On 27-11-2020, the UK Supreme Court finally adjudicated on its most awaited decision: Halliburton Co. v. Chubb Bermuda Insurance Ltd. This case has substantially clarified the issue of arbitrator’s bias to a great extent by providing a high degree of jurisprudential analysis to the same. Considering this case, in the Indian context, firstly, this article shall scrutinise the points at which the Indian arbitration law on arbitrator’s bias differs from that of the international arbitration law i.e. the Halliburton case. Secondly, this article shall also explore the point on which both the Indian and international arbitration law are similar regarding arbitrator’s bias. Thirdly and in finality, this article shall provide a psychological analysis to arbitrator’s bias and suggest ways to mitigate the same.
- Introduction: 2015 Amendment of the Arbitration and Conciliation Act, 1996 and factual matrix of the Halliburton case
Prior to 2015, there was no law in the Arbitration and Conciliation Act, 1996 (hereinafter “1996 Act”) to ensure the unbiasedness of the Indian arbitrators. However, the 1996 Act was amended through the Arbitration and Conciliation (Amendment) Act, 2015 that introduced “safeguards” to ensure the fairness, impartiality, and independence of arbitrators while mitigating the inherent conundrum of human bias. In the newly amended 1996 Act, Section 12 was amended that now introduced grounds for challenging an arbitrator’s bias alongside introduction of the Fifth and Seventh Schedule. Section 14 was also amended that allowed the substitution of another arbitrator upon termination of the arbitrator’s mandate.
In Halliburton case, claims arose out of the deepwater horizon incident. These claims were made against Halliburton who provided offshore services. In return, Halliburton sought claim against Chubb under Halliburton’s excess liability insurance policy with Chubb. Halliburton’s claim was rejected and in January 2015, they commenced arbitration proceedings against Chubb. Halliburton’s claim was then brought under the policy of Bermuda form in question; this arbitration proceeding was governed by the law of New York and provided for a London–seated ad hoc arbitration. After this, both the parties were unable to agree on the selection of a presiding arbitrator due to which, the English High Court appointed Kenneth Rokison as the arbitrator in June 2015. It is pertinent to note that Mr Rokison was proposed by Chubb but opposed by Halliburton on the ground that Mr Rokison was an English lawyer whereas the Bermuda form policy was governed by the law of New York.
Prior to 2015, Mr Rokison disclosed that he had been an arbitrator in arbitrations that involved Chubb and was currently in the midst of two arbitration proceedings involving Chubb. The conundrum regarding Mr Rokison arose because, after his appointment as an arbitrator between Halliburton and Chubb, he took up two additional arbitration proceedings that were related to the deepwater horizon incident. He did not disclose these two additional proceedings to Halliburton that later became aware of the same in November 2016. Due to this, Halliburton asked Mr Rokison to resign; however, he denied and noted that the issues in the Halliburton arbitration proceedings were neither the same nor similar to the two additional arbitration proceedings he took up in 2015 and 2016. Further, he also stated that he has been independent and impartial throughout the Halliburton arbitration proceedings and would continue to be the same.
However, this was unacceptable to Halliburton that made an application to the English Court for Mr Rokison’s removal. The original application and the appeal in the court of appeal were both unsuccessful due to which Halliburton appealed to the Supreme Court. The UK Supreme Court held that it was not possible for a fair-minded and informed observer to conclude biasedness on the part of Mr Rokison.
- Analysis: What is the status of Indian arbitration law regarding arbitrator’s bias in the context of the Halliburton case?
Section 12(3)(a) of the 1996 Act states that an arbitrator’s independence or impartiality may be challenged only if “circumstances exist that gives rise to justifiable doubts” regarding his independence or impartiality. Before we comprehend what these “circumstances” are, it is imperative for us to distinguish between independence and impartiality. In Voestalpine Schienen GmbH v. Delhi Metro Rail Corpn. Ltd., the Supreme Court held the following:
… independence and impartiality are two different concepts. An arbitrator may be independent and yet, lack impartiality, or vice versa. Impartiality, as is well accepted, is a more subjective concept as compared to independence. Independence, which is more an objective concept, may, thus, be more straightforwardly ascertained by the parties at the outset of the arbitration proceedings in light of the circumstances disclosed by the arbitrator, while partiality will more likely surface during the arbitration proceedings.
Thus, from the aforementioned judgment, the line of distinction between independence and impartiality is clear that both are inherently quintessential characteristics and ethical values for an arbitrator to possess for just, effective, and efficient award delivery.
Moving on, under Section 12 of the 1996 Act, it is imperative for us to comprehend what these “circumstances” are that give rise to “justifiable doubts” regarding an arbitrator’s impartiality and independence. The Fifth Schedule of the 1996 Act lays down the “grounds” or circumstances that give rise to such doubts. These grounds are thirty-four in number and range from arbitrator’s relationship with the parties or counsel, direct or indirect interest in the dispute to having provided services for one of the parties involved in the dispute. The Seventh Schedule of the 1996 Act provides certain “categories” that when “any person whose relationship, with the parties or counsel or the subject-matter of the dispute, falls under any of the categories specified … shall be ineligible to be appointed as an arbitrator”. Thus, these Two Schedules of the 1996 Act essentially act as grounds/categories to substantiate these “justifiable doubts” regarding an arbitrator’s impartiality and independence. In other words, these Two Schedules act as factors to guide in determining whether “circumstances” exist which give rise to such “justifiable doubts”.
At this point in time, it is pertinent to note that in Halliburton case, the Supreme Court observed that different codes use different expressions to describe the nature of “doubts”. For example, General Standard 3 of the IBA Guidelines only state “doubts”; Article 11 of the International Chamber of Commerce (hereinafter “ICC”) Rules state “reasonable doubts”; and Article 5.4 of the London Court of International Arbitration (hereinafter “LCIA”) Rules speak of “justifiable doubts”. In this context, Lord Hodge observed that: “[B]ut I do not think that there is a material difference between those formulation as I do not construe the IBA Guidelines or the institutions’ rules as requiring disclosure when the only doubts to which the circumstances might give rise would be unreasonable or unjustified.” In other words, it is lucidly evident that “circumstances” which give rise to any type of “doubts” regarding an arbitrator’s impartiality or independence is essentially for the purpose of determining and ascertaining an arbitrator’s biasedness. Thus, it is safe for us to infer that the factors mentioned in the Fifth and Seventh Schedule of the 1996 Act are to enable the better determination and ascertainment of arbitrator’s bias. The same was reiterated in Voestalpine Schienen GmbH case wherein it was lucidly held that the amended Section 12, and the newly added Fifth and Seventh Schedule conform to International Bar Association (hereinafter “IBA”) guidelines for inducing independence and impartiality in arbitrators.
However, the first point of conundrum arises is that these factors mentioned in the Two Schedules of the 1996 Act are not in conformity with the International Law, particularly the English law. This is because, in Locabail (UK) Ltd. v. Bayfield Properties Ltd., the court of appeal lucidly held that an attempt to define or list factors to ascertain arbitrator’s bias, which may or may not necessarily give rise to “bias,” is a dangerous and futile attempt. In this regard, the court of appeal held that: “[e]verything will depend on the facts, which may include the nature of the issue to be decided”. The same stance was repeated in Halliburton case which laid out that for arbitrator’s biasedness to be identified, firstly, the objective test of the fair-minded and informed observer (or also called the objective test of appearance of bias) must be applied. In this test, this “fair-minded and informed” observer does not reach to a judgment or a conclusion on any point unless and until this observer has acquired a full understanding of both sides of the argument. Further, the following was observed:
- 3. Then there is the attribute that the observer is “informed”. It makes the point that, before she takes a balanced approach to any information she is given, she will take the trouble to inform herself on all matters that are relevant. She is the sort of person who takes the trouble to read the text of an article as well as the headlines. She is able to put whatever she has read or seen into its overall social, political or geographical context. She is fair-minded, so she will appreciate that the context forms an important part of the material which she must consider before passing judgment.
Further, as per Lord Hodge, the observer is “neither complacent nor unduly sensitive or suspicious”. Thus, Lord Hodge, relying on Porter v. Magill, concluded that this fair-minded and informed observer, after having considered all facts, would conclude that whether there was a real possibility that the Tribunal was biased. In other words, if this observer, only after meeting the aforementioned standards, concludes that an arbitrator was biased, then the same decision regarding an arbitrator being biased shall stand. This would mean that the arbitrator has failed to pass the objective test.
Secondly, while deciding on the arbitrator’s biasedness, certain characteristics of arbitration must be taken into consideration. These characteristics are as follows: although the arbitrator is under a statutory obligation to disclose information, privacy and confidentiality of the arbitration proceedings is also of quintessence importance that must be balanced with this legal duty to disclose; there are very limited powers for reviewing an arbitrator’s decision; whether the arbitrator has a financial interest in the dispute; arbitrators may be highly experienced or be limited in terms of their experience alongside their varying views on “what constitutes ethically acceptable conduct”; “due to the private nature of arbitration, where an arbitrator is appointed in relation to multiple overlapping references the non-common party cannot discover what evidence or submissions have been put before the Tribunal, or the arbitrator’s response”; a party-appointed arbitrator is expected to be of the same high standards of fairness and impartiality as the person chairing the Tribunal; and lastly, the reputation and experience of an arbitrator can only be accorded “limited” weight. Therefore, it is safe for us to infer that no particular “guiding” factors can be listed or made for the purposing of determining an arbitrator’s bias because it must be decided on a case to case bias while applying the objective test and the certain characteristics of arbitration as laid above.
Section 12(1)(a) of the 1996 Act makes it compulsory for the arbitrator to disclose, in writing, any circumstances “… such as the existence either direct or indirect, of any past or present relationship with or interest in any of the parties or in relation to the subject-matter in dispute, whether financial, business, professional or other kind, which is likely to give rise to justifiable doubts as to his independence or impartiality”. This legal duty of disclosure and its procedure has been clarified in HRD Corpn. v. GAIL (India) Ltd. wherein it was lucidly held that an arbitrator must disclose the grounds — mentioned in the Fifth Schedule of the 1996 Act — which give rise to justifiable doubts regarding an arbitrator’s independence and impartiality. Further, if any of the party wants to challenge an arbitrator’s appointment on the ground of arbitrator’s impartiality and independence being doubtful, then they can pursue this challenge under Section 13 of the 1996 Act. The same clarification and procedural aspect were repeated by the Supreme Court in Bharat Broadband Network Ltd. v. United Telecoms Ltd.
However, the second point of conundrum arises is that the Indian jurisprudence on this legal duty of disclosure is not adequately developed to cover the substantial aspect as it fails to move beyond the procedural technicalities. This substantial aspect has been covered in Halliburton case that has answered several burning queries regarding this legal duty of disclosure. Firstly, Lord Hodge lucidly stated that an arbitrator may avoid an appearance of bias on his part if he discloses all the matters that he is legally supposed to disclose because such disclosure “… allows the parties to consider the disclosed circumstances, obtain necessary advice, and decide whether there is a problem with the involvement of the arbitrator in the reference and, if so, whether to object or otherwise to act to mitigate or remove the problem”.
Secondly, even if the arbitrator fails to make disclosure due to reasons such as oversight, forgetfulness or failure to properly recognise how matters would appear to the objective observer, it will still amount to non-disclosure that would inevitably colour the thinking of the arbitrator.
Thirdly, if there is a statutory obligation to make disclosure, then this is called as statutory obligation of fairness that is essentially a corollary of the statutory obligation of impartiality. In other words, if an arbitrator fails to make a disclosure, then he fails to act according to the standards that fairness requires him to act. This failure to adhere to fairness standards, in turn, potentially detriments a party, thereby rendering the arbitrator guilty of partiality. In the Indian context, this statutory obligation to make disclosure is enshrined in Section 12(1) of the 1996 Act.
Fourthly, there is a blurry area that exists regarding the relationship between disclosure (of information that would give rise to doubts regarding an arbitrator’s impartiality and independence i.e. appearance of bias) and confidentiality and privacy. As the arbitrator acquires information during an arbitration proceeding wherein the parties put a certain degree of faith, trust and confidence in this arbitrator regarding their sensitive information, the arbitrator is bound to uphold the privacy and confidentiality of arbitration. This brings to the following question: what are the boundaries concerning an arbitrator’s ability to make disclosure while upholding privacy and confidentiality? This question was also answered in Halliburton case.
Lord Hodge gave the example of “multiple appointments” or in other words, the arbitrator being involved in multiple arbitration proceedings that involves a common party. He observed that each type of arbitration requires a different standard of disclosure. In the context of the “multiple appointments” example, Lord Hodge lucidly observed that there is no “general practice” to disclose multiple overlapping arbitrations in maritime, sports and commodities arbitrations as such non-disclosure is not considered to constitute arbitrator’s bias.
Further, Lord Hodge observed that for an arbitrator to make disclosure, he has to obtain the consent of the relevant party. Such consent can be expressed or implied. In cases where an arbitration takes place in institutions such as LCIA, ICC, and so on, consent of the parties is “implied” because such institutions mandate disclosure to the institution or the parties of matters [Article 11(2) of ICC Arbitration Rules and Article 5.4 of LCIA Rules] wherein such disclosure may include information about other arbitrations. This mandatory disclosure by certain institutions have become a “general practice”. In other words, when parties’ arbitration proceedings take place in such arbitral institutions, it is assumed that they have consented to such “general practice”.
On the other hand, in Grain and Feed Trade Association (hereinafter “GAFTA”) and London Maritime Arbitrators Association (hereinafter “LMAA”), it is a “general practice” that arbitrators can act in multiple arbitrations, often arising out of the same events, thereby making disclosure regarding multiple appointments not mandatory.
When it comes to the aspect of “expressed” consent, this legal duty of disclosure is also dependent on the relevant arbitration agreement. In other words, the “… parties to an arbitration can determine as a matter of contract the extent to which they wish matters to be treated as confidential”.
However, in case an expressed or implied consent is absent, then the general practice and custom prevailing in the particular field and place is scrutinised. In English-seated arbitrations, the arbitrators are allowed to make high-level disclosures of their involvement in other relevant arbitrations without obtaining the express consent of the parties. Thus, it can be lucidly inferred that the legal duty of disclosure is highly dependent on the general practice and custom prevailing in a particular field and place.
Failure to make a disclosure when it is legally mandated either by such general practice and custom or by expressed consent, then this failure “… may demonstrate a lack of regard to the interests of the non-common party and may in certain circumstances amount to apparent bias”. Such a failure shall be considered by the fair-minded and informed observer to ascertain whether there existed a real possibility of bias. Further, such observer must consider the facts and circumstances as at and from the date when the legal duty to disclose arose alongside the facts and circumstances known at the date of the hearing to remove the arbitrator.
Fifthly and in finality, the legal duty to make disclosure arises when the arbitrator acquires knowledge of the requisite circumstances that may lead to a doubt to his independence or impartiality. In other words, this legal duty of disclosure must be considered prospectively rather than retrospectively i.e. by disregarding the matter which the arbitrator could not have possibly known at that time. For example, on January 1, A is appointed as an arbitrator between B and C. On February 1, A is also appointed as an arbitrator between B and D. On March 1, A is appointed as an arbitrator between B and E. Thus, a legal duty to disclose such multiple appointments to C arises on February 1 and March 1 respectively and not on January 1. This is because it would not have been practically possible to disclose such overlapping appointments on January 1 due to their non-existence and the arbitrator being unaware of such new appointments. Thus, the duty of disclosure is a continuing one.
- Conclusion: Conformity of Indian arbitration jurisprudence on arbitrator’s bias with that of international standards – Not a myopic goal
One point where the Indian arbitration law and the international arbitration law coexist regarding arbitrator’s bias is on the concept of existence of “apparent bias”. This type of bias is different from “actual bias” which was distinguished in Director General of Fair Trading v. Pty. Assn. of Great Britain. The UK Court of Appeal essentially held that “actual bias” represents a “demonstrable” situation wherein a Judge’s decision has been influenced by partiality or prejudice. On the other hand, “apparent bias” represents the mere existence of a reasonable “apprehension” and “possibility” that the Judge may have been, or may be, biased. In the Indian case of A.K. Kraipak v. Union of India, the Supreme Court lucidly held that: “The real question is not whether he was biased. It is difficult to prove the state of mind of a person. Therefore, what we have to see is whether there is reasonable ground for believing that he was likely to have been biased.” Thus, this concept of “apparent bias” is directly in conformity with the principle of apprehension or appearance of bias regarding an arbitrator’s bias found in Halliburton case. In this context, it is imperative to note that Article 57(13) of the International Centre for Settlement of Investment Disputes (hereinafter “ICSID”) Convention mandates a manifest lack of independence and impartiality, as per the qualities laid down in Article 14(1) of ICSID Convention, to bring a challenge against an arbitrator. This “manifest lack” essentially means that the lack of independence and impartiality must be vividly evident or clear. However, this “manifest lack” proves counterproductive as an arbitrator may be subtly and indirectly biased i.e. unconscious bias, that may not be prima facie evident. Due to this, such problematic subtleties can flow in any part of his arbitral award and considerations, thereby leading to a potential detriment to a party.
One may argue that the test of “justifiable doubts,” as evident in the 1996 Act, cannot be equated with the objective test of fair-minded and informed observer. However, it is quintessential to comprehend that, in Halliburton case, Lord Hodge lucidly observed that this test of “justifiable doubts” — as adopted in the United Nations Commission on International Trade Law Model Law on International Commercial Arbitration, 1985 (hereinafter “UNCITRAL Model Law”) – is the same as that of the objective test. This is because the tests as to the nature of doubts is more or less the same in English law as such tests essentially mandate objective assessment of arbitrator’s bias. Thus, as we are already aware that the 1996 Act is based on the UNCITRAL Model Law, implying that the test of “justifiable doubts” found in the Model Law and the 1996 Act are one and the same, the test of “justifiable doubts” in the 1996 Act can be equated to that of the objective test.
One final aspect that this article shall delve into is that of the psychological aspect regarding arbitrator’s bias. Multiple cases have acknowledged the existence of unconscious bias within the minds of Judges and arbitrators. In Reg. v. Gough, Lord Goff stated that: “[T]here is also the simple fact that bias is such an insidious thing that, even though a person may in good faith believe that he was acting impartially, his mind may unconsciously be affected by bias.” This bias essentially takes place because the Judges, or the arbitrators, that are involved in the decision-making process, are human beings who have been exposed to different scenarios in life that has led to their subjective social, cultural, and psychological conditioning. Further, any form or extent of legal training cannot and does not alter the inherent and fundamental reality of their subconscious that has been developed throughout their early years. It has also been highlighted that “hindsight” is the most troublesome conundrum for Judges. Hindsight essentially means that a situation or an event is understood only after it has taken place. Arbitrators face similar issues and often advocate to not be influenced by such a conundrum. Apart from hindsight, Judges and arbitrators also face the problem of egocentrism regarding the correctness of their decisions. The same has been summed up as follows by one of the most highly regarded American jurists, Judge Posner: “the psychology of judging includes the belief that one is almost always (some Judges think always) right”.
Thus, to counteract and minimise this bias, J.P. Linahan, In re, Judge Frank suggested the following remedy: “The conscientious Judge will, as far as possible, make himself aware of his biases … and by that very self-knowledge nullify their effect.” Further, two systems had been proposed by Daniel Kahneman, a noble prize-winning psychologist: System 1 wherein decisions are taken according to our fast, automatic, high capacity, low effort and intuitive mode; and System 2 wherein decision are taken according to our slow, deliberate, limited capacity and high-effort mode. In System 1, we comprehend that the probability of biasedness is higher because there is a higher reliance on our “intuition” for the purpose of decision-making. This “intuition” is problematic because it is inherently shaped and influenced by our psychological conditioning, thereby leading to no voluntary filtering of such subconscious biases. However, in System 2, we observe that the probability of biasedness is low because there is a voluntary exercise of our deliberation and decision-making capabilities. Due to such voluntary exercise, we comprehend that the ability and agency to put filters for weeding out biases is higher and more effective and efficient because now, the arbitrators would consciously weigh the pros and cons of the matter alongside the evidence and other concerned material. Thus, System 2 must be actively promoted and encouraged in the Indian arbitration regime to remove the possibility of arbitrator’s biasedness wherein training can be provided for the same.
Therefore, although we comprehend that the Halliburton case is an international case based on English law, it is of quintessence importance to note that this case has resolved multiple ambiguities in the area of arbitrator’s bias by providing a substantial amount of jurisprudence on the same, thereby becoming a landmark superior in the international community and setting an international standard for arbitrator’s bias. Due to the setting of such a new international standard, it is quintessential for the Indian arbitration regime to adopt and catch up to the same for ensuring an appropriate blooming, growth, and development from its nascent stage.
† Iram Majid, Director of Indian Institute of Arbitration and Mediation (ILAM) and Executive Director of Asia Pacific Centre for Arbitration and Mediation.
 Arbitration and Conciliation Act, 1996, S. 12 (hereinafter 1996 Act).
 1996 Act, S. 14.
 1996 Act, S. 12(3)(a).
 1996 Act, S. 12(5).
 Porter v. Magill, (2002) 2 AC 357 : 2001 UKHL 67.
 UK Supreme Court Judgment in Halliburton v Chubb Clarifies English Law on Arbitrator Apparent Bias, Herbert Smith Freehills (1-12-2020), <https://hsfnotes.com/arbitration/2020/12/01/uk-supreme-court-judgment-in-halliburton-v-chubb-clarifies-english-law-on-arbitrator-apparent-bias/>.
 1996 Act, S. 12(1)(a).
 1996 Act, S. 13.
 1996 Act, S. 12(1).
 (2001) 1 WLR 700.
 Halliburton Co. v. Chubb Bermuda Insurance Ltd., (2020) 3 WLR 1474 : 2020 UKSC 48.
 1993 AC 646 : (1993) 2 WLR 883.
 Edna Sussman, Biases and Heuristics in Arbitrator Decision-Making: Reflections on how to Counteract or Play to them, in The Roles of Psychology in International Arbitration, 45, 45 (Tony Cole, Ed., 2017).
 Sussman, id. at 46.
 Sussman, id. at 53.
 Sussman, id. at 53.
 Sussman, id. at 57.
 Richard A. Posner, The Problems of Jurisprudence, 192 (1990).
 138 F 2d 650, 652 (2nd Cir, 1943).
 Sussman, supra note 56 at 47.