Mustafaen Kamal1
This article serves as a follow up to my February 19 2026 Article which commented on the Court of Appeal’s judgment in the case which has since been considered and decided by the Supreme Court of the United Kingdom, in substance endorsing the approach taken below.
The question that framed my earlier article — whether state immunity should be able to prevent the recognition of an ICSID award — has now been answered by the highest court in the United Kingdom. The Supreme Court’s judgment in the conjoined appeals brought by Spain and Zimbabwe, handed down in Infrastructure Services Luxembourg v Spain and Border Timbers v Zimbabwe [2026] UKSC 9, confirms with welcome clarity that once a state joins the ICSID Convention, it accepts that national courts of other contracting states may recognise and enforce awards rendered against it. The judgment resolves a debate that had begun to unsettle the coherence of the enforcement regime and addresses the very concerns that animated the case.
When I first wrote about the Court of Appeal’s decision, I described the ICSID Convention as a bargain where nations willingly forego state immunity in the name of neutrality and enforceability. Investors agree to take their disputes out of domestic courts and into an international forum and states agree to honour the resulting awards as if they were final judgments of their own courts. The upside for states is the prospect of increased investment by commercial entities who feel safer at the prospect of a neutral dispute resolution forum, in case things go wrong. Since the inception of the ICSID convention, this understanding held relatively firm, not because it was untested but because compliance was common and the scale of awards relatively modest. It was only when larger, politically charged claims began to increase that states turned more readily to immunity in an effort to resist enforcement. The tension that had always been latent in the system, between sovereignty on the one hand and treaty-based consent on the other, moved centre stage.
The Supreme Court’s judgment settles the matter in a way that restores the conceptual symmetry of the ICSID regime. The Court held that Article 54 of the Convention, properly understood, necessarily requires contracting states to accept the jurisdiction of national courts for the purposes of recognising and enforcing ICSID awards. If states are under an obligation to treat such awards as if they were final judgments of their own courts, then they must also accept that those awards will be recognised as such in the courts of other contracting states. Recognition cannot occur in the abstract. It is a judicial act performed by a national court. In England, recognition takes the form of registration as a judgment of the High Court under section 101 of the Arbitration Act 1996, which gives the award creditor access to the full range of ordinary enforcement mechanisms available against a judgment debtor, though it does not, of itself, override execution immunity in respect of state assets. A treaty obligation to recognise an award is therefore also, by implication, an acceptance that courts will exercise the jurisdiction required to give effect to that obligation.
This conclusion speaks directly to the central question of my original article: should immunity be allowed to defeat recognition? The Supreme Court’s answer is that immunity cannot be invoked because it has already been waived. The waiver is not implicit in the sense of being inferred from silence; it constitutes, in the Court’s own terms, a clear and unequivocal submission to the adjudicative jurisdiction of domestic courts — one that is inherent in and integral to the treaty obligation undertaken. By acceding to the Convention, states commit themselves to a mechanism that operates through national courts. To accept the benefits of the ICSID system without accepting its enforcement architecture would be to unpick the very fabric of the Convention.
Equally important is the way the Court reached this conclusion, which is broadly consistent with the reasoning of the Court of Appeal below. Both courts. Rather than confining themselves to a narrow parsing of statutory language or isolating the question within the confines of the UK State Immunity Act 1978, the Supreme Court affirmed the appellate court’s purposive reading of the ICSID Convention. It emphasised that the Convention was designed to depoliticise investment disputes by channelling them away from domestic litigation and by insulating awards from national judicial review. Recognition was meant to be automatic and uncontroversial. The Convention’s structure reflects this. Articles 53 to 55 form a carefully calibrated sequence: Article 53 obliges parties to comply with awards; Article 54 requires contracting states to recognise and enforce them as if they were final domestic judgments; and Article 55 expressly preserves execution immunity. Together, they constitute what the Court described as a self-contained scheme, within which there is no role for domestic courts to scrutinise the merits of an ICSID award, and no basis for applying local public-policy exceptions. The only sphere left untouched by the Convention is execution, where states retain immunity from measures directed at their property except to the extent that they have separately waived that protection.
That structure would be distorted if immunity could be invoked at the threshold stage of recognition. Without recognition, an award has no juridical foothold in domestic law; it cannot be converted into a judgment, cannot be used as leverage, and cannot serve as the basis for negotiations. The Supreme Court understood that allowing immunity to prevail at that stage would effectively collapse the Convention’s enforcement mechanism. The Court’s commitment to interpreting the Convention in a way that preserves its internal logic mirrors the concerns outlined in my earlier writing: that failure to do so would re-politicise disputes and weaken investor confidence.
The Supreme Court also placed considerable weight on consistency with the broader international jurisprudence. One of the anxieties I noted previously was the risk of fragmentation, of a world in which some courts accepted the ICSID Convention’s obligations while others allowed immunity to block recognition. Such division would invite forum shopping, encourage states to seek out sympathetic jurisdictions, and undermine the predictability that is essential to cross-border investment. The Supreme Court acknowledged this risk and noted that courts in Australia, New Zealand, Malaysia, the United States, and several European jurisdictions have taken a broadly similar approach: accession to ICSID entails acceptance of recognition proceedings in other contracting states. By affirming that understanding, the UK ensures that it does not become an outlier and that the Convention is interpreted uniformly, as its framers intended.
Importantly, the judgment maintains the careful distinction between “adjudicative immunity” and “immunity from execution.” While immunity cannot prevent the registration of an award, it still protects states from the seizure of non-commercial assets, and from other intrusive measures directed at sovereign property. This is entirely consistent with the Convention’s design; indeed, the drafters expressly left immunity from execution untouched, recognising the sensitivity that attaches to state assets, particularly those used for diplomatic or governmental functions. The Supreme Court’s ruling therefore rebalances, rather than redefines, the relationship between immunity and treaty-based consent. Recognition flows from the Convention; execution remains subject to domestic law and the narrower waivers states may choose to give.
That balance answers another of the questions I raised in the earlier article: whether overcoming immunity at the recognition stage would amount to an undue intrusion on sovereignty. The Supreme Court’s approach demonstrates that it does not. Recognition does not compel payment. It does not force a state to surrender property. What it does is ensure that the award is given legal effect within the jurisdiction, reflecting the state’s own commitment when it joined ICSID. From there, the path to enforcement remains constrained by the long-standing protections that international law affords.
In many ways, the Supreme Court’s judgment is a vindication of the ICSID system itself. It reaffirms the core proposition that states are bound by the commitments they choose to make, particularly in the context of a treaty designed to remove political considerations from the resolution of investment disputes. It also reinforces the rule of law in the investment sphere: states cannot accept the benefits of investor confidence and treaty protection while resisting the corresponding obligations they freely undertook.
Looking back to the questions that framed the earlier article, the answers now appear clear. Should immunity trump recognition? No. Does ICSID require domestic courts to exercise jurisdiction over recognition proceedings? Yes. Is the UK aligned with the global trend? It is, and decisively so.
- Mustafaen Kamal, LLM (2022, Harvard) is an Associate in the International Arbitration Group at Baker McKenzie in London.
Baker McKenzie represented Border Timbers Limited in the aforementioned enforcement proceedings in the UK. The views expressed in the article are the author’s own, and should not be attributed to Baker McKenzie. ↩︎