Full disclosure – what do we really need to know?

By Rebecca Ardagh, Senior Associate

The appearance of justice

A maxim of the judicial world is that justice not only be done, but that it be seen to be done. The importance of faith and trust in the system cannot be overstated when it comes to its continued successful operation. 

With international arbitration, the dispensation of justice relies on the decision making of the arbitral tribunal, and that these decisions are based solely on the merits of the case without deference to any other influences. As above; it is just as important that this be the appearance given by the tribunal as it is that it is the reality behind the scenes.  

Independence and/or impartiality

Accordingly, the international community holds arbitrators to certain standards of impartiality and independence through domestic arbitration law, institutional rules, and even various soft-law mechanisms (such as the IBA Guidelines on Conflicts of Interest in International Arbitration 2014 and UNCITRAL Model Law on International Commercial Arbitration 1985). Though there are some differences between these standards across the various sources,1 the approach is mostly consistent and is akin to that set out in the IBA Guidelines, which can be summarised as follows: 

1. The arbitrator has an obligation to be independent and impartial. 
2. This obligation continues throughout the proceeding and until the final award has been rendered. 
3. The test of independence or impartiality is an objective one, considering whether a reasonable third party with full knowledge of the facts and circumstances would consider it likely the arbitrator may decide a case based on factors other than its merits. This test is founded on the appearance of bias or assessment of “apparent” bias, rather than identifying actual bias. 

An arbitrator can be challenged on his or her lack of independence and/or impartiality either through the institutional body governing the arbitration or under the applicable domestic law. If successful, this will result in the arbitrator being removed and replaced or, in the event the challenge is taken after the award has been rendered, the award being nullified. Clearly, a challenge to an arbitrator’s independence and impartiality can be incredibly disruptive and costly depending on when it occurs. Ideally a challenge will be made at the beginning of proceedings to ensure minimal cost and delay (whether successful or otherwise). 

Disclosure

In order to know whether there are grounds for a challenge at the beginning of a proceeding or when an arbitrator is first appointed, the parties need to be aware of any facts and circumstances that may give rise to a justifiable doubt as to the arbitrator’s independence or impartiality. This information is likely held solely by the arbitrator and therefore, in order to best ensure the obligations of independence and impartiality are maintained, there is generally considered to be an obligation of disclosure held by arbitrators throughout a proceeding. 

The interdependence of the obligations of impartiality and disclosure was somewhat confirmed by the English Supreme Court in Halliburton Company v Chubb Bermuda Insurance Ltd.2 There is currently no express obligation of disclosure under the English Arbitration Act (though this is likely to change through the ongoing reform3, however, the Supreme Court found that there was necessarily a legal obligation of disclosure implied under section 33 of the Act, which requires arbitrators to remain impartial; in order to act impartially, one must disclose any matters that have the potential to impact this ability. In other jurisdictions, institutional rules, and guidelines, there is an express obligation on an arbitrator to disclose such matters. 

There is a natural tension between the idealistic (though, of course, necessary) expectations of impartiality and independence, and the realities of international arbitration in practice; arbitrators do not exist in a vacuum and, particularly depending on the industry, are likely to have some sort of experience with counsel or parties and existing professional views on legal issues in consideration, and, in any event, are usually appointed directly by one of the disputing parties. None of these influences necessarily render an arbitrator partial, however, there can be instances where a pre-existing relationship with counsel or a party is one of influence, where an arbitrator has a direct financial interest in one of the parties or even the outcome of the arbitration, or when an arbitrator’s existing view on an issue of law is so entrenched that he or she has a pre-determined decision rather than considering the merits of the case. 

Somewhere on this continuum, facts and matters concerning these potential influences become relevant rather than innocent; the question is, how do we know and what should be disclosed? 

In order to ensure the parties have the best information available as early as possible, arbitrators are generally encouraged to resolve this question in favour of disclosure. In fact, most laws, rules, and guidelines that give guidance as to the test to be applied for disclosure prefer a subjective approach rather than an objective one, such as the IBA Guidelines which confirm that the arbitrator’s duty of disclosure “rests on the principle that the parties have an interest in being fully informed of any facts or circumstances that may be relevant in their view” [Emphasis added]. 

Too much of a good thing

The concern with this approach is that disclosure of every circumstance may, in some cases, be extensive and yet still not contain any fact or matter that would necessarily give rise to a justifiable doubt as to the arbitrator’s impartiality or independence. Ultimately, it cannot be up to an arbitrator to consider and form a view as to the probative value of his or her own information. In order to ensure the parties can accurately consider whether a challenge is justifiable as early as possible, there must be a preference for disclosure. However, parties must be realistic about the nature of the information that is disclosed and the facts or matters the may evidence; not all information will be grounds for a successful challenge (in fact, it will more frequently not be the case). 

This assessment is inherently fact specific and approached on a case-by-case basis. For example, the recent Paris Court of Appeal decision of Douala International Terminal (DIT) v Port Autonome de Douala4 considered an undisclosed relationship between an arbitrator and the late lawyer for one of the parties. This relationship became apparent after the final award had been rendered when the arbitrator wrote a eulogy for the lawyer. In this eulogy, the arbitrator confirmed that the two would meet regularly, that he loved and admired the lawyer, and that the lawyer would often advise him in his decision-making process. In this case, it was not the existence of the un-disclosed relationship that met the threshold to void the award; it was the fact that the arbitrator expressly confirmed the relationship was one that held particular influence over him and his decisions. The existence of a relationship with counsel alone, particularly in smaller legal industries, has been frequently held not to be indicative of apparent bias (see, for example, Grupo Unidos por el Canal, S.A. v Autoridad del Canal de Panamá5 in the United States). 

The wide-ranging nature of the duty may provide some comfort to arbitrators in alleviating the need to make an assessment as to pertinence him or herself, though it risks becoming overly burdensome. In France, this obligation is fettered by a “notoriety” exception, where there is no need to disclose information that is a matter of public knowledge. Similar exceptions exist in other jurisdictions, such as Egypt. This, to some extent, shifts some of that burden to the parties to carry out their own due diligence as to potential arbitrators at the time of appointment and familiarise themselves with the information available at least in the public arena. 

The extent to which this exception evolves into a positive duty on the parties remains to be seen; it is not drafted with the intention of requiring the parties to undertake significant research, however, the courts have been open to invoking the exception on a relatively liberal basis. For example, in Delta Dragon v BYD,6 the Paris Court of Appeal held that an arbitrator’s connection to the automotive industry (being a member of the advisory board for a company that was the parent company to a strategic partner to one of the parties) would have been evident by typing the arbitrator’s name with the German word “automobil” in quotations into an internet search engine, and therefore fell within the notoriety exception. The court went so far as to state that parties are required to demonstrate a “modicum of curiosity” and conduct their own research (in this case, even in other languages) in order to identify such possible objections and raise them in a timely manner, otherwise they risk waiving this right after a reasonable time has passed from the original appointment.  

The notoriety exception may go some way to tempering the potential landmine of social media and disclosure obligations, which gives rise to such questions as whether “contacts” warrant disclosure or only offline relationships, and whether posts or views shared or expressed on personal platforms are captured, or only those from professional pages. We should be cautious, however, of placing too high of a burden on parties to uncover such information through onerous research and investigation.  

Further clarity

There is currently a large amount of reform going on in this space; there are current draft updates to the IBA Guidelines, the English Arbitration Act 1996, and the Draft Code of Conduct for Adjudicators in Investor-State Dispute Settlements. All these proposed updates provide further recognition of the obligation of disclosure and more refined attempts at guidance as to the application of this obligation. 

In particular, the IBA Guidelines provide guidance by way of practical examples of facts and matters that warrant disclosure and may give the appearance of bias in an arbitration. These are colour coded into a traffic light system ranging from the Red list, which includes examples of situations in which an arbitrator must be removed or expressly require the consent of the parties to continue with the appointment, to the Green list which includes examples that are so innocuous they do not require disclosure. While further updates to this list are being made to provide more clarity, there are still those who criticise the utility of these guidelines practically. In particular, there is often criticism of the middling Orange list, which includes situations that must be disclosed but do not necessarily imply a conflict of interest. Matters that fall within or outside the Orange list necessarily require examination on a case-by-case basis and so it has been argued that the mere existence of this list creates more confusion than clarity.  

Finally, there are also a number of cases currently awaiting decisions at the time of publication that may provide more clarity on the scope of disclosure obligations. Aroma Franchise Company v Aroma Espresso Bar, which was heard by the Ontario Court of Appeal in December 2023 and is currently awaiting decision, concerns the impact of non-disclosure of multiple appointments by common counsel. Occidental Petroleum v Andes Petroleum is currently awaiting leave from the US Supreme Court to appeal following a petition for writ of certiorari filed in November 2023. This relates to an allegation of apparent bias based on the failure of an arbitrator to disclose an appointment to an unrelated tribunal as co-arbitrator with a member of the counsel team for one of the parties. The Second Circuit concluded “a reasonable person would have to conclude that an arbitrator was partial” where Occidental Petroleum argues that Supreme Court precedent confirms that an appearance of bias is sufficient to demonstrate “evident partiality” under the Federal Arbitration Act. The appearance of bias would clearly be a lower bar than that applied by the Second Circuit. 

Conclusion

These pending reforms and decisions signal a big year ahead for the disclosure obligation, however, expectations must be tempered; the appearance of bias is so fact and circumstance specific that, though it would be welcome, it is difficult to see how guidance could be any more detailed or prescriptive and still have any practical application for parties and arbitrators.  

In order to ensure the appearance of an impartial and independent decision maker, there must be complete transparency. It is, however, up to the parties to approach this transparency reasonably and not take the mere action of disclosure by an arbitrator as evidence of partiality; the bar applied to warrant disclosure is much lower than that applied to an assessment of partiality, and expectations should be sensibly set, and challenges taken cautiously. Arbitrators should look to err on the side of disclosure whenever a question arises. Parties should also carry out brief exercises of due diligence to ascertain at least what is already in the public domain (including web searches and a review of platforms such as LinkedIn). 

Previous article | Next article

• 1. For example, under the English Arbitration Act 1996, the US Federal Arbitration Act and the Swedish Arbitration Act 1999, arbitrators are required to be impartial but there is no express obligation of independence.
• 2. [2020] UKSC 48. See Dispatch, Issue 246.
• 3. See “So, how does the UK plan to remain a world leader in international arbitration? By doing very little”.
• 4. Paris Court of Appeal, 10 Jan 2023, No. 20/18330.
• 5. No. 21-14408 (11th Cir. 2023).
• 6. Paris Court of Appeal, 25 May 2021, No. 18/20625.