In November this year, the UK Supreme Court heard the appeal in the Halliburton v Chubb, which has been the subject of much scrutiny by the arbitration community recently.
Background
The proceedings related to the claim by Halliburton Company against Chubb Bermuda Insurance Ltd, seeking indemnification under an insurance policy for its liability for 3% of the damage caused by the Deepwater Horizon oil rig disaster in the Gulf of Mexico in 2010. The underlying claim was referred to arbitration in June 2015.
In the arbitration, the parties were unable to agree on a chair for the arbitral tribunal so the English Commercial Court, acting as an appointing authority pursuant to section 18(3)(d) of the English Arbitration Action 1996 (AA 1996), appointed an arbitrator, “M” (who has not been named in the court proceedings), to chair the proceedings, notwithstanding objections from Halliburton. M was Chubb’s preferred appointee and had disclosed, prior to his appointment, that he had previously acted and was currently acting in a number of arbitrations which involved Chubb, including some matters where he was Chubb’s appointee.
However, M subsequently failed to make a further disclosure following his appointment, to the effect that he had also accepted a nomination by Chubb, as party-appointed arbitrator, in two other insurance-related arbitrations connected to the Deepwater Horizon disaster albeit with a different claimant (Transocean). When Halliburton learned of these subsequent appointments some 18 months into the proceedings, it questioned M’s impartiality and requested that he resign on the basis that M was faced with similar issues in all three Deepwater Horizon cases, including the arguments raised by Chubb in its defence. M said that he had not considered the need to disclose these appointments but offered to resign, albeit only with Chubb’s consent.
Chubb did not consent to M’s resignation. Halliburton therefore sought to have M removed under section 24(1)(a) of the AA 1996, which allows for removal if “circumstances exist that give rise to justifiable doubts as to his impartiality.” The premise of Halliburton’s application was:
- M’s acceptance of the further appointments.
- His failure to disclose them.
- That his response to the challenge to his impartiality resulted in an appearance of bias.
The road to the Supreme Court
At first instance, Popplewell J held that there was nothing which gave rise to an appearance of bias. In particular, he highlighted the need for party autonomy when appointing arbitrators. He also considered that an experienced arbitrator would be able to ignore facts and arguments which may only have been presented elsewhere when deciding the present case. Halliburton therefore appealed to the Court of Appeal.
The Court of Appeal upheld Popplewell J’s decision. The court considered that, although M should have disclosed the subsequent appointments, the acceptance of the subsequent appointments of itself was not sufficient to create an impression of bias; to do so, “something [more] of substance” was required. The court deemed the non-disclosure to be inadvertent and found that there was only a limited degree of overlap between the arbitrations in which M was appointed. This decision was then also appealed to the Supreme Court.
Arguments in the Supreme Court
The core issues before the Supreme Court were:
- When should an arbitrator disclose circumstances which may give rise to justifiable doubts as to his impartiality?
- What are the consequences of a failure to do so?
The nature of the dispute also led to interventions and submissions from several arbitral institutions.
Halliburton argued that English law should apply a “gold standard” when it comes to questions of impartiality or bias and that, therefore, the starting presumption should be that an arbitrator should never accept appointments in multiple references involving overlapping issues and only one common party without giving disclosure. Both the LCIA and the ICC were supportive of such an interpretation (which is not of itself surprising given the obligations imposed with regard to arbitrator disclosure in their respective arbitration rules) and referred to an “international pro-disclosure consensus”. Their view was that the English law approach on this issue was not strict enough.
In response, Chubb argued that the accepted practice in maritime and insurance disputes where arbitrators are routinely appointed in multiple arbitrations is difficult to reconcile with a presumption that concurrent appointments in different arbitrations must always be disclosed. Chubb also argued that section 24(1)(a) of the AA 1996 referred to “impartiality”, not “independence”, and that this was a deliberate act on the part of the draftsman in recognition of the need for parties to choose from a relatively small pool of arbitrators with specific expertise in specialist fields. The London Maritime Arbitration Association and the Grain and Feed Trade Association provided written submissions to the court in support of this view, highlighting the practices in their particular areas of law in which fewer arbitrators and appointments in multiple, concurrent arbitrations are generally accepted. Chubb also advanced the possibility that parties could use the disclosure process in a tactical way effectively to veto arbitrators appointed by the other party.
However, one interest does not necessarily preclude the other, since parties would always have the freedom to accept the appointment of an arbitrator who had disclosed other connected appointments. While in this case Halliburton felt that the subsequent appointments, and their non-disclosure, potentially compromised the arbitrator’s impartiality, they could have just as easily approved his appointment had they felt differently and been inclined to do so. It remains an open question whether a timely disclosure would have alleviated any concerns. Furthermore, Chubb’s and M’s failure to disclose the subsequent appointments arguably meant that significant (and potentially wasted) time and costs had been expended prior to Halliburton’s discovery of M’s further appointments by Chubb. Had M or Chubb made immediate disclosures, these issues may have been resolved at an earlier stage of the proceedings.
The LCIA further argued that courts should give “little or no weight” to an arbitrator’s reputation and experience when considering the risk of bias. In this respect, Halliburton also questioned Popplewell J’s reasoning at first instance that experienced arbitrators should be able to compartmentalise their minds and decide cases purely on the facts presented to them in each one. Such an approach inevitably posed the question of how English law and international practice could effectively ensure that this was always the case. Halliburton felt that such an expectation would subject parties to an unreasonable risk of conflict.
The Court of Appeal’s decision to recognise an obligation to disclose but only enforce that obligation if there is “something more” to the situation was also addressed by the Chartered Institute of Arbitrators (CIArb). The CIArb argued that this was not a case of M failing to consider whether he needed to disclose the further appointments, but rather that he had considered it and then decided not to disclose it. The Supreme Court may need to consider whether by “something more” the Court of Appeal intended that the non-disclosure had to be deliberate and ill-intentioned. If so, this approach would shift the burden to the innocent party to prove ill intent (which may not always be possible or easy to do) and potentially let the party with actual knowledge of the circumstances off the hook.
There certainly does not appear to be any clear cut answer to the conundrum. The Supreme Court will need to balance the need for arbitrators to be able to have overlapping appointments in areas of law where the number of subject matter experts may be limited, and on the other hand, the need to preserve global confidence in the impartiality of the arbitration process as well as policy considerations regarding England’s reputation as a centre and preferred seat for international arbitration. The English courts’ approach thus far appears to lean towards a more flexible and less prescriptive approach. It remains to be seen whether the Supreme Court will follow in that vein or whether it will opt for the “gold standard” as espoused in the rules of most of the leading arbitral institutions. The global arbitration community awaits the outcome with bated breath.
