Back to Arbitration Committee publications

Duncan Bagshaw
Howard Kennedy, London
duncan.kennedy@howardkennedy.com

The Chartered Institute of Arbitrators (CIArb) is a global organisation, based in the UK, which trains, certifies and regulates arbitration practitioners. Like several other international arbitration bodies, CIArb was granted permission to intervene in the Supreme Court case of Halliburton Company v Chubb Bermuda Insurance.

An intervenor is a third party, not a party to the case in question, who is permitted to make submissions with the permission of the Court, to provide assistance to the court on the issues, or because the case involves significant issues of public interest.¹ Around 30 per cent of cases before the UK Supreme Court involve submissions from at least one intervenor.

CIArb’s reasons for intervening were expressed in its submissions as follows:

‘First, it is the world’s leading institution for the education and training of arbitration practitioners and arbitrators and can therefore assist the Court by providing a unique insight as to the generally accepted standard of arbitrator disclosure. Second, CIArb as a body representing a vast body of members of the international arbitration community, and with a wide global outreach, will be able to provide the Court with balanced observations in relation to the effects of the outcome of this appeal, representing the views of the international arbitration community at large.‘

CIArb also referred in its submissions to its role in addressing the behaviour of arbitrators who are members of CIArb. As part of its function, CIArb can censure members who behave in a way which is injurious to the good name of CIArb, renders a person unfit to be a member of CIArb or is likely to bring CIArb into disrepute, amongst other matters. This is provided in section 15.2 of the bye-laws of CIArb.²

It seems appropriate to discuss here the extent to which the Supreme Court’s judgment addresses the concerns which CIArb expressed about the conduct of Kenneth Rokison in its submissions, and which it would be concerned about, if its members were to exhibit similar conduct.

Those concerns were essentially expressed in two points.

First, the arbitrator seemed to have ignored his obligation to treat all of the parties fairly. By putting himself in a position where one party (Chubb) had a practical advantage over the other parties to the arbitrations (Halliwell and Transocean), he seemed willingly to accept multiple appointments with the immediate risk of favouring the party which appointed him (in one arbitration, and proposed him for appointment by the Court in the other), and therefore possessed knowledge not held by the other party

Second, the arbitrator appeared not merely to have failed to apply his mind to the problem of whether to disclose this situation, as the Court of Appeal appeared to conclude. It appeared to us that he had in fact considered it and decided that there was no obstacle to his accepting the appointment in the Transocean arbitration without disclosing it to Halliburton.

Unfairness in the acceptance of multiple appointments without disclosure

The Supreme Court was satisfied that the circumstances existing when Rokison was appointed in the second case were sufficient to give rise to potential unfairness to Halliburton and Transocean. As Lord Hodge said:

‘I do not interpret the Court of Appeal as saying that the acceptance of multiple appointments can never be sufficient of itself to give rise to the appearance of bias. But if that is what the court meant, I would respectfully disagree, especially because the inequality of knowledge between the common party and the other party or parties has the potential to confer an unfair advantage of which an arbitrator ought to be aware.‘

The need to ensure fairness is fundamental to the obligations of an arbitrator. It is enshrined in section 33 of the Arbitration Act 1996 (the 1996 Act), which provides that arbitrators shall ‘act fairly and impartially as between the parties‘. CIArb would expect its members to comply strictly with that duty.

Lady Arden went somewhat further than Lord Hodge in her separate judgment.

‘[A]n arbitrator should proceed on the basis that a proposal to take on a further appointment involving a common party and overlapping subject-matter (in that it arises out of the same event) is likely to require disclosure of a potential conflict of interest. The fact that an arbitrator is to be trusted to decide the case on the evidence is not a complete answer to the objections based on inequality of arms and material asymmetry of information that have been raised by Halliburton.‘

Lord Hodge even acknowledged implicitly that the result of the case might have been different if the assessment was made at the time of the acceptance of the further appointments:

‘I am of the view that the fair-minded and informed observer, if he or she had considered the question at or around the date of his acceptance of appointment in reference 2, may well have concluded that there was a real possibility of bias. But it is not necessary to express a concluded view on this as that is not the correct time to ask the question [emphasis added].‘

The concern is that the decision of the Supreme Court in the result (finding that that award should stand) may distract from the fact that the Court found that the arbitrator allowed himself to be in the position of giving one party an advantage over others at the time of his second and third appointments. That is a position which CIArb would generally expect its members to avoid.

To think or not to think?

CIArb expressed in its submissions some concern regarding the treatment of the arbitrator’s state of mind in failing to make any disclosure of the later appointments to Halliburton. It was accepted at all levels of the court process that ‘the failure to disclose was accidental and not deliberate‘ (as the Court of Appeal put it). Popplewell J said that ‘it did not occur to [Rokison] at the time that he was under any obligation under the International Bar Association (IBA) Guidelines to do so.‘

The arbitrator’s actual words when asked to provide an explanation for not making a disclosure were ‘I do not think and did not think that the above circumstances put any obligation upon me to make any disclosure to you or your clients under the IBA Guidelines.‘

These words made it very difficult for the CIArb to accept the Court of Appeal’s characterisation of Rokison’s failure to disclose as ‘accidental‘, or Popplewell J’s description of his state of mind which suggested that disclosure ‘did not occur to him‘. There is a significant difference between ‘I considered it and decided I was not obliged to disclose‘ and ‘I did not consider it at all‘. The latter is difficult to reconcile with Rokison’s expression ‘I did not think that the above circumstances put any obligation upon me to make any disclosure’. Would anyone who had not turned their mind to an issue at all use those words to describe their thought process? It is arguably not a natural way to express it.

This stance was not taken lightly on behalf of CIArb. No other party questioned the Court of Appeal’s position on the issue. But given the above analysis of the arbitrator’s words, and CIArb’s concern that the decision was made on a proper factual footing, we raised this issue for consideration.

This reflected CIArb’s concern to emphasise that they would expect their members to consider the possibility of an appearance of bias in any similar case and to make a disclosure to the parties, or decline to accept the later appointments. This does not mean that CIArb members could never act in multiple arbitrations in related cases. But it does mean that CIArb would expect an arbitrator in Mr Rokison’s position to at least turn their mind to the issue. Mr Rokison himself acknowledged, with the benefit of hindsight, that it would have been prudent for him to have informed Halliburton through its lawyers and apologised for not having done so.

Bootstraps?

Having accepted the potential advantage conferred by the inequality of knowledge between the common party and the other parties at the time of appointment, the Supreme Court found that Rokison ought to have disclosed his involvement, but that his failure to do so did not mean that his award should be set aside ‘having regard to the circumstances known to the court at the date of the hearing at first instance‘ because:

‘It was likely that references 2 and 3 would be resolved by the preliminary issue and that there would not be any overlap in evidence or legal submissions between them and reference 1. As the arbitral tribunal had held hearings on the preliminary issues in November 2016, Mr Rokison would have been aware of its likely decision when he corresponded with K & L Gates in December 2016 and January 2017 (paras 19-22 above). Indeed, the awards handed down on 1 March 2017 revealed that his discreet prediction was correct. If that had not been the outcome of the preliminary issues, he had also offered to consider resigning from his appointments in references 2 and 3.’

My concern about this conclusion is that it fails to acknowledge the irreversible impact of a tribunal member having laboured under a disqualifying appearance of bias for part of the arbitrations, including while he was making substantive decisions. Once that has occurred, it seems difficult to understand how the arbitrator’s own actions could serve to avoid removal. Can arbitrators pull themselves up by their own bootstraps?

Section 24 of the 1996 Act allows an arbitrator to be removed when circumstances exist that give rise to justifiable doubts as to their impartiality. There is no requirement to show substantial injustice. If that is the case, it is not easy to see why events subsequent to an appearance of unfairness can relieve the arbitrator from being removed. Is there much difference between requiring substantial injustice to be shown, and the Supreme Court’s finding that, in spite of an appearance of bias, the arbitrator would not be removed because of the practical implications of the way that the case was subsequently decided by that arbitrator?

For Halliburton, it may be felt that this clear policy decision in the 1996 Act not to tolerate any doubts as to partiality (even where substantial injustice has not been shown) has been overridden by the application in this case of the rule that the decision on removal is made taking into account the facts at the date of the hearing. That feels rather like adding a requirement of substantial injustice by the back door. The Supreme Court appears to have given a judgment which says (to paraphrase) ‘well, the way the case turns out shows that you did not suffer injustice in the end, notwithstanding the appearance of a possibility of bias in the earlier stages.’

The decision also seemed to ignore the fact that the events before the first instance decision, which informed the decision that the award could stand, were largely events involving acts by the arbitrator himself, or by a panel of which he was a member. The primary event was the decision on the preliminary issue. It is difficult to understand why a disqualifying characteristic of an arbitrator can be overridden by one or more decisions made by that arbitrator, or any communications by that arbitrator with the parties, which have the result that the potential unfairness has been eliminated by the result of those decisions or communications. To do so does have a feeling of ‘bootstraps’ about it.

From CIArb’s point of view, the message in the Supreme Court was clear: one would expect an arbitrator in this position to consider and identify the potential unfairness and, at the least, to disclose it before continuing with the appointment.