You are here

Inside Arbitration: Party autonomy prevails Revisiting London after the latest Supreme Court decisions

25 February 2021 | Global
Legal Briefings


Against the backdrop of Brexit, there has been an increased interest in London as an arbitral seat, intensifying with the UK’s exit from the EU in January 2020 and over the few months leading up to the end of the transitional arrangements on 31 December 2020.  This greater  focus on London has coincided  with several high profile Supreme Court cases which have addressed key aspects of arbitration law. In this regard, Q4 of 2020 saw the most significant English arbitration decisions since 2011 and the judgments in question have placed both arbitrator impartiality and the governing law of the arbitration agreement under the lens. In this article, we consider the recent Supreme Court decisions and where these leave London as a seat of arbitration.

What happened in Halliburton?1

The underlying arbitration between Halliburton and Chubb related to the Deepwater Horizon incident and Halliburton’s claim under an insurance policy. As the presiding arbitrator could not be agreed, the English High Court appointed Kenneth Rokison QC in June 2015. Mr Rokison  was subsequently appointed by Chubb in a further arbitration in relation to Transocean’s separate claim against Chubb under the same insurance cover (December 2015). Transocean then appointed him in relation to Transocean’s claim against a different insurer under the same layer of insurance (August 2016). Halliburton became aware of these arbitrations in November 2016 and then applied to the English Court for Mr Rokison’s removal under s24 of the English Arbitration Act 1996. Both this application and a subsequent appeal to the Court of Appeal, were rejected.

Halliburton appealed to the Supreme Court, arguing that there was apparent bias on Mr Rokison’s part. This was because Chubb could communicate with the arbitrator by way of submissions and evidence submitted in the later proceedings on questions  potentially relevant to the arbitration between Halliburton and Chubb. Halliburton contended that this gave Chubb an unfair advantage. Halliburton also took the position that Mr Rokison’s non-disclosure amounted to apparent bias.

As the case raised questions of general importance in the law of international arbitration, the Supreme Court permitted interventions from multiple arbitral institutions and organisations, notably the LCIA, ICC, CIArb, LMAA and GAFTA.

The Supreme Court was asked to decide two issues:

  1. Whether overlapping appointments could be accepted without this resulting in the appearance of bias on the part of the arbitrator?
  2. Could such appointments be accepted without giving disclosure?

Supreme Court decision

The Supreme Court made it clear that impartiality was vital in international arbitration. There was no suggestion of actual bias, so the Supreme Court was required to decide whether there was an appearance of bias, applying the well-established test of “whether the fair-minded and informed observer, having considered the facts, would conclude that there was a real possibility that the tribunal was biased.” The Supreme Court decided that when the apparent bias test is applied to arbitrators, the unique features of arbitration must be considered, including the private nature of arbitration, the limited rights of appeal, the appointment process and the potential “financial interest in obtaining further appointments as arbitrator”. The Court emphasised that in English law, party-appointed and presiding arbitrators must meet exactly the same high standards of fairness and impartiality.

The decision confirmed that all arbitrators have a duty of disclosure concerning facts and circumstances which would or might reasonably give rise to the appearance of bias. In considering compliance with this duty, the Court will assess the circumstances at the time the disclosure fell to be made. The Court emphasised that the duty aligned with best practice set out in the International Bar Association ("IBA") Guidelines on Conflicts of Interest in International Arbitration 2014 and the approach of arbitration institutions such as the LCIA and ICC.

The Supreme Court determined that acceptance of multiple overlapping appointments of this kind may have to be disclosed, in accordance with usual practice in the relevant type of arbitration. Failure to disclose appointment in overlapping arbitrations may amount to apparent bias and will be one factor taken into account in deciding whether there was a real possibility of bias. Apparent bias is assessed in light of the circumstances at the time that the challenge to the arbitrator is heard.

The Supreme Court decided that Mr Rokison had been under a legal duty to disclose the multiple appointments, because disclosure of this kind was the norm in Bermuda Form arbitration and because at the time of that appointment it “might reasonably give rise to the real possibility of bias”.2

However, the Supreme Court found that there was no appearance of bias in this case, for the following five reasons:

  1. At the relevant time the English law on arbitrator disclosure was unclear.
  2. The time sequence of the three arbitrations in question may explain why Mr Rokison disclosed the arbitration between Halliburton and Chubb to Transocean, but did not identify the need to tell Halliburton about the arbitration between Chubb and Transocean.
  3. Both the later arbitrations were to be decided by way of preliminary issue, so there would be no overlap in evidence or submissions. Mr Rokison had offered to resign from the later arbitrations should this prove not to be the case.
  4. There was no secret financial benefit.
  5. Mr Rokison had dealt with the challenge fairly and there was no evidence that he “bore any animus towards Halliburton as a result”.

What was decided in Enka v Chubb?3

The English Supreme Court case was an appeal from the decision of the Court of Appeal, which had granted an anti-suit injunction preventing Chubb Russia4 from continuing a subrogation claim in the  Russian courts. This was on the basis that the Russian claim was brought in breach of the arbitration agreement in question, which the Court of Appeal decided was governed by English law.

In the Supreme Court appeal, Chubb Russia took the position that the arbitration agreement constituted an integral part of the contract and was therefore governed by the law governing the main contract. Chubb Russia argued that the parties had made an implied choice of Russian governing law for the main contract.

Supreme Court decision

The majority of the Supreme Court decided that an arbitration agreement will usually be governed by the same law that has been chosen to govern the main contract and this choice of law will not be displaced by the selection of a different seat of arbitration. Additional factors may be relevant in some circumstances, including where there is a serious risk that the arbitration agreement would be rendered ineffective if it were to be governed by the main contractual governing law.

The majority decided that there had been no express or implied choice of Russian governing law for the main contract. Under Article 4(3) of the Rome I Regulation, the contract was governed by Russian law, on the basis that this was the country with which the contract was most closely connected. This meant that it was necessary to consider the system of law with which the arbitration agreement had its closest and most real connection, which was determined to be English law, being the law of the seat.

In relation to the anti-suit injunction, the Supreme Court decided that it was irrelevant whether an arbitration agreement was governed by English or foreign law, confirming the availability of anti-suit relief.

Both these judgments have generated a significant volume of comment in the international arbitration community. There has been a particularly mixed reaction internationally to the Halliburton decision, with a number of criticisms made of the judgment, including whether the application of the principles outlined in the judgment to the facts in question met the expectations of the international arbitration community. There have been criticisms of the long-standing objective nature of the apparent bias test in English law and some commentators have also raised concerns around the differing timings set by the Supreme Court for assessing compliance with arbitrator duties of disclosure and impartiality.

In relation to the Enka v Chubb case, there are certainly differing approaches internationally to the question of determining governing law of an arbitration clause. The French courts, for example, will review the scope of an arbitration clause in a Paris-seated arbitration with regard to mandatory French law/public policy, the parties’ intentions and the case background, but without reference to any domestic law. These different approaches to governing law can put national courts on something of a collision course, as in the recent Kabab-Ji SAL saga. In that case, the Paris Court of Appeal refused to set aside an arbitral award handed down by an ICC Tribunal seated in Paris, despite the same award having been denied enforcement by the English Court of Appeal, on the basis that English law governed the arbitration agreement and that the award was made against a non-party.5

While there are clearly a range of standpoints on both decisions, these definitive judgments have brought a welcome clarity to two important areas of English arbitration law, with the courts taking a pro-arbitration stance and reinforcing the centrality of party autonomy. After these decisions there is greater certainty on how governing law of an arbitration agreement will be determined by the English courts and how the courts will assess arbitrator conflicts. While these decisions have not won entirely unanimous support, importantly they are both pro-arbitration, in that Enka v Chubb has confirmed the ability to use anti-suit injunctions in support of arbitration and the Halliburton decision has underlined that the English courts will judge any allegations of apparent bias carefully in the context of the specific circumstances of each particular case.

Party autonomy could be said to lie at the heart of both these decisions, with the approach to governing law in Enka v Chubb leaving the parties free to expressly state their preferred governing law for the arbitration agreement itself. In contracts where no express choice of governing law has been referred to in the arbitration clause, the English courts will recognise the parties’ choice of governing law for the underlying contract as a clear choice of governing law for the arbitration clause.

The Halliburton decision leaves parties free to ensure what the Halliburton judgment referred to as a “best practice” approach, by opting for institutional arbitration, such as LCIA or ICC arbitration, where decisions on conflicts are generally informed by the IBA Guidelines. Parties also continue to have the choice to opt for the different approach to conflicts taken in specialist types of arbitration, including for example, LMAA and GAFTA maritime and commodities arbitration, which is expressly recognised by the IBA Guidelines.6

Where does this leave London?

The finding in Halliburton of no apparent bias may have been criticised by some, but that decision was founded on the narrow facts of the particular ad hoc arbitration in issue and the reasoning is pro-arbitration. Further, selecting London-seated institutional arbitration provides a measure of certainty that conflicts will be dealt with in the manner expected by the international arbitration community. This is because challenge applications are first made to the relevant institution, so that challenges are considered by specialists in international arbitration, who are well aware of the standards of arbitrator ethics expected internationally. In relation to governing law, some parties from jurisdictions with a different approach to interpreting arbitration agreements might not be inclined to agree with the English law principles set out in Enka v Chubb. That said, the decision ensures certainty and this issue can in any event effectively be side-stepped by expressly referring to the chosen governing law in the arbitration clause itself.7

London is not only a leading institutional seat, but also continues to be a popular seat for ad hoc arbitration, with over 1,000 London-seated ad hoc arbitrations commenced in 2019.8 Some users of ad hoc arbitration may well move to institutional arbitration in the wake of the Halliburton case, in recognition of the strong institutional approach to conflicts. However, in a number of the sectors prominent in London-seated ad hoc arbitration, such as maritime and commodities, there is a small pool of specialised arbitrators and the relevant market accepts that multiple and overlapping appointments are an inevitable feature. Given the lack of consolidation provisions in these types of arbitration, it is accepted that parties will often wish to appoint the same tribunal and deal with the arbitrations concurrently.

Ultimately, the key factors behind London’s leading position as a seat for both institutional and ad hoc arbitration remain unaltered by the latest judgments and parties are likely to continue to be attracted by the arbitration-friendly legislative framework and supportive courts, along with the existence of a thriving arbitration hub.


  1. Halliburton Company (Appellant) v Chubb Bermuda Insurance Ltd [2020] UKSC 48.
  2. Emphasis is from the original judgment.
  3. Enka Insaat Ve Sanayi AS v OOO Insurance Company Chubb [2020] UKSC 38.
  4. While Chubb entities were parties to both the Halliburton case and the Enka v Chubb case, the two cases were not related.
  5. See the French proceedings in Kabab-Ji SAL (Lebanon) v Kout Food Group (Kuwait) [CA Paris, 23 June 2020, n°17/22943] and the English Court of Appeal judgment, Kabab-Ji SAL (Lebanon) v Kout Food Group (Kuwait) [2020] EWCA Civ 6.
  6. See Footnote 5 to 3.1.3 IBA Guidelines on Conflicts of Interest in International Arbitration 2014, which recognises that no disclosure of multiple appointments is required in types of arbitration such as maritime, sports and commodities, with a smaller/specialised pool of arbitrators, where parties frequently appoint the same arbitrator to different cases and where all parties should be familiar with this.
  7. For users of LCIA arbitration, the issue is dealt with by Article 16.4 of the LCIA Rules, which provides that the law governing the arbitration agreement will be the law of the seat, unless the parties have agreed otherwise.
  8. For the latest published statistics on ad hoc arbitration, see our blog post: Ad hoc arbitration alive and well in London: the latest statistics.


More Inside Arbitration

See how we help our clients in

International Arbitration

Learn More