Micula: ICSID enforcement before the English Court of Appeal
28 August 2018
On 27 July 2018, the Court of Appeal delivered its judgment in Viorel Micula and others v Romania and European Commission (Intervener). It maintained the stay of enforcement granted by Blair J in the High Court, but overturned his decision on security, ordering that Romania should provide £150 million as a term of the stay (but not making compliance with the order on security a condition of the stay). The decision reveals a delicate balancing act by the court between the UK’s obligations under the ICSID Convention and its duties under EU law.
In December 2013, the Micula brothers obtained an ICSID award against Romania (award), ordering it to pay compensation breaches of the Sweden-Romania bilateral investment treaty (BIT). In March 2015, the European Commission (Commission) issued a decision (decision) concluding that enforcement of the award would constitute new state aid under Article 107(1) of the Treaty on the Functioning of the European Union (TFEU) and was, therefore, prohibited. The award had already been registered in the UK in October 2014 (registration order).
In July 2015, Romania filed an application in the Commercial Court to set aside, vary or stay the registration order. The Miculas later commenced proceedings before the General Court of the European Union (GCEU) seeking an annulment of the decision. In September 2016, they filed a cross-application for security for damages, pending the outcome of the GCEU proceedings. In January 2017, Blair J refused to set aside the registration order, but granted a stay of enforcement pending determination of the GCEU proceedings. In June 2017, the Miculas’ application for security was refused. They appealed against the grant of a stay and the refusal of security. The Commission appeared as an intervening party, supporting Romania’s position.
The Court of Appeal rejected the stay appeal and granted the security appeal. The decision carefully attempts to navigate between the UK’s obligation to enforce arbitral awards under the ICSID Convention incorporated into English law through the Arbitration (International Investment Disputes) Act in 1966, and the UK’s EU law obligations. While the judgment was unanimous, the reasoning of the judges on the stay appeal differed, reflecting the difficulties in treading a line between these two sets of obligations. This post considers the tension between the judges’ reasoning, with reference to two of the principal discussions in the court’s judgment concerning international law.
Section 2 of the Arbitration (International Investment Disputes) Act 1966
The first concerned whether an English court was obliged to stay enforcement of an ICSID award simply because it would be contrary to EU law. Central to this issue was section 2(1) of the 1966 Act, which provides that a registered award shall be of the “same force and effect for the purposes of execution as if it had been a judgment of the High Court”. Romania argued that:
- The effect of section 2(1) was to give the award on registration the same status in English law as any other judgment.
- The enforcement of any other judgment would be subject to any restrictions on enforcement under EU law.
- The same principle should apply to the enforcement of an ICSID award (referred to as Romania’s “equivalence argument”; that an ICSID award should be treated as equivalent to an English judgment for enforcement purposes).
Hamblen LJ broadly accepted Romania’s equivalence argument, recognising that, in some respects, the restrictions on enforcement of ICSID awards are unusually narrow (for example, there is no public policy exception to enforcement). Nevertheless, he held that “neither the ICSID Convention nor the 1966 Act confers a status on a registered award for the purposes of enforcement which is different or superior to that of a final judgment”. Where there was a reason that a final domestic judgment would not be enforced, that should equally apply to the award.
Arden and Leggatt LJJ took a slightly different approach, holding that section 2(1) could not be interpreted out of context and solely as a domestic statute. Rather, by enacting section 2, Parliament had intended to perform its international treaty obligations. Under the ICSID Convention, the UK had an obligation to enforce ICSID awards (“automatic[ally]”, said Leggatt LJ). The 1966 Act could not logically be understood as overriding the UK’s enforcement obligation under the ICSID Convention by requiring the court to decline to enforce a judgment contrary to EU law.
Leggatt LJ further stated that section 2(1) should be interpreted as limiting the analogy with a High Court judgment to defining the “force and effect of a registered award” for the purpose of execution. It did not give the High Court power to refuse to enforce for a reason that would justify staying enforcement of an ordinary domestic judgment.
These divergent approaches highlight the difficult balance between the UK’s enforcement obligations under the ICSID Convention and potentially conflicting EU obligations. Arden and Leggatt LJJ’s narrower approach to the permissibility of granting a stay is surely to be preferred; Hamblen LJ’s alternative option seems to derogate from a fundamental purpose of the ICSID Convention: effective enforcement. When section 2(1) was enacted to give effect to that purpose, it cannot have been intended by Parliament to have the effect that registered ICSID awards should be brought within the scope of a later international treaty (that is, EU treaties), which did not expressly affect the UK’s ICSID obligations.
Arden LJ explained that granting the stay was within the powers conferred on a domestic court by the ICSID Convention, provided the stay was consistent with the ICSID Convention’s purpose (by which the court presumably meant the automatic enforcement of awards). It would therefore not be open to impose more than a temporary stay on execution if an award was enforceable. Thus, there could only be a limited discretion under Article 54 of the ICSID Convention, otherwise the courts could use powers granted by domestic law to effectively thwart execution. Here, a stay was justified pending the outcome of the GCEU proceedings because (in Leggatt LJ’s reasoning) it would then be apparent whether there was a conflict between the UK’s obligations requiring resolution.
Article 351 of the Treaty on the Functioning of the European Union FEU
The second discussion pertains to Article 351 TFEU, which provides that:
“[t]he rights and obligations arising from agreements concluded before 1 January 1958 or, for acceding States, before the date of their accession, between one or more Member States on the one hand, and one or more third countries on the other, shall not be affected by the provisions of the Treaties”.
If it were held that there was a conflict between the UK’s obligations under the Convention and its EU law duties, Article 351 would resolve which should take precedence. Thus, if the English court were not to stay enforcement of the Award, and if it considered there was a conflict, it would have to consider Article 351.
The issue before the court was not the substantive question of how Article 351 should be interpreted, but rather a procedural one. Hamblen and Leggatt LJJ held that a further reason for granting a stay was that the GCEU was considering Article 351 TFEU; there was a risk, therefore, of conflicting views being reached.
In Arden LJ’s opinion, however, there was little overlap, as the GCEU proceedings were focused on the effect of the Sweden-Romania BIT, while the UK proceedings concerned the UK’s obligations under the ICSID Convention. In her view, a stay would not, therefore, have been justified on this ground.
Arden LJ also considered other issues about Article 351’s interpretation and application, particularly Romania’s argument that EU law applied to the issues before the English court pursuant to Article 351 because the relevant ICSID Convention obligations should be treated as being solely between member states. This approach, she believed, ignored the ICSID Convention’s significance as a multilateral treaty. The UK’s obligation to enforce ICSID awards in its courts may (subject to the EU treaties) have been owed to Sweden, as the Miculas were its nationals; but the UK also owed its ICSID Convention obligations to other ICSID Convention parties who were not parties to the Sweden-Romania BIT. For example, it was possible that any ICSID contracting party could commence proceedings against the UK as to whether the UK was complying with its enforcement obligation under Article 54.
The court granted security as a term of the stay on the basis that non-compliance would not itself lead to the termination of the stay. It rejected Romania’s and the Commission’s argument that the decision prohibited payment of the award and provision of security. Security instead simply represented compliance with a court order. The purpose of such an order was apparently to preserve (insofar as EU law permitted) the Miculas’ position under international law. If it was not possible currently to comply with enforcement, it “should in principle do the ‘next best’ [thing]”. This would at least assist the Miculas if the GCEU rules in their favour.
What happens next?
The court’s decision in Micula is both principled and pragmatic, seeking to reach a decision that does not place the UK’s international law obligations under the ICSID Convention in direct conflict with its EU duties. The court noted that, if the decision is set aside and there is no appeal, it is difficult to see how the stay could be extended.
Both Arden and Leggatt LJJ emphasised that any stay could only be for a temporary purpose. As such, it may be necessary for the court hearing any further stay application to decide upon questions so far left unanswered. In particular, these questions may include whether enforcement under the ICSID Convention is a pre-accession treaty obligation of the UK, and therefore not displaced by entry into the EU Treaties (including, self-evidently, state aid obligations).
The Micula saga continues.