the obligation to remove incompatibilities between pre-accession agreements and EU Treaties was in itself a strong enough argument to impose the stay of enforcement of an intra-EU ICSID award. Although now irrelevant within the UK, this issue might still be relevant for all Member States which have signed the ICSID Convention before acceding to the EU (even though the CJEU’s Romatsa ruling has clarified that such awards are not to be enforced by Member States).

The second tenet, although sufficiently developed in the reasoning of the UKSC (§105-108), is not intuitive. At this stage, it would take us (too) far to address the collective character of the ICSID’s convention obligations or lack thereof. But, leaning on a prior writing by Joost Pauwelyn, some remarks are in order: (i) the ICSID Convention does not expressly qualify the obligations under it as having collective or even interdependent character (contrary to the EU Treaties, which do that precisely in article 351(3) TFEU); (ii) notwithstanding the importance and relevance of the ICSID Convention, it is more common to think of collective or interdependent obligations as those relating to environment, security, disarmament, world heritage or human rights concerns; (iii) it is up for discussion whether a third state has a specific interest in the enforcement of an intra-EU ICSID award for as long as the awards concerning one of its nationals and another contracting state (i.e., extra-EU) are indeed enforced. Moreover, the dispute underlying any award is bilateral (a private party against a State) and so is, in most instances, the very international legal instrument which enables investment arbitration in the first place (such as a BIT). For example, Joost Pauwelyn has considered that WTO obligations have bilateral nature with arguments that, mutatis mutandis, could be easily transposed to the investment law realm.

All in all, it is plausible to think that other courts might find the obligations under the ICSID Convention to have a merely bilateral character, which would then prejudice the applicability of article 351(1) as so far interpreted by the CJEU.

  1. Contingent value of the High Court of Justice decision

In any case, with the UK out of the European Union, the High Court judgement has its own value not because of what it says about the relationship between Member States’ domestic law and European Union, but rather because it states the UK (favourable) stand as to the enforcement of intra-EU ICSID awards. Certainly, the UK courts are entitled to consider that European Union law does not trump the UK’s enforcement obligations under the ICSID Convention scheme, and much less so when the UK is no longer part of the Union. They are also entitled to find that enforcement is still owed when the underlying dispute is purely intra-EU. If that, however, is an answer that could be similarly provided by the court of a Member State is an altogether different issue – which we shall tackle when discussing, in a next post, the anti-arbitration injunction issued by the Bundesgerichtshof in its decision of 27 July 2023.

[1] The views expressed in this blog post are those of the author solely and do not represent the views of the firm nor do they bind it in any way.