The ability to select an arbitrator is widely considered one of the most valuable characteristics of international arbitration. According to the Queen Mary University and White & Case 2015 International Arbitration Survey, selection of arbitrators was considered its fourth most important characteristic, with 38% of respondents rating it among their top three.
Surely, then, it is all-important that parties can remove an arbitrator if he/she proves unable, or unwilling, to conduct the proceedings properly, acting fairly as between the parties?
Following the recent case of Enterprise v U-Drive, in which the applicant made several allegations against the arbitrator, all of which insufficient to warrant his removal, this two-part series is intended as a timely reminder of just how difficult it can be to remove a troublesome arbitrator.
This Part 1 focuses on allegations of apparent bias.
The test for apparent bias
The right of a party to arbitration proceedings seated in England to apply to court to remove an arbitrator is set out in section 24 of the English Arbitration Act 1996 (the Act). Section 24(1)(a) provides that a party may do so on the ground “that circumstances exist that give rise to justifiable doubts as to [the arbitrator’s] impartiality”.
In Cofely Limited v Bingham & Knowles Limited  EWHC 240 (Comm), Hamblen J set out at  a thorough exposition of the test to be applied under section 24(1)(a):
- The section 24 test for apparent bias is the same as the common law test, e., whether “the fair minded and informed observer, having considered the facts, would conclude that there was a real possibility that the tribunal was biased”.
- Such an observer, although not a lawyer, is assumed to be in possession of all the facts and aware of how the legal profession operates in practice.
- A “fair-minded” observer reserves judgment until he/she has fully understood both sides of the argument: the assumptions made by the complainer are not to be attributed to the observer unless they can be justified objectively.
- An “informed” observer takes a balanced approached and appreciates the importance of context.
Since the Act came into force, other than Cofely itself (which is considered below) there have been very few successful applications under section 24(1)(a):
- In Save and Prosper Pensions v Homebase  L. & T.R. 11, the arbitrator was a partner in a firm which had been instructed in a different matter by a company which had the same parent company as Homebase.
- In Sphere Drake Insurance v American Reliable Insurance  EWHC 796 (Comm), the arbitrator had been retained by American Reliable’s underwriters as consulting expert in the context of earlier proceedings in which findings were made against them, on which findings Sphere Drake relied in the arbitration. The arbitrator’s involvement had included meeting persons who might give evidence in the arbitration.
- In ASM Shipping v TTMI  1 Lloyd’s Rep 375, the arbitrator had previously appeared as advocate in proceedings in which serious allegations of impropriety had been made against ASM’s main witness.
- In Sierra Fishing v Farran  1 All ER (Comm) 560, the arbitrator was a partner in his father’s law firm. That firm and his father acted for one of the parties and derived significant financial benefit from doing so. The arbitrator had also been involved in drafting, and had advised one of the parties in relation to, an agreement which contained the arbitration clause out of which his contested jurisdiction was said to arise. The arbitrator failed to disclose those matters to the parties.
Applying the test in practice
There may frequently be doubt in the mind of a party as to whether an arbitrator has unfairly, or prematurely, taken against it. But when are those doubts justifiable? The answer is not always obvious.
In 2016, the English courts have already been asked three times to grapple with the issue of apparent bias.
Cofely v Bingham & Knowles (8 February 2016)
In this case, Cofely successfully applied to remove the arbitrator, Mr Anthony Bingham, on grounds of bias.
The facts in Cofely were fairly stark, and the decision unsurprising. Over the previous three years, 18% of the arbitrator’s appointments and 25% of his income as arbitrator had derived from cases involving Knowles (either as party to the proceedings or acting for a party as claims consultant). The arbitrator did not disclose his past involvement with Knowles to Cofely, and behaved evasively and defensively when asked questions about it, both in correspondence and at a hearing. In response to Cofely’s section 24 application, he filed an aggressive and unapologetic witness statement seeking to exonerate himself.
Hamblen J held that, cumulatively, those matters demonstrated a real possibility of apparent bias, and made an order to remove the arbitrator if he would not resign.
Enterprise v U-Drive (15 June 2016)
In Enterprise Insurance Company plc v U-Drive Solutions (Gibraltar) Limited & James Drake QC  EWHC 1301 (QB), HHJ Moulder QC refused Enterprise’s application to remove the arbitrator.
Enterprise’s section 24(1)(a) application alleged bias on the basis that the arbitrator had:
- repeatedly made procedural findings in favour of U-Drive (of which Enterprise listed 11 instances); and
- given undue latitude to U-Drive, notwithstanding its history of procedural default, in particular by way of several extensions of time, including in respect of compliance with a peremptory order in respect of which he had previously ruled that there would “be no further extension of time for compliance”.
The judge held that the test for bias was not made out, in that:
- it was “not for this court to substitute its view for the decisions made by the arbitrator” and that “the mere fact that Enterprise failed in its various applications…cannot…possibly lead to any inference that there was a real possibility that the tribunal was biased”; and
- “[t]he fact that a different arbitrator may have taking [sic] a different view on the facts does not mean that the fair-minded and informed observer would conclude that there was a real possibility that the tribunal was biased”.
It cannot compellingly be said that Enterprise is an example of arbitrator bias. At worst, the arbitrator may have been afflicted by “due process paranoia”, an increasingly common phenomenon characterised by arbitrators’ reluctance to make decisive case management decisions, for fear of their award being challenged on the basis of a party not having had a proper opportunity to present its case. Perhaps more likely, however, is that the arbitrator simply wished, as he put it, “to seek a substantive resolution of the dispute between the parties, and not to allow form to trump substance” by striking out U-Drive’s case.
In effect, Enterprise’s argument was that the arbitrator’s paranoia (if that is what it was) gave rise to such a degree of leniency towards U-Drive as to amount to bias in its favour. Ultimately, that was not sufficient.
Harb v Aziz (16 June 2016)
The most recent authority on bias is the decision of the Court of Appeal in Harb v Prince Abdul Aziz bin Fahd  EWCA Civ 556, involving allegations of bias against a judge of the English High Court, Peter Smith J.
Although the case involved the removal of a judge, it is instructive because the test under section 24(1)(a) is the same.
Harb has been the subject of considerable commentary; it is sufficient for present purposes to note the Court of Appeal’s conclusion that even the judge’s express animosity toward counsel for the applicant did not necessarily, and did not in that case, lead to a reasonable apprehension of bias against the party itself.
The distinction between animosity toward a party and its advisers may, depending on the circumstances, be justifiable. However, it is not a distinction which has previously been drawn: in a previous decision involving the same judge’s animosity towards a party’s solicitors, Howell v Millais  EWCA 720, the Court of Appeal held that the judge’s animosity toward the applicant’s solicitors, together with his intemperate dealing with the initial recusal application, was sufficient evidence of apparent bias.
Howell was not referred to in Harb, and it remains to be seen whether Harb is the start of a trend toward raising the bar for demonstrating apparent bias.
Conclusions and practice notes
Although it is not always clear from the authorities how the test for apparent bias will be applied, in light of Harb the hurdle may be rising. When advising a party as to whether it is likely to succeed in removing an arbitrator for apparent bias, the starting point must be that the evidence must not only be very compelling, but must also demonstrate bias against the party itself as opposed to its advisers.
Part 2 of this series will consider other routes to removal under section 24 and how to avoid losing the right to object to the arbitrator under section 73.