In a tribunal comprising three members, a unanimous arbitral award depicts the agreement of all members to the award. It has been argued that a single arbitrator’s bias in such a tribunal does not impact the integrity of the award as the outcome remains unaffected. Consequently, applications for setting aside arbitral awards on the grounds of independence and impartiality of such a single arbitrator are rejected. In this regard, while dealing with a challenge against one of the members of the tribunal for reasonable apprehension of bias, the Ontario Superior Court of Justice (“ONSC”) in Vento Motorcycles, Inc. v. United Mexican States, 2023 ONSC 5964 (“Vento I”) refused to set aside the award observing that reasonable apprehension of bias by a single arbitrator is a procedural error and unanimity infers that all three members shared the same view on the decision and reasoning. However, this presumption of a shared view against a single arbitrator’s bias not only seemed to be misplaced from an objective standard but also from the fundamental standard, as the parties intended for not only a fair award but also adjudication by an independent and impartial arbitral tribunal.
Recently, in February 2025, a three-judge bench of the Ontario Court of Appeal (“ONCA”) in Vento Motorcycles Inc v The United Mexican States, 2025 ONCA 82(“Vento II”) set aside the award and overturned the above reasoning of ONSC in Vento I. The decision is a welcome move and a significant addition to the jurisprudence surrounding the issue.
A prior publication on the Blog discussed the factual background and procedural posture of Vento II in particular. In furtherance of the same and building upon the brief discussion on Vento II here, this post provides a broader and international analysis of the issue of permeability of bias within an arbitral tribunal and the validity of such awards. Additionally, it provides insight into the interplay between objective standards of independence and impartiality, the influence of bias within the tribunal, and unanimity in arbitral awards. In particular, this post while dealing with judicial precedents in India, Germany, the United States (US), and the International Centre for Settlement of Investment Disputes (“ICSID”) arbitral award shall analyze the relevant standard for assessing the validity of such awards. Moreover, this post shall discuss relevance of party expectations and arbitral deliberations in context of such awards.
Unanimous arbitral awards under the allegation of bias: Is quantum of influence the correct standard?
Unanimous arbitral awards implicitly carry a stamp of consensus as compared to bias, which conveys a sense of favourability to one party in the proceedings. Accordingly, even if one of the tribunal members was biased, the award is valid, as the majority of the tribunal members agreed with the findings. This argument stems from the established practice and provisions of institutional rules such as the International Chamber of Commerce (ICC) in article 32(1), Stockholm Chamber of Commerce (SCC) in article 41(1), London Court of International Arbitration (LCIA) in article 26.5 which prescribes that the award may be decided by a majority. This argument for unanimous awards overextends the provisions, ignoring arbitrators’ influence on each other and its impact on the award. Prima facie, such influence seems a reasonable standard for assessing an award’s validity. As noted in the prior post, even though there was a reasonable apprehension of bias, the court in Vento I focused on the potential impact of the procedural error on the award while observing that it did not necessarily ‘taint’ the award and the entire panel.
Notably, Vento I was based on speculation on two fronts: (1) a unanimous award ensures non-influence by a biased arbitrator, and (2) the procedural error caused no real unfairness in the proceedings. As distinguished in Vento II, the Court in Vento I paradoxically relied on the Supreme Court’s distinct decision-making process while acknowledging that it does not represent the tribunal’s decision-making process. As discussed in a prior post, Vento I made several assumptions about the tribunal’s inner workings to assess the potential impact of the disclosure breach, which Vento II rejected.
Thus, as per Vento II, under Canadian law mere participation of the biased arbitrator mandates setting aside the decision regardless of its unanimity, thereby ousting any requirement of assessing the quantum of influence. It aligns with the common law approach of not affording discretion once reasonable apprehension of bias is established. Likewise, in India, dealing with similar circumstances, the Delhi High Court, in the decision of Lanco-Rani (JV) v. National Highways Authority of India O.M.P. 199/2008, observed that mere fact that one of the members of the tribunal was biased compromises fairness and vitiates the entire proceedings. These observations further align with the objective standard assessment of independence and impartiality. Similarly, the German Supreme Court, in the proceedings to set aside a domestic arbitral award in I ZB 23/14, dismissed the argument that unanimity ensures that the biased arbitrator had no causal effect on the award. Moreover, deviating from the strict objective assessment, the Court took into account the aspect of influence by stating that the arbitrator’s conduct during deliberations may influence the formation of opinion and voting behaviour of other arbitrators. Interestingly, the annulment committee in Eiser v. Spain ICSID Case No. ARB/13/36, (“Eiser”) took a slightly stricter approach by stating unanimity does not impede annulment and that each member is expected to have influenced the other two with their views and analysis during deliberations. The US Supreme Court in the decision of Commonwealth Coatings v. Continental Cas., 393 U.S. 145 (1968) (“Commonwealth Coatings”), took a split approach on the issue, the majority while considering ‘appearance of bias’ as an evident partiality set aside the award while the dissenting judges considering that the award was unanimous stated that only biased conduct towards the other party constitutes evident partiality. The dissenters rejected setting aside the award as there was no suggestion of partiality, bias, irregularity, or unfairness. Aligning with the dissenters, the 2020 Supreme Court of New York state in Carter v. Royal All. Assocs (2020 N.Y. Slip Op. 32086) held that the challenge of bias in a unanimous arbitral award was an afterthought, as the party failed to do so during proceedings.
Does single arbitrator bias ensue setting aside of unanimous arbitral award?
Although objectively, a unanimous arbitral award may be set aside in the event there exists a reasonable apprehension of bias against a single arbitrator, the following factors play a crucial role in understanding the interplay between single arbitrator bias and annulment of unanimous arbitral awards.
- Party expectations
As established in Vento II, parties not only expect the final award to be fair and just but also expect an independent and impartial tribunal to render it. Thus, even if the bias does not impact the outcome of the dispute, it still derogates from the parties’ expectations. Additionally, a reasonable apprehension of bias against a single arbitrator permeates not only to the tribunal but also to the arbitral award. In the Commonwealth Coatings context, it was argued that evident partiality depends on party expectations, meaning that in niche industries, an arbitrator’s expertise may involve prior dealings with the parties.
- Arbitral deliberation
As discussed by the Spanish Supreme Court in Puma case 102/2017, deliberations are a key aspect of arbitral collegiality among members, and its violation may constitute grounds for the annulment of the arbitral award. However, such deliberations are often unobtainable by the parties on grounds of confidentiality and secrecy of arbitral deliberations,1) 2003 Freshfields – Lecture Dissenting Opinions in International Commercial Arbitration: The Good, the Bad and the Ugly, Arbitration International Vol. 20 Issue 3, September, 2004, p. 223. and may escape the eyes of the parties. As described by Yves Derains, such pathological deliberations are triggered by personal interest rather than objective assessment of issues. As such pathological deliberations are difficult to sever from a unanimous arbitral award, the ICSID annulment committee in Eiser considered piercing the veil of tribunal deliberations impossible. Thus, given the presumption of influence in such awards, distinguishing pathological deliberations from the overall arbitral deliberations in a unanimous arbitral award is as tricky as finding a needle in a haystack.
Conclusion
The decision in Vento II has significantly clarified the issue of single arbitrator bias in a unanimous arbitral award. Subject to certain exceptions existing international jurisprudence does align with the objective standard for assessment of bias in such awards. Although, it does not require an assessment of, the quantum of influence by a biased arbitrator, an implicit influence of bias within the tribunal may be presumed as it is often difficult to distinguish pathological deliberations in a unanimous arbitral award. Thus, both these ‘objective’ and ‘presumed influence’ standards allow for the setting aside of a unanimous arbitral award, even if there was a reasonable apprehension of bias or actual bias against one of the tribunal members.
________________________
To make sure you do not miss out on regular updates from the Kluwer Arbitration Blog, please subscribe here. To submit a proposal for a blog post, please consult our Editorial Guidelines.
References
| ↑1 | 2003 Freshfields – Lecture Dissenting Opinions in International Commercial Arbitration: The Good, the Bad and the Ugly, Arbitration International Vol. 20 Issue 3, September, 2004, p. 223. |
|---|
