
Every day is Independence Day – Two successful arbitrator challenges for “apparent bias”
In this excellent summary (produced here by courtesy of Mr Markus Esley of Haynes and Boone LLP), Mr Esley looks as recent cases where arbitrator appointments have been successfully challenged.
With thanks to Mr Esley for allowing me to reproduce here.
Every day is Independence Day – Two successful arbitrator challenges for “apparent bias”
The independence and impartiality of arbitrators matters more than ever. Parties choose to have their disputes resolved by arbitral tribunals because they wish to have their differences decided by experienced decision-makers who can be trusted to apply the applicable law without fear or favour, rather than litigate before national courts in particular those of the home state of their counterparty. While tactical or unmeritorious challenges to arbitrators, alleging lack of impartiality, are not uncommon in international arbitration, arbitral tribunals can generally be relied on to uphold these principles and live up to the high standards expected of them.
Sometimes though arbitrators slip up. They must remain vigilant to ensure they do not do so, be it unintentionally, for want of experience, or due to a lack of appreciation of the appearance that certain actions can create. In the last six months, the English Commercial Court has upheld challenges to two arbitrators for lack of impartiality, or `apparent bias’ as English law terms it, in (i) H1, H2 v W, D, F [2024] EWHC 382 (Comm) (February 2024) (“H v W”) and (ii) Aiteo Eastern E & P Company Ltd v Shell Western Supply and Trading Ltd [2024] EWHC 1993 (Comm) (August 2024) (“Aiteo v Shell”). In H v W, Calver J was faced with a challenge arising out of unguarded and illadvised statements made by a sole arbitrator at a procedural hearing, as to the weight of certain expert evidence: the decision illustrates the importance of not pre-judging the merits. In Aiteo v Shell, Jacobs J dealt with another, perhaps more common situation, namely where an arbitrator has previously been appointed or otherwise professionally engaged by the same firm of solicitors that is representing one of the parties to the arbitration. This article considers both these decisions, and the lessons that can be learnt from them.
The principles of independence and impartiality in international arbitration
These principles of independence and impartiality are enshrined in the leading institutional arbitration rules and national laws. Looking first at institutional rules, they bind the parties and also the arbitrators as a matter of contract, and is of course possible to undertake an obligation of independence and impartiality by agreement. Article 22(4) of the ICC Rules of Arbitration (2021) states that: “In all cases, the arbitral tribunal shall act fairly and impartially and ensure that each party has a reasonable opportunity to present its case.” A breach by an arbitrator of that duty entitles a party to challenge him or her under Article 14(1) “… for an alleged lack of impartiality or independence, or otherwise.” Similar provisions are found in the LCIA Rules of Arbitration, Article 10.1 which permit a party to challenge an arbitrator if he or she proves to be “unfit to act” or “circumstances exist that give rise to justifiable doubts as to that arbitrator’s impartiality or independence”.
The UNCITRAL Model Law on International Commercial Arbitration, which forms the basis for many national arbitration laws worldwide, also imposes an ongoing duty on arbitrators to disclose without delay “… any circumstances likely to give rise to justifiable doubts as to his impartiality or independence.” (Article 12(1)): a similar continuing duty of disclosure is imposed by the LCIA Rules (Article 5.4) and the ICC Rules (Article 11(3)). Where such justifiable doubts exist, an arbitrator may be challenged (Article 12(2) of the UNCITRAL Model Law).
In English law, the Arbitration Act 1996 governs the issue. The 1996 Act applies to all arbitration proceedings with their seat in England and Wales. Section 33(1) of the 1996 Act sets out the general duty of any arbitral tribunal, requiring it to:
“(a) act fairly and impartially as between the parties, giving each party a reasonable opportunity of putting his case and dealing with that of his opponent, and (b) adopt procedures suitable to the circumstances of the particular case, avoiding unnecessary delay or expense, so as to provide a fair means for the resolution of the matters falling to be determined.”
Section 33(2) provides that this duty applies to the conduct of the arbitration, decisions on procedural and evidentiary matters, and the exercise of all other powers conferred on arbitrators. Under Section 24(1)(a) of the 1996 Act, arbitrators can be challenged where “circumstances exist that give rise to justifiable doubts as to the arbitrator’s impartiality.” One point to note is that under the 1996 Act, there is presently no statutory duty on arbitrators to disclose matters that could put their impartiality into question, but such a duty arises at common law as discussed further below. In October 2023, the English Law Commission published a final report on its proposals for reform of the 1996 Act, suggesting that the duty of disclosure should be codified, placing all arbitrators in proceedings with an English seat under a ” … continuing duty to disclose any circumstances which might reasonably give rise to justifiable doubts as to their impartiality.” No such legislative reform has yet to be enacted.
English law: the `fair-minded observer’ and the appearance of bias
Under English law, arbitrators must not only be impartial, they must also be seen to be impartial. Arbitrators can be challenged for lack of impartiality not only where there is actual bias, but also in cases of `apparent bias’. The question is whether a fair-minded and informed observer (a judicial construct), having considered the facts, would conclude that there was a real possibility that the tribunal was biased. The leading decision is that of the Supreme Court in Halliburton Co v Chubb Bermuda Insurance [2020] UKSC 48., where the principles were authoritatively restated:
The obligation of impartiality is a core principle of arbitration law, and it applies equally to party-appointed arbitrators and to presiding arbitrators appointed, no matter how they are appointed (either by the agreement of the parties, the party-appointed arbitrators, by an arbitral institution pursuant to the applicable rules, or by the Court).
The test for apparent bias is entirely objective. It is made out where the fair-minded and informed observer (not the parties themselves) would consider that there was a real possibility that the tribunal was biased.
English law imposes on arbitrators a duty of disclosure that operates by reference to the same objective standard: “… the arbitrator’s legal obligation of disclosure imposes an objective test. This differs from the rules of many arbitral institutions which look to the perceptions of the parties to the particular arbitration and ask whether they might have justifiable doubts as to the arbitrator’s impartiality. The legal obligation can arise when the matters to be disclosed fall short of matters which would cause the informed observer to conclude that there was a real possibility of a lack of impartiality. It is sufficient that the matters are such that they are relevant and material to such an assessment of the arbitrator’s impartiality and could reasonably lead to such an adverse conclusion.”
The fair-minded observer does not reach judgment before acquiring a full understanding of both sides of the argument. Their conclusions must be justified objectively. In reaching those conclusions, the observer displays neither complacency nor undue suspiciousness.
In applying the test, the English Courts pay regard to the particular characteristics of international arbitration, which strongly favour proper disclosure of prior relationships or other matters, in order to maintain the integrity of the process.
These characteristics include (i) that arbitration is a private process with limited public oversight; (ii) an unsuccessful party has limited opportunity to appeal an award; (iii) arbitrators have a vested financial interest in obtaining further appointments from the parties, and so may wish to avoid alienating a party to an arbitration; (iv) arbitrators come from a wide range of backgrounds and jurisdictions, and may have different notions of what is and is not ethically or professionally acceptable conduct; (v) under English law, however, all arbitrators must adhere to the same standard of behaviour; and (vi) opportunistic or tactical challenges are (unfortunately) a possibility.
In conclusion, as Jacobs J put it in Aiteo v Shell:
“… the assessment of the fair-minded and informed observer, as to whether there is a real possibility of bias, is an objective assessment which has regard to the realities of international arbitration and the customs and practices in the relevant field of arbitration.”
H v W: the dangers of appearing to make a pre-determination
Turning to the first of our two cases, H v W concerned an arbitration in the film industry, which the judge described as close-knit with relatively few participants. The identity of the cast of characters in the case remains unknown, since Calver J made an order preserving the anonymity of the parties and of the sole arbitrator.
The facts were as follows. A film production company, the claimant in the eventual arbitration, had taken out an insurance policy with the respondent insurer. The policy was to cover costs and expenses resulting from any accidents causing death or injury on set. Sadly, during filming in Sweden, the lead actor suffered serious injuries when a stunt scene led to an accident. The production company made a claim under the policy for resulting expenses and the costs of delays to concluding filming. The insurer rejected the claim, arguing that the policy did not apply since the injuries had been suffered as a result of deliberate participation in a hazardous activity, and that the production had breached a condition of the policy which required it to take reasonable care to avoid or mitigate the risk of any injuries.
The production company commenced arbitration proceedings under the policy, arguing that the stunt scene had been properly planned and was not an excluded hazardous activity. It also said that it had discharged its duty of care under the policy by hiring a stunt coordinator who had duly signed off on the scene as planned. The production company relied on the fact that, it said, it was the usual industry practice – in Sweden, Scandinavia and indeed worldwide – to rely on the safety assessments of stunt coordinators. As against that, the insurer argued that the duty could not be delegated in this way, and the production company remained ultimately responsible for safety.
The parties could not agree on the identity of an arbitrator. Pursuant to the arbitration clause in the policy, the British Film Institute nominated “W” as sole arbitrator. He had several decades of experience as producer in the film industry, but apparently lacked any arbitral experience. The arbitration progressed quickly following the constitution of the tribunal. In July 2023, the Statement of Claim was served. The insurers served their Defence in August 2023. A four-day evidentiary hearing was set down early December 2023. In mid-October 2023, the first procedural hearing then took place. Witness statements were served by the end of October 2023, with expert reports a couple of weeks later in early November 2023.
The production company gave notice that it would rely on the evidence of a number of experts on the issue of market / industry practice, as to whether a stunt coordinator was indeed considered as ultimately responsible for safety on set. That evidence would go to the central issues in the arbitration, namely the interpretation of the exclusion clause (`hazardous activities’) and whether the insured complied with the condition to take reasonable steps to ensure safety on set in the policy. Both sides duly adduced expert reports by experienced film and television producers and stunt coordinators. Unexpectedly, the insurer also served an expert statement by a particular stuntman and stunt coordinator who had previously been instructed by the claimant insured to investigate the accident that had led to the claim under the policy. In that capacity as independent investigator, the expert had issued a (apparently non-privileged) report to the Swedish film industry regulatory body. In his written statement for the insurer in the arbitration, the expert witness took the position that ultimately, safety on set always remained the responsibility of the production company supporting the insurer’s position.
In mid-November 2023, a second procedural hearing took place. At this hearing, the sole arbitrator identified the factual witnesses whom he wanted to appear at the hearing, but (according to the parties’ notes of the procedural hearing), added that “For me, I don’t need to hear any of the expert witnesses. I don’t think they will add any value. I know what they are saying. They are exceptional people in their fields. They are the best, but I don’t need them to say what is normal on a film. I know what is normal on film.” Counsel for the insurer responded that he did want to cross-examine the experts for the insured. This led the arbitrator to reply “Look, if you want to cross examine the expert witness that is fine by me, but I don’t think we need to listen to them. I know them all personally extremely well on the insured side. I don’t know your expert …” adding in relation to the insured’s experts “I know all three very well, I am extremely good friends with them …”. With respect to one of the insured’s experts, a wellknown Norwegian producer who testified as to the industry practice in Scandinavia, the arbitrator said “… what he says is what I will believe but what he says is not how it operates in the rest of the world. He says stunt coordinator is in charge of safety on set in Scandinavia. I absolutely believe what he says. But it’s not what happens elsewhere in the world.” Based on the parties’ notes, then, the sole arbitrator had made it plain that this expert was so well-respected that the arbitrator would believe his evidence even though it went against what the arbitrator himself thought.
The debate at the second procedural hearing also extended to the insurer’s expert who had previously been engaged by the insured as independent investigator of the accident. The sole arbitrator said that while that expert’s evidence seemed material, it appeared to the arbitrator that the expert had a conflict of interest, a notion that counsel for the insured picked up on and reinforced in his submissions, suggesting that it would be “highly inappropriate” if this expert were allowed to appear for his new paymasters. The sole arbitrator commented: “From my point of view, I believe that his evidence to either party should be disallowed. I don’t think that we should allow it. He cannot change sides half-way through. I think it is absolutely wrong.” Counsel for the insurer argued against the evidence being excluded, submitting that there was no property in a witness and noting that the expert in question had been brought in as an independent investigator and had issued a report to the Swedish industry regulatory body. Counsel for insurer was right in making this point: by acting as an independent investigator, the expert had not disqualified himself from giving evidence as to the result of his investigations even if that evidence was adduced by a different party than the one who had commissioned his independent report in the first place. The sole arbitrator however did not seem to be swayed by such submissions, stating that while the expert’s interview notes and results of the investigation were in his view “very relevant”, “… now that he has switched, I find his action incredible.”
Following the second procedural hearing, the insurer’s solicitors wrote to the sole arbitrator to require further disclosure of his relationship with the insured’s experts, further to the arbitrator’s comments such as “I know them all personally extremely well on the insured side”. In response, the sole arbitrator confirmed that he knew them and had worked with them all on at least one film and had met them socially over the years on a few occasions, but that he had no shared financial interests with any of the individuals. He added that, given the nature of the industry, it would be very surprising if he did not know, or had not previously worked with, anyone who was qualified to act as an expert.
Apparent bias and the proper role of `industry’ arbitrators
The insurer challenged the sole arbitrator under Section 24 of the Arbitration Act 1996 for lack of impartiality, arguing that his conduct showed that he favoured the insured’s experts simply because he knew and respected them, and had made an impermissible pre-determination of the issues or evidence against the insurer. Calver J applied the Supreme Court’s test in Halliburton. He held that the fair-minded observer would take note of the fact that this was an arbitration in a relatively small industry, where it was likely that specialist arbitrators would, as a result of their (non-arbitrator) work, become acquainted with other industry participants who might end up playing a role in the proceedings. This would not by itself give rise to a risk of apparent bias. Indeed, as previously held, the parties are taken to agree on an industry arbitrator (or to having their arbitrator nominated by an industry body such as the British Film Institute in H v W) precisely because of their first-hand experience in the trade, and would be unlikely to consider it significant that the arbitrator’s ability to act impartially was affected because of previous commercial dealings with other industry members (see Rustal v Gill & Duffus [2000] 1 Lloyd’s Rep).
An industry arbitrator chosen for their technical skill and knowledge might not adopt the same sophisticated case management procedures for an arbitration as would a King’s Counsel or an international arbitration practitioner. Again, Calver J noted that this would not surprise the fair-minded observer. An industry arbitrator would also be expected to use their own expertise and knowledge to evaluate the evidence presented by the parties, but – like any other arbitrator – he or she must always conduct an objective assessment of that evidence without supplementing or supplanting it, or prejudging or dismissing it out of hand. In Bubbles & Wine Ltd v Lusha [2018] EWCA Civ 468 at [34] Leggatt LJ stated:
“There is nothing wrong with a judge indicating provisional views, and advocates are generally grateful for such indications as it gives them an opportunity to correct any misconception which the judge may have and to concentrate in their submissions on those points which appear to be influencing the judge’s thinking. The expression of such views could only be thought to indicate bias if they are stated in terms which suggest that the judge has already reached a final decision before hearing all the evidence and argument.”
Arbitrators, like judges, may properly indicate a predisposition towards a particular outcome, but they must allow the parties the opportunity to persuade them otherwise, always remaining prepared to weigh up and consider submissions and evidence. If the arbitrator’s conduct is such that it suggests instead that he or she has made a predetermination, that will give rise to a real possibility of bias. That same risk arises where there is a “… real possibility that extraneous factors have played a part in the arbitrator’s decision, by the arbitrator taking into account something that should not legitimately be part of the decision-making process at all.” In each situation, the `fair minded observer’ test of apparent bias is met.
Applying these principles to the facts, Calver J concluded that the sole arbitrator’s prior relationships (such as they were) with the expert witnesses were not a cause for concern. They were nothing more than would be expected in the case of someone with more than three decades of experience in the industry. Turning to the sole arbitrators’ comments about the expert investigator who had allegedly `switched sides’, while these statements betrayed inexperience, they were on balance not such as to show a closed mind, and could have been corrected at the evidentiary hearing. While the sole arbitrator had at one stage (regrettably) said that the expert’s evidence “should be disallowed” because of their having agreed to testify for the insurer, he had not in the end made such a ruling.
Where the sole arbitrator fell down, however, was in relation to the comments he had made about the insured’s other experts: the fair-minded observer would conclude that the arbitrator had prejudged the merits of the dispute. By describing the insured’s experts as “exceptional people in their fields” and noting that “what [the expert] says is what I will believe”, the sole arbitrator had created the appearance of being influenced by extraneous factors placing weight solely on the identity of the experts and their standing in the industry, as opposed to assessing their actual evidence. When counsel for the insurer had urged the arbitrator to allow cross-examination, the sole arbitrator had not helped himself by saying that “I will of course reserve my judgement” but immediately adding “… but I have read the statements and I know the professionals. I can say now what I think.” In conclusion, Calver J upheld the Section 24 challenge, finding that:
“… the parties might expect the arbitrator to use his special knowledge of the film industry to understand the evidence that is given at the arbitration and any usages of the trade. But they would not expect him to use his special knowledge to pre-determine that he would prefer the evidence of an industry insider whom he knew (both personally and by reputation) over one whom he did not know. Rather, they would expect him to undertake an objective assessment of the evidence after he had heard it and heard it tested in cross examination.”
Aiteo v Shell: The dangers of relational contacts between arbitrators and counsel
The second case, Aiteo v Shell, is concerned with a different issue: when do appointments of an arbitrator by a law firm representing one of the parties, or other professional engagements of the arbitrator by that law firm, give rise to apparent bias under English law? The case concerned a dispute between Aiteo, a privately-owned oil and gas company based in Nigeria, and Shell and other lenders concerning facility agreements for around US$ 2 billion which Aiteo was to use to acquire interests in Nigerian oilfields and related facilities. The loans were split between two contracts, an Onshore and an Offshore Facility Agreement. The Onshore Facility Agreement was governed by Nigerian law, with English law governing the Offshore Facility Agreement. Both contracts provided for all disputes to be resolved by ICC arbitration seated in London.
In December 2020, Shell, represented by Freshfields Bruckhaus Deringer, commenced an arbitration under the Offshore Facility Agreement, arguing that Aiteo had failed to repay principal and interest when it had fallen due. The other lenders, also represented by Freshfields, followed suit under the Onshore Facility Agreement. Shell and the other lenders each requested that the two ICC arbitrations be consolidated into a single reference the Onshore and Offshore Facility Agreements contained specific clauses dealing with this. A tribunal of three arbitrators was to be appointed for each arbitration, pending a decision on consolidation. Shell nominated Dame Elizabeth Gloster DBE (“DEG”) as its party-appointed arbitrator in the offshore arbitration. Since retiring from the Court of Appeal in 2018, DEG has become a sought-after and high-profile international arbitrator. The other lenders, again through Freshfields, also nominated DEG in the onshore arbitration. Aiteo immediately objected to DEG’s nomination in both arbitrations. This was because Aiteo also objected to consolidation of the two arbitrations and therefore, Aiteo argued, it was inappropriate for one arbitrator to have access to material produced in both references.
DEG, having been nominated, had signed the ICC’s statement of acceptance, availability, impartiality and independence for both the arbitrations (the “ICC Arbitrator Statement”). The ICC Arbitrator Statement is governed by Article 11.2 of the ICC Rules, which provides that:
“The prospective arbitrator shall disclose in writing to the Secretariat any facts or circumstances which might be of such a nature as to call into question the arbitrator’s independence in the eyes of the parties, as well as any circumstances that could give rise to reasonable doubts as to the arbitrator’s impartiality.”
In DEG’s ICC Arbitrator Statement, she disclosed that she had been appointed by Freshfields as arbitrator in two unrelated arbitrations over the last two years, so from December 2018 to December 2020. As subsequently transpired, DEG did not however disclose an instruction by Freshfields to give `expert advice’ in conference over the period of June to July 2020. The number of disclosed professional contacts between DEG and Freshfields was thus two, with one undisclosed professional contact. To that one has to add two more nominations by Freshfields: the onshore and offshore arbitrations with Aiteo.
Aiteo’s challenge to DEG as arbitrator in both these, as-yet unconsolidated, arbitrations, was decided by the ICC Court pursuant to Article 14(3) of the ICC Rules. On 17 June 2021, the ICC Court determined that DEG could be appointed in either the onshore or the offshore arbitration, but not in both. Shell then confirmed DEG’s nomination for the offshore arbitration. The other lenders nominated an English KC as their replacement arbitrator in the onshore arbitration. At this time, Freshfields took over as counsel in another unrelated arbitration where the party whom Freshfields now represented had previously nominated DEG as arbitrator. This circumstance was not disclosed in the arbitrations with Aiteo.
Once the tribunal in the offshore arbitration was fully constituted in August 2021, Aiteo then made a jurisdictional challenge. On 15 March 2022, the offshore tribunal (with DEG) dismissed that challenge by a partial award. Shortly before the offshore tribunal issued the jurisdictional award, in February and March 2022, Freshfields instructed DEG to produce an expert report on English law for the purposes of foreign proceedings. This engagement was not disclosed. In April 2022, DEG then did make a further disclosure to the ICC and Aiteo, namely that she had been appointed as presiding arbitrator by the two other party-appointed arbitrators in an arbitration in which Freshfields acted for one of the parties.
The offshore tribunal (with DEG) then made two further awards: one relating to the costs of Aiteo’s unsuccessful jurisdictional challenge, and importantly another award, consolidating the offshore and onshore arbitrations. That consolidation award was issued in August 2022. It was made by the offshore tribunal pursuant to specific provisions in arbitration clauses in Offshore and Onshore Facility Agreements.
Aiteo’s initial challenges and further disclosures by the arbitrator
Aiteo promptly challenged the offshore jurisdiction award and the consolidation award in the Commercial Court, under Section 67 of the Arbitration Act 1996. Aiteo argued that since the offshore tribunal had no jurisdiction over the offshore arbitration in the first place (and the offshore jurisdiction award was wrong), the tribunal also had no power to order that the offshore and onshore arbitrations should be consolidated. It is well-established under English law that arbitral tribunals cannot finally decide on its own jurisdiction, and that a jurisdictional challenge under Section 67 of the 1996 Act lies and proceeds by way of a full re-hearing of the issue, as opposed to a review of a jurisdictional ruling by the arbitral tribunal. On 17 November 2022, Foxton J dismissed Aiteo’s jurisdictional challenges (Aiteo v Shell [2022] EWHC 2912 (Comm)). The arbitral proceedings continued. Aiteo had not, however, exhausted its procedural challenges, and contested the jurisdiction of the (post-consolidation) tribunal (with DEG) over the onshore arbitration. In August 2023, the tribunal issued a further jurisdictional award, confirming that it did have jurisdiction over all the disputes under the onshore facility agreements, and confirming that the onshore arbitration had been properly consolidated with the offshore arbitration.
In November 2023, DEG made a further disclosure to the ICC and Aiteo: she had been instructed by Freshfields to provide another expert opinion on English law for the use by a different client in potential foreign proceedings. This prompted Aiteo’s counsel in the arbitration to request further information about the full extent of DEG’s relationship with Freshfields. In early December 2023, DEG provided a response, from which Aiteo learnt that:
The most recent expert instruction disclosed in November 2023 related to work that DEG had already carried out for Freshfields over the period from 17 to 25 October 2023.
DEG had previously been instructed by Freshfields to provide an expert declaration, with that work having been done over 25 February to 21 March 2022. DEG wrote that it “… did not cross my mind at the time to disclose this retainer. If I should have done so, I can only apologise”.
DEG had given expert advice in conference on English law to a Freshfields client in June and July 2020, prior to her appointment to the tribunal in the Aiteo proceedings. She had disclosed this to Freshfields and had asked her clerk to include this as a disclosure in her ICC Arbitrator’s Statement, but the clerk had omitted to do so an error for which DEG accepted responsibility in her December 2023 email.
In an unrelated arbitration, Freshfields had replaced the law firm who had previously appointed DEG as that party’s arbitrator.
The ICC Court removes the arbitrator
In mid-December 2013, Aiteo challenged DEG pursuant to Article 14(1) of the ICC Rules, arguing that from the perspective of an informed third party, there were justifiable doubts as to DEG’s impartiality in light of the information she had (belatedly) disclosed in the December 2023 email. The other two arbitrators and the remaining parties commented on this challenge, with the tribunal members confirmed that they had not seen any evidence of a lack of independence or impartiality on the part of DEG. DEG also provided her view, that the matters referred to did not give rise to apparent bias: however, it is accepted that the arbitrator’s own view of their independence does not add anything of substance. Everyone in the ICC challenge proceeded on the assumption that the English law test, or an objective test that was substantially the same, applied. On 17 January 2024, the ICC Court considered the matter at one of its closed sessions and upheld the challenge. The outcome was communicated to the parties by the ICC Secretariat. No reasons were given, since the parties had not asked for a reasoned decision beforehand: the ICC’s official guidance expressly requires this.
Aiteo challenges the awards as made by a compromised tribunal
On 31 January 2024, Aiteo then challenged the tribunal’s awards under Section 68 of the Arbitration Act 1996, on the basis that there had been a “… serious irregularity affecting the tribunal, the proceedings or the award.” Section 68(2) describes what may amount to a serious irregularity and specifically refers to “(a) failure by the tribunal to comply with section 33 (general duty of tribunal)”. That general duty of any arbitral tribunal extends to “… act[ing] fairly and impartially as between the parties.” Aiteo argued that apparent (not actual) bias on the part of DEG, as found to have existed by the ICC Court, amounted to a “serious irregularity”, and caused Aiteo to suffer “substantial injustice” as is specifically required by Section 68: Aiteo said it was deprived of its fundamental right to argue, and have its case determined by, a tribunal that complied with its Section 33 duty of impartiality.
The matter came before Jacobs J. The parties each made submissions on why the application of the Halliburton test should lead to their desired result. Aiteo highlighted passages in Halliburton in which the Supreme Court had said that the private nature of arbitration, with limited oversight, “puts a premium of frank disclosure”. Aiteo further submitted that DEG had failed to provide disclosure as required by the ICC Rules, which applied contractually, but that she had also breached the common law duty of disclosure that existed independently under English law. Shell and the other lender’s main point was that Aiteo had simply not explained why or how all the professional engagements that Aiteo relied on would trouble the fair-minded observer, or give rise to apparent bias. Shell argued it was not axiomatic that one should immediately proceed from a particular number of relational contacts between arbitrator and counsel to a conclusion of apparent bias. There had to be something more warranting concern by the fair-minded observer. In that regard, Shell pointed to DEG’s standing, reputation and her very considerable experience as a member of the judiciary first and then as arbitrator, being accustomed to dealing impartially with all matters that she was asked to decide.
Prior appointments and nominations as arbitrator
Jacobs J began his analysis by considering the provisions of the ICC Rules (2017), which expressly required DEG to disclose in the ICC Arbitrator’s Statement “… any facts or circumstances which might be of such a nature as to call into question the arbitrator’s independence in the eyes of the parties, as well as any circumstances that could give rise to reasonable doubts as to the arbitrator’s impartiality”, and also to disclose immediately any facts or circumstances of a similar nature that may arise during the arbitration. Starting with the position in December 2020 when DEG was appointed, Jacobs J held that her two prior nominations as party-appointed arbitrator by Freshfields (which had been disclosed) were clearly legally disclosable. He arrived at that conclusion by two different routes. The first was by considering the IBA Guidelines on Conflicts of Interest, which include a prior appointment by the same firm of solicitors on more than three occasions on the so-called `orange’ list (suggesting disclosure may be warranted). Here, including the two recent prior nominations, and the two present nominations in the Aiteo arbitrations, DEG and Freshfields had reached a problematic number (four).
The second route was by a close review of the ICC’s own documentation. The ICC Arbitrator Statement advises arbitrators to consult the ICC’s Note to Parties and Arbitral Tribunals on the Conduct of the Arbitration (the “ICC Note”) when preparing their disclosure. The ICC Note includes a checklist of “potentially relevant circumstances”, which includes: “The arbitrator or prospective arbitrator has in the past been appointed as arbitrator by one of the parties or one of its affiliates, or by counsel to one of the parties or the counsel’s law firm.” Jacobs J concluded that, while it did not follow that an arbitrator had to disclose all prior appointments by the law firm in question, two relatively recent such appointments were clearly disclosable. Those two arbitrator appointments had of course been disclosed, but they still formed part of the factual background that the fair-minded observer would take into account. DEG’s disclosure did not mean that they could be entirely ignored.
Expert and advisory engagements
The Judge then turned to DEG’s first expert engagement by Freshfields in June and July 2020. He noted that DEG had been asked to advise on English law (there is some doubt as to precisely how this engagement may really have differed from simply asking DEG as counsel to advise on the law), and that such an advisory engagement “… gives rise to a closer and different relationship to that which exists between arbitrator and the firm of solicitors which has appointed him or her. Thus in contrast to an arbitral appointment, an advisory engagement for a client requires the advisor, whether barrister or retired judge, to consider the client’s best interests and to advise and assist accordingly. It may also involve, as here, one or more conferences with the client. It is equivalent or at least very similar to a relationship, to use the words of the IBA Guidelines, of “cocounsel” between the barrister or retired judge and the law firm which has instructed him or her.”
Shell and the other lenders argued that this engagement did not require to be disclosed, relying in particular on the decision of Flaux J in A v B [2011] EWHC 2345. In that case, a party in the arbitration in question was represented by Dewey & Leboeuf. The arbitrator was a QC who had previously been instructed as counsel by Dewey in another case. That case was however the subject of a stay, and so no disclosure was made initially. When the case reactivated, the arbitrator did disclose his engagement, but this was late in the day after the hearing, and shortly before the award was to be issued. Flaux J rejected the resulting challenge. Jacobs J noted that the law had moved on since that decision in 2011. Certain aspects of the reasoning in A v B required reconsideration in light of Halliburton. Flaux J had pointed to how the legal profession in England and Wales operates, with barristers traditionally being instructed by a number of law firms to appear in court and arbitration. The Supreme Court had, however, specifically emphasised the differences between litigation and arbitration and placed a “premium on frank disclosure”. It was also no longer right to suggest, as Flaux J had done, that “disclosure and apparent bias are quite separate questions”. The Supreme Court in Halliburton has now explained that non-disclosure will inevitably colour the thinking of the fair-minded observer (even if it is inadvertent). A failure to disclose is a factor in deciding whether there are justifiable doubts as to an arbitrator’s impartiality: it may “in certain circumstances amount to apparent bias”. It was not possible to say that, because the arbitrator was a very experienced professional or a retired judge such as DEG, the possibility of unconscious bias in favour of the law firm instructing them could simply be ruled out. The fair-minded observer would ask whether the judgment of even such an experienced individual was subconsciously biased in favour of the law firm giving them work. Considering that there were already two prior and two contemporaneous nominations of DEG by Freshfields, Jacobs J concluded that the expert engagement of June and July 2020 was certainly disclosable. This had also been DEG’s own view but her clerk had inadvertently omitted to make the disclosure.
Moving on chronologically, Jacobs J held that the fact of Freshfields in June 2021 taking over as counsel in an unrelated arbitration for a party whose previous legal representatives had nominated DEG as their arbitrator would be of no significance to the fair-minded observer. However, the position was rather different with regard to DEG’s engagement to provide an expert declaration on English law for a Freshfields client engaged in foreign proceedings. This instruction, which lasted from 25 February to 21 March 2022, was only disclosed by DEG in the email of 9 December 2023. While the provision of an expert report presupposed the independence of the person doing so (and in this case, an English law expert would owe duties to the foreign court receiving the evidence), Jacobs J pointed to how such an engagement would play out in reality. He held that the relationship between an expert and the instructing law firm would not be vastly different to that between solicitor and counsel, and the expert would be advising the client, together with the instructing law firm, on the position under English law. There would also be discussions on strategy between expert, law firm and client. Those discussions could be expected to cover what points could properly be presented to the foreign court, and drafts of the expert report or declaration would be reviewed and commented upon by all. Again, Jacobs J concluded that an expert appointment had elements of an advisory engagement, and that “… there are aspects of the relationship which are similar to a “co-counsel” relationship, albeit that the expert will consider that his or her overriding duty is to the foreign court.” The same points also applied to the October 2023 expert engagement, disclosed by DEG in the December 2023 email. On that occasion, Freshfields had instructed her to “… expert opinion on English law in the context of potential foreign insolvency proceedings.” Jacobs J held that this was a disclosable engagement or relational contract. There were, therefore, three failures by DEG to make timely disclosure (the June July 2020 advisory engagement and the two expert instructions of February March 2022 and October 2023).
Res judicata?
In its submissions to the Commercial Court, Aiteo also relied on the decision of the ICC Court to remove DEG from the tribunal, arguing that this meant the matter was res judicata and had been conclusively decided in Aiteo’s favour, without the Commercial Court being able to reach a different view. Jacobs J rejected this argument. The learned judge noted that the ICC Court was not a `court’ in the ordinary sense of the word, as it does not convene hearings or hear evidence and arguments before handing down a judgment. Instead, the ICC Court is part of the ICC as an arbitral institution. By its statutes (Appendix 1 to the ICC Rules), the ICC Court is tasked with ensuring the proper application of the ICC Rules of Arbitration: “As an autonomous body, it carries out these functions in complete independence from the ICC and its organs.” The members of the ICC Court are drawn from the ICC’s national committees, with members and alternative members from around 118 countries worldwide. The ICC Court has one President and 17 Vice-Presidents. The composition of the ICC Court changes from (closed) session to session, and its work is generally confidential. As noted, the ICC Court will, when deciding an arbitrator challenge, only provide reasons if a party has expressly asked for this in advance. If not, the ICC Secretariat will merely communicate the outcome to the parties in essentially a one-line letter.
The key question was whether the ICC Court was, for the purposes of English law, a `judicial tribunal’ whose decisions would bring the res judicata doctrine into play. In Unite the Union v McFadden [2021] EWCA Civ 199, the Court of Appeal found that the decisions of all arbitrators or other tribunals who have been invested with the power to hear and determine disputes by the consent of the parties were binding until set aside, so that the doctrine applied. The dividing line was between a tribunal “… which is independent of the parties and is invested by law with the power to determine an issue which establishes the existence of a legal right, and other bodies, which are not”. That determination of an issue or legal right must, however, relate to the merits as opposed to being merely a procedural decision. A decision on the merits requires the tribunal to establish the facts and applicable legal principles, and then to apply those principles to those facts, thereby adjudicating on a particular cause of action. That, Jacobs J held, is not what the ICC Court does: it does not determine any legal rights, but only decides procedural matters arising under the ICC Rules. It was also highly material that a challenge of an arbitrator under Section 24 of the 1996 Act can only be brought once any arbitral institution has addressed the matter (here, the ICC Court): see Section 24(2). If such a decision by an arbitral institution meant the issue became res judicata, then there would be very little point in the statutory power to remove an arbitrator under the 1996 Act.
However, even though the decision of the ICC Court did not render the issue res judicata, the fact that the challenge had been upheld under the ICC Rules was a factor to be taken into account:
“… the fair-minded and informed observer would accord considerable respect to the decision of the ICC Court, and realistically could not avoid being struck by the fact that this was one of the rare challenges that succeeded. However, the observer would recognise that he or she should make up his or her own mind on the basis of the underlying facts, and that it would be wrong to reach a conclusion simply by reference to what the ICC Court had decided. The decision of the ICC Court could serve as a useful cross-check on the observer’s own conclusions based on the underlying facts, but ultimately the observer needed to make up his or her own mind.”
That was consistent with how the English Courts had treated the decisions of other arbitral institutions prior to Halliburton such as, for instance, a determination made by a three-member division of the LCIA Court in P v Q [2017] EWHC 194 (Comm).
What would the fair-minded observer think?
Against that background, Jacobs J held that Aiteo’s submissions were well-founded and that there was apparent bias. The fair-minded observer would consider all relational contacts between DEG and Freshfields, so both the disclosed prior and current arbitral appointments and the undisclosed (or belatedly disclosed) advisory and expert engagements. The picture presented was that Freshfields had appointed or engaged DEG six or seven times in a period of five years: three successful arbitral appointments (including the offshore arbitration), one unsuccessful arbitral nomination (the onshore arbitration) and three expert / advisory engagements. If all that information had been promptly disclosed, the fair-minded observer might have concluded that DEG had stayed on the right side of the line. However, that had not been the case. Jacobs J noted that the non-disclosures had occurred relatively soon after the Supreme Court decision in Halliburton, which highlighted the need for arbitrators to be full and frank with the parties. He rejected Shell’s submission that Aiteo had failed to explain why the fair-minded observer would be concerned by these cumulative appointments and engagements, or their non-disclosure. The question was not whether DEG was biased in favour of Freshfields’ clients. Rather, the issue was the appearance of bias, and unconscious bias. Further, the Supreme Court had already explained why cumulative appointments or relational contacts between arbitrators and counsel were self-evidently relevant and required no further explanation as to why they gave rise to concerns:
“… a judge is the holder of a public office, is funded by general taxation and has a high degree of security of tenure of office and therefore of remuneration. An arbitrator is nominated to act by one or both of the parties to the arbitration either directly or by submitting names to the appointing body, whether an institution or the court, for appointment. The arbitrator is remunerated by the parties to the arbitration in accordance with the terms set out in the reference, and often is ultimately funded by the losing party. He or she is appointed only for the particular reference and, if arbitral work is a significant part of the arbitrator’s professional practice, he or she has a financial interest in obtaining further appointments as arbitrator. Nomination as an arbitrator gives the arbitrator a financial benefit. There are many practitioners whose livelihood depends to a significant degree on acting as arbitrators. This may give an arbitrator an interest in avoiding action which would alienate the parties to an arbitration, for example by assertive case management against the wishes of the legal teams who are presenting their clients’ cases. It also may give those legal teams an incentive to be more assertive of their side’s interests in the conduct of the arbitration than might be the case in a commercial court.”
DEG’s reputation and great distinction as a judge were a relevant factor, but not a determinative one: the Supreme Court in Halliburton indicated that reputation and experience of arbitrators would only carry limited weight, and Jacobs J agreed with that. The fair-minded observer would also take note of DEG’s prompt and courteous responses to Aiteo’s questions, and of the assurances given by the other two arbitrators that they had seen no sign of bias, but would attach more importance to the objective facts specifically, the number of appointments and the failures to disclosure. In reaching his ultimate conclusion, Jacobs J then pointed to the ICC Court’s decision to remove DEG (a rare occurrence) as tipping the balance:
“… the observer would consider that there was a real possibility of unconscious bias, notwithstanding that there were some factors which would favour a different conclusion. The observer would feel comfortable in reaching that conclusion in circumstances where the ICC Court had removed DEG as arbitrator. Any possible doubt as to the answer to the question for the observer would be resolved by the consideration of the decision of the ICC Court, which would strike the observer as rational and well-founded.”
Substantial injustice
Aiteo argued that a finding of apparent bias invariably meant that there was “substantial injustice” for the purposes of Section 68(2) of the 1996 Act, such that all the awards made by the tribunal featuring DEG were subject to challenge. Jacobs J disagreed. He followed the decision in Africa Sourcing Cameroun Ltd v LMBS [2023] EWHC 150 (Comm), concluding that in the case of apparent bias, substantial injustice will normally be inferred or anticipated, but that is not inevitably the case. Whilst it was inherently likely or in the nature of things that such apparent bias would support an inference of substantial injustices, that inference could be rebutted. Shell had argued (in the event that it lost on apparent bias) that a number of matters served to provide such a rebuttal in the present case. The first point was that the offshore jurisdiction award had already been the subject of an unsuccessful jurisdictional challenge under Section 67 of the 1996 Act. Foxton J had carried out his own independent investigation, concluding (for his own reasons) that the tribunal did have jurisdiction. That decision by the Court had a curative effect on the award itself, and also on the costs award against Aiteo as the unsuccessful party. Aiteo’s Section 68 challenge therefore failed. Turning to the consolidation award, Jacobs J also concluded that there was no substantial injustice, because of an unusual feature in the consolidation provision that formed part of the arbitration agreement, which provided that the “first arbitrator” to be appointed under the Offshore Facility Agreement had the power to decide whether any related arbitrations should be consolidated because of a close connection between the disputes. In the arbitration, the issue had been whether the “first arbitrator” meant DEG (she was literally the first tribunal member to be nominated) or the first tribunal of three duly constituted. The tribunal reached the conclusion that the latter view was correct, but because of the nature of the arguments, each tribunal member then set out their own individual views on consolidation in the award. All three were in favour. This led Jacobs J to conclude that there was no substantial injustice, since:
“… each member of the Tribunal did in fact consider the question … , prior to discussing the question and reaching a collegiate view. There is no reason for the court to doubt that this is indeed what happened. In these unusual circumstances, it is clear that the Tribunal did in fact reach its conclusion, that there should be consolidation, independently of the individual views of DEG. In those circumstances, I consider that Aiteo has failed to show substantial injustice in relation to the Consolidation Award. … it is necessary for the court to tread carefully when considering statements by the tribunal members, unaffected by apparent bias, that they would have reached the same conclusion irrespective of the involvement of the member affected by apparent bias. Here, however, the statement was not made in response to any challenge, but rather in “real time” solely because of the nature of the issue being argued.”
However, there were no factors warranting the rebuttal of the inference of substantial injustice relating to the onshore jurisdiction award. Jacobs J found that whilst Shell’s jurisdictional arguments which the tribunal had upheld were powerful, it could not be said that Aiteo had had nothing in response, or that Aiteo’s arguments were so weak that they were unworthy of consideration by a tribunal unaffected by apparent bias. The learned judge also granted Aiteo an extension of time for the Section 68 application, since more than 28 days had passed since the award that was being challenged had been made. He held that Aiteo had only learnt of the circumstances that allowed it to argue the challenge long after the 28 days had already elapsed, and (in summary) it would be unfair in the round to prevent Aiteo from challenging an award that had been made by a tribunal affected by apparent bias. Jacobs J remitted the onshore jurisdiction award to the (reconstituted) tribunal for reconsideration, in accordance with Article 15(4) of the ICC Rules: the ICC Court has discretion to decide on the process for nominating the replacement arbitration, and the newly completed tribunal is to then decide whether and to what extent any part of the prior proceedings are to be repeated.
Conclusion
Arbitrator challenges rarely succeed, and it is unusual to have two judgments of the English Courts removing arbitrators follow in relatively quick succession. The arbitrators who fell foul of the objective test for apparent bias were at different ends of the spectrum: one was an experienced and senior industry arbitrator appointed for the first time, and the other a very senior retired member of the English judiciary. The first arbitrator failed to pay heed to the appearance he created when making unguarded comments about the evidence at a procedural hearing.
The second arbitrator may perhaps have underestimated the impact that accepting repeat instructions from counsel for one of the parties would have as a matter of objective assessment. The presumed independence of counsel does not trump the appearance of bias. Care must be taken to disclose all such matters up front, to avoid creating the wrong impressions.