Much has been said about the duty of international arbitrators to disclose potential conflicts of interest with parties and counsel, but relationships with experts is an area that still needs clarification. A recent decision to annul a final award by an ICSID ad hoc Committee (the “Committee”) in Eiser Infrastructure Limited and Energia Solar Luxembourg S.à.r.l. (“Eiser” or “Claimants”) v. Kingdom of Spain (“Spain” or “Respondent”), ICSID Case No. ARB/13/36 (“Eiser v. Spain”), based on the failure of an arbitrator to disclose extensive contacts with an expert in the matter, provides important guidance on this issue.[i]
On June, 11 2020, the Committee rendered its decision on Respondent Spain’s application to annul the award in Eiser v. Spain (the “Annulment Decision”) based on arbitrator, Dr. Stanimir A. Alexandrov’s failure to disclose a prior and continuing relationship with expert Carlos Lapuerta of the Brattle Group. Claimants had appointed both expert Lapuerta and arbitrator Alexandrov.
The annulment application was brought pursuant to article 52(1)(a) of the ICSID Convention, which lists the grounds on which a party “may request annulment of the award.” Two of the grounds relevant to the disclosure issue required the Respondent to show “that the [arbitral] Tribunal was not properly constituted,” and “that there has been a serious departure from a fundamental rule of procedure.” The Committee accepted that both grounds could be proven if it was shown that arbitrator Alexandrov’s failure to disclose his ties to Brattle and expert Lapuerta raised the appearance of a “lack of impartiality or independence” that would be obvious to a third party, and material to the award.[ii]
The Committee considered various sources of guidance on conflicts of interest between arbitrators and experts, but did not find any published standards that were directly applicable. The well-known IBA Guidelines on Conflicts of Interest in International Arbitration (the “IBA Guidelines on Conflicts”), which memorialize the most widely recognized principles in this area, do not speak to relationships between arbitrators and experts, except for in very limited circumstances.[iii] The Committee noted this omission, but also considered that the IBA Guidelines are not exhaustive or binding, and thus took no instruction from their silence on this issue.[iv] Nevertheless, the Committee did find some support for its Decision in the IBA Guidelines general standard holding that any questionable contacts maintained by an Arbitrator should be resolved in favor of disclosure.[v]
A second source of potential guidance, the IBA Rules on the Taking of Evidence in International Arbitration (the “IBA Rules of Evidence”), do state in article 5(1) that expert reports should include disclosures of relevant connections between experts and arbitrators. Nevertheless, this is a duty incumbent on experts, and does not directly implicate a disclosure obligation on the part of arbitrators.
In addition to these sources of guidance, the Committee also considered general principles of law and prior arbitral precedents, some of which were relied on.
Annulment Committee’s Decision
It is not uncommon in the relatively small world of international arbitration for experts to work with legal counsel on repeat engagements. Here, arbitrator Alexandrov’s relationship with Brattle Group, and specifically Lapuerta, had developed over 15 years during which on at least eight separate occasions he acted as counsel in matters in which his clients also retained Brattle.[vi]
The previous cooperation between Alexandrov and Lapuerta became an issue once the former began to serve as an arbitrator in matters where Brattle was present. Challenges to Alexandrov’s neutrality were raised on this basis on two different occasions, one being a separate ICSID arbitration and the other an UNCITRAL Rules arbitration administered by the Permanent Court of Arbitration (“PCA”). In the first instance, arbitrator Alexandrov resigned after his fellow arbitrators were unable to agree on whether a conflict was present; and, in the second, the challenge was rebuffed after an opinion from the Secretary General of the PCA found that a conflict of interest had not arisen as a result of Brattle’s appearance in the matter.[vii]
In its Decision, the Committee distinguished both of these cases on the fact that Alexandrov had not worked with Brattle as counsel while sitting as arbitrator. In contrast, the Committee noted that here, while the underlying arbitration in Eiser v. Spain was pending, Brattle was retained on at least three separate cases, (two that involved expert Lapuerta) for clients which Alexandrov also acted as counsel.[viii] In the Committee’s view, these parallel situations gave rise to an intense working relationship on behalf of a shared client, placing Alexandrov in the position to advocate for Lapuerta’s (or Brattle’s) analysis in one capacity while sitting in adjudication of it in the other.
The Committee found that Alexandrov’s concurrent relationship and extensive history of working with Lapuerta and Brattle, created a “manifest appearance of bias” and should have been disclosed.[ix] Since Alexandrov had not disclosed this relationship [Nor had Lapuerta per IBA Rules art. 5.1(a).], the Committee ruled that the appearance of a “lack of impartiality or independence” tainted the award sufficient to justify annulling it on the grounds set forth in article 52(1)(a) of the ICSID Convention.
The Annulment Committee’s analysis in this case is instructive in regard to the murky area of conflicts of interest between arbitrators and experts. The following points summarize some of the key takeaways from their Decision:
(a) The argument that party-appointed experts are hired by clients to give a neutral opinion, and thus do not have a relationship with counsel that could raise a later conflict, was flatly rejected by the Committee due to the close working relationship that exists between the two roles in practice.[x]
(b) The Committee’s Decision turned on the fact that this situation involved an arbitrator sitting in adjudication of an expert’s work in one mater while simultaneously, as counsel, advocating for the same expert’s analysis in another. The concurrent nature of the two situations seemingly distinguished this case from one where an arbitrator’s contacts with an expert were exclusively in the past and no longer active – nevertheless, one cannot rule out the possibility that historical relationships should be disclosed if the number and nature of such contacts would warrant it;[xi]
(c) Although the obligation under IBA Rules of Evidence art. 5(1) requiring any connections with the arbitrator to be disclosed is incumbent on the expert, not the arbitrator, Lapuerta’s failure to divulge such information likely contributed to the Committee’s dim view of this situation.[xii] This serves as a reminder to experts to be circumspect in their disclosures to ensure they do not inadvertently create grounds for challenging a final award by omitting important information;
(d) The annulment of an award constitutes a drastic measure. That the Committee decided to do so in respect of a relationship between arbitrator and expert, even in the absence of express rules governing this situation, suggests that the development of standards around this issue is warranted.
[i] ICSID ad hoc Committees are convened to consider challenges to final awards issued within the ICSID arbitration system. Although unique to ICSID arbitration, the remit of an ad hoc Committee is often to apply principles ubiquitous to arbitration laws found throughout the world. Consequently, as these committees are generally composed of international jurists, the body of decisions by ICSID ad hoc Committees provides insight into widely held views on core due process principles and related laws.
[ii] Annulment Decision, para. 180.
[iii] For example, IBA Guideline on Conflicts arts. 3.4.3 (regarding close personal friendship between an arbitrator and an expert) and 3.4.4 (regarding enmity between an arbitrator and an expert).
[iv] Annulment Decision, para. 226.
[v] IBA Guidelines, General Standard 3(d).
[vi] Annulment Decision, para. 205(f).
[vii] Annulment Decision, para. 214. The Committee recalled the fact that in both prior incidents, the work done by Alexandrov as counsel with Brattle had ceased by the time a challenge was brought against him. Thus, it could not be said he was simultaneously sitting in review of Brattle’s expert testimony in one instance while working with them as counsel in another. Nevertheless, even here, the Committee noted that the opinion issued by the Secretary General of the PCA was instructive, providing the following:“ (…) it cannot be ruled out that the concurrent service of an arbitrator as counsel in another pending arbitration, in which the damages expert of the part that he represents has submitted a similar and “innovative” damage evaluation method may prevent an arbitrator from evaluating this method with an open mind.” Id.
[viii] Annulment Decision, para. 218.
[ix] Annulment Decision, para. 228: “In the Committee’s view, in this case, the duty to disclose was warranted due to the respective roles of a damages expert and counsel in an arbitration. It was warranted not only because of the existence of such a relationship but also by the extent of the past and present interactions, at issue. These taken together triggered Dr. Alexandrov’s obligation to disclose.”
[x] Annulment Decision, para. 227: “In the opinion of the Committee, nothing turns on the answer to the questions about who appoints the expert or what are the qualifications or expertise of such a damages expert. What is important is that damage experts work closely with counsel in preparation of a case. In the course of an arbitration there are multiple exchanges between them. They do not and cannot possibly maintain between them the kind of professional distance which is required to be maintained between a party, its counsel and its experts in a case, on the one hand and the member of the tribunal hearing the case, on the other.”
[xi] Annulment Decision, para. 217.
[xii] Annulment Decision, para(s). 247-248.