Thank you very much for the introduction. Thank you to the ITA for this kind invitation. I am only sorry that we are not together in person. Ladies and gentlemen, let me begin at the ICCA Congress in Singapore that took place in 2012,1For more information, see generally Breakout Session C3 The Relationship Between International Arbitration and the Regulator(s): The Need for Ethical Codes, Guidelines and Best Practices for Arbitration Counsel, Arbitrators, Arbitral Secretaries and Arbitral Institutions: The DB/MS Rio Code, the ILA Code and the CCBE Draft Code, in INTERNATIONAL ARBITRATION: THE COMING OF A NEW AGE?, 17 ICCA Congress Series 465, 465-537 (Albert Jan van den Berg, ed., 2013).X1 a great debate took place between Houston’s own Doak Bishop and Toby Landau, then of London. Their battleground and topic of debate was whether international arbitration needed a code of ethics. Their focus was on the code of ethics for counsel, but the philosophical joust that ensued between them equally could have applied to the positives and negatives of regulating the system of international arbitration more generally.

Mr. Bishop, in one corner, argued that public competence is an essential element of our system, that international arbitration needs to be able to police itself, and that this should take the form of a uniform code of ethics for international arbitration.

Mr. Landau, in the opposite corner, sounded an alarm bell against the specter of ever expanding regulation. The practice of arbitration he warned was increasingly overburdened with protocols, codes, and guidelines. “We are witnessing,” he said, “the pandemic spread of a highly contagious condition: legislitis, a virulent affliction that manifests itself in an involuntary urge to publish booklets of rules, guidelines, and principles on every conceivable arbitration subject.”

Without commenting on whether the Houstonian or the Londoner won the battle of the soundbites during that memorable debate, there can be no doubt as to who has since won the war of ideas. Because since that debate, the world of arbitration has continued to come down firmly on the side of increasing regulation, most prominently and most relevantly for my subject today, in the form of introducing guidelines with respect to the regulation of arbitrators’ duties of impartiality and independence, as well as the related duty of disclosure.

Arbitral institutions are similarly issuing more guidance on the meaning and application of those ubiquitous standards, with some institutions (such as the LCIA) are going further still and publishing their reasoned arbitrator challenge decisions, which provides practitioners with the application of those principles inconcreto. The International Bar Association’s (“IBA”) has also played a significant role, since first publishing2See generally International Bar Association, Guidelines on Conflicts of Interest in International Arbitration, May 22, 2004 (first issuance).X2  (and subsequently updating3See generally International Bar Association, Guidelines on Conflicts of Interest in International Arbitration, Oct. 23, 2014 (updated Aug. 10, 2015).X3) their debated yet still widely relied upon the IBA Guidelines on Conflicts of Interest in International Arbitration (“IBA Guidelines”).

As a practitioner, I have certainly supported and even advocated for the growth of this kind of guidance, on the basis that the growing number of participants in the arbitral process has rightly heightened demands for clarity as to the meaning of those important standards of independence and impartiality, and predictability and transparency as to the manner of their application.
We perhaps should be a little more blunt about the importance of that guidance. The prime paucity of guidance that often handed a broader discretion to a largely well-meaning arbitral elite was no longer good enough for a process of now global importance in which, unlike in national court justice, parties select their arbitrator. In addition, we need to keep reminding ourselves of that fundamental difference between judges and arbitrators when we consider the question of regulating arbitrator duties.

Let us look at what the U.K. Supreme Court said about that difference only a few months ago, in the case of Halliburton v. Chubb4See generally Halliburton Co. v Chubb Bermuda Insurance Ltd [2020] UKSC 48.X4 pertaining to a challenge to a well-known arbitrator for his failure to disclose certain arbitrator appointments that had given rise to doubts as to his impartiality. There the Court observed that “arbitrators have a legal duty to make disclosure of facts and circumstances which would or might reasonably give rise to the appearance of bias. The fact that an arbitrator has accepted appointments in multiple references concerning the same or overlapping subject matter with only one common party is a matter which may have to be disclosed, depending upon the customs and practice in the relevant field. In cases in which disclosure is called for, the acceptance of those appointments and the failure by the arbitrator to disclose appointments taken in combination might well give rise to the appearance of bias.”5Id. ¶ 136.X5

As we look at the words of the U.K. Supreme Court, arbitration proponents may take issue with the description offered in that passage, however, coming as they do from the highest court in the most popular seat of arbitration in recent years, they are not to be ignored.

I believe that there are particularly good reasons for us to have clear standards of conduct for arbitrators and to regulate them robustly. So I do not appear before you today as someone who is against regulation, but I do appear before you to say, almost exactly ten years after that great debate in Singapore took place, that I now believe that in different ways both of our debaters that day have been proven to be right.

I would like to propose two propositions that are relevant to the discussion, which will be illustrated by reference to a recent, prominent national court decision that demonstrated the value of international arbitral codes and guidelines in encouraging consistency and resisting parochialism. I will also refer to a recent and ongoing initiative relating to a new code for arbitrators, which I fear is in danger of going too far.

The first proposition is that we have seen the value over the last decade of greater regulation of arbitrator ethics. To illustrate this first proposition, I am going to return to the case that I have mentioned, the U.K. Supreme Court’s recent decision on arbitrator challenges in the seminal case of Halliburton v. Chubb.1See generally id. I note that I appeared as counsel for one of the intervenors before the Supreme Court in that case, the ICC Court of Arbitration.X1

The Halliburton case concerned two primary issues, (1) whether an arbitrator may accept multiple appointments in interrelated cases with only one common party without thereby giving rise to an appearance of bias; and (2) the related question of whether and to what extent an arbitrator may do so without disclosure. In this case, the underlying dispute concerned the Deepwater Horizon oil spill of 2010, with which many may be familiar, in which Halliburton had provided well cementing services that were implicated as one possible contributing cause of the disaster. After settling a number of U.S. civil claims brought against it, Halliburton (a multinational corporation and one of the world's largest oil field service companies) sought to recover those settled amounts under its liability insurance with Chubb Insurance. Chubb, however, disputed coverage and the relevant insurance policy provided for ad hoc arbitration in London before a tribunal composed of three members.

While the parties each appointed an arbitrator without issue, they could not reach an agreement concerning the appointment of the tribunal chair. Following a contested application to the English high court, Chubb’s preferred candidate was subsequently appointed to chair the tribunal. Almost a year into that first arbitration, however, the same chairman then accepted a separate appointment by Chubb, as its party appointed arbitrator involving the same legal team in another arbitration also related to the Deepwater Horizon incident, but this time involving the owner of the oil rig, Transocean (i.e., not Halliburton), and the chairman failed to disclose this new arbitrator appointment by Chubb to Halliburton. Many months later, upon discovering the chair’s subsequent new appointment, Halliburton applied for his removal under Section 24 of the English Arbitration Act, arguing that circumstances existed that gave rise to justifiable doubts as to his impartiality.2Arbitration Act 1996, Section 24(1)(a) (provision titled “Power of court to remove arbitrator,” and providing that “[a] party to arbitral proceedings may (upon notice to the other parties, to the arbitrator concerned and to any other arbitrator) apply to the court to remove an arbitrator on any of the following grounds— (a) that circumstances exist that give rise to justifiable doubts as to his impartiality . . .”).X2

These circumstances show the difficulties intrinsic to a system of justice where parties can select their decision-maker, and in this case, how a party’s subsequent selection may have a potential to affect an already impaneled tribunal. They also show an important counterbalance of disclosure, and how disruptive of the entire process a failure to disclose early on, or at all, can be. In these circumstances, it is not difficult to see Halliburton’s resulting concern with the repeat appointment. Indeed, the chairman had accepted an appointment in a second case by one of the existing disputing parties, in a related dispute in which only one of the two parties to the first arbitration would have access to him. What is more, their chairman did not disclose this separate appointment, thereby not only leaving Halliburton in an unequal position to Chubb, but entirely ignorant in that fact until it discovered the subsequent appointment by happenstance. On these facts one would think that Halliburton had compelling grounds for challenge, however its challenge was rejected both by the English High Court and then again by the Court of Appeals, in decisions which raised more than a few eyebrows within the arbitration community.

The Court of Appeals’ reasoning on the two key issues that I identified earlier is informative for present purposes. On the first issue of multiple appointments, the Court of Appeals held that the mere fact of the multiple appointments in interrelated arbitrations with one common party did not itself give rise to an appearance of bias, without there being, in the Court of Appeals’ words, something more.3Halliburton Co. v Chubb Bermuda Insurance Ltd [2018] EWCA Civ 817 ¶¶ 53, 76.X3 Nevertheless, the court did not elaborate what that something more needed to be, notwithstanding that it acknowledged that multiple appointments in these circumstances can give rise to what it described as legitimate concerns.4See id. ¶¶ 48, 77.X4 Those concerns related primarily to the creation of an inequality between the parties by putting Chubb in a privileged position of having unilateral access to the common arbitrator on related issues.5Id. ¶ 77.X5 And by putting the common arbitrator in a position of having information on those related issues that have not been derived from the parties submissions in the arbitration in question.

On the second issue of disclosure, although the Court of Appeals agreed with Halliburton that these circumstance ought to have been disclosed, and indeed that such nondisclosure itself reached a legal duty to disclose,6Id. ¶¶ 83-91.X6 it found that the breach had no consequences on the facts of the case because there was, as it subsequently transpired, ultimately only a limited degree of overlap between the two arbitrations.7Id. ¶ 96.X7 The court considered that in the end, the Transocean arbitration was determined on a preliminary issue that did not arise in an earlier Halliburton case.8Id. ¶¶ 99-100.X8

The Court of Appeals decision demonstrates how difficult it is to assess fact-specific conflicts questions, how arbitrator disclosure practices vary widely, and how the outcomes of challenges to impartiality – even in mature arbitration jurisdictions – still have the capacity to surprise.

Because the Court of Appeals decision appeared to many to be out of line with international standards, both on the question of multiple appointments of interrelated cases and on the possible consequences of nondisclosure, when Halliburton appealed that decision to the U.K. Supreme Court, the ICC, the LCIA and a number of other arbitral institutions applied to intervene as nonparties to give the court the benefit of their international perspective. That Supreme Court hearing took place in November 2019, and the Supreme Court’s judgement was rendered less than a year ago in November 2020. In its decision, the Supreme Court, too, rejected the challenge of the arbitrator, on the basis that the English Arbitration Act requires that the court ascertain whether the circumstances that give rise to a justifiable doubt as to impartiality exists at the time – not of the nondisclosure, not of the challenge, but at the time subsequently of the court's decision on the challenge.9Halliburton Co. v Chubb Bermuda Insurance Ltd [2020] UKSC 48 ¶¶ 121-123.X9 In this regard, the Court considered that by the time of those court decisions rejecting the challenge, the dismissal of the Transocean case on a preliminary issue that did not arise in the Halliburton case only served to underscore that the two cases were insufficiently interrelated.10Id. ¶ 149.X10

Whether one agrees or disagrees with the Supreme Court’s decision, it illustrates the difficult and fact-specific nature of arbitrator conflict issues, and confirms the risks to the process if arbitrators under-disclose. In addition, although the Supreme Court did not remove the arbitrator, it did clarify certain disquieting elements in the Court of Appeals decision, thereby finally bringing the English law on independence, impartially and the duty of disclosure squarely back in line with prevailing international standards.11See id. ¶¶ 151-157.X11

With regards to the first specific issue in the appeal, the Supreme Court helpfully confirmed that where an arbitrator accepts appointments, in multiple references concerning overlapping subject matter, this may, depending on those circumstances, give rise to an appearance of bias.12Id. at ¶ 130.X12

On the second issue of disclosure, the Supreme Court confirmed that disclosure was indeed a legal obligation and that the failure to disclose in circumstances of multiple appointments in interrelated cases itself might well give rise to an appearance of bias.13Id. at ¶ 136.X13

The U.K. Supreme Court’s judgment offers a profoundly important analysis on the subject of arbitrators’ duties. For present purposes, just as important as the outcome is the way in which the Court arrived at its judgement, and the role that international standards guidelines played in contributing to the Court’s analysis.

Remarkably, all of the parties relied on the IBA Guidelines on Conflicts of Interest in making submissions to the Supreme Court as to the prevailing international practice today, and were in agreement that the IBA Guidelines represent an international consensus and that there was an important interest in ensuring that English law was not at odds with that international consensus. Although the differing parties did not always have precisely the same view as to the meaning and effect of the IBA Guidelines, notably the applicant Halliburton, the intervenors the ICC, the LCIA (as well as other intervening institutions), and the respondent Chubb (whose legal team included one Toby Landau) all relied upon and placed emphasis on the IBA Guidelines. As counsel for the intervenors at the hearing, it was striking to see how those guidelines afforded all of the parties an immensely valuable common framework from which that hard fought debate could ensue.

From the perspective of an international practitioner interested in ensuring that English law did not depart from emerging international norms, without the IBA Guidelines, it would have been far more difficult to present to the court prevailing international practice in a reliable and objective way. The U.K. Supreme Court itself similarly relied with confidence on the IBA Guidelines in various places,14See id. ¶¶ 71, 80.X14 in a decision that was quite clearly drafted carefully to avoid any charge of departure from that international consensus.

Although one of many court decisions that have referred to and relied on the IBA Guidelines around the world, the Halliburton case provides a striking real life illustration of how the rules and regulations generated by the international arbitration community can assist in ensuring common standards are interpreted and applied in a manner that is consistent across different jurisdictions, in a way befitting of a system of justice which aspires to be truly global. This decision eloquently speaks to the positive contribution that rules, standards and guidelines created by the international arbitration community can have on the outcome of proceedings.

I fear that we are seeing increasing signs of overzealous overregulation, and the still evolving ICSID and UNCITRAL Draft Code of Conduct for Adjudicators in Investor State Dispute Settlement (the “Draft Code”) illustrates some of the concerns that many practitioners have in this regard.1For more information, see generally ICSID, Code of Conduct for Adjudicators in International Investment Disputes and Code of Conduct Resources, https://icsid.worldbank.org/resources/code-of-conduct (providing an overview as well as draft versions, comments, and working papers).X1 A collaborative initiative, the Draft Code was drawn up by ICSID and UNCITRAL and has been the subject of consultation within the forum, in particular of the UNCITRAL Working Group III on ISDS reform. Although the draft code remains a work in progress,2A first version of the Code was published on May 1, 2020, following which ICSID and UNCITRAL received extensive input on the draft through consultation with State delegates and other interested stakeholders. A second version of the Code was published on April 19, 2021 was published taking into account the feedback received. See generally id.X2 the versions of the Draft Code circulated so far are ripe for discussion, as there is still an opportunity for concerned arbitral citizens to make their views known before the Draft Code is finalized, which is presently expected sometime in 2022.

The new Draft Code could be set to have a laudable central aim, mainly addressing the particular ethical issues that may arise for an arbitrator in investment treaty arbitration, the most obvious of which is the double-hatting syndrome in which the same individuals serving as arbitrator and counsel in different cases under different investment treaties that may raise the same correlated questions of law, which justifiably gives rise to concern that warrant increased regulation. The Draft Code addresses this issue, and includes draft proposals, which would prohibit such double-hatting subject to party consent. Whether the prohibition will be limited to narrower circumstances involving only overlapping facts and parties remains to be seen, as that debate is ongoing.

While the initiative to introduce regulations addressing ethical issues that commonly arise in investor-State arbitration, there is cause for concern that the new regulatory initiative has mushroomed in scope. Rather than seeking to address only those issues that might be said to be particular to arbitrators appointed in investment treaty arbitrations, it has been prepared as a comprehensive code that defines the duties of independence and impartiality in a manner that is not entirely consistent with arbitral practice or the way in which those duties have been violated, identifies other duties not addressed by courts or arbitral institutions around the world, and arguably overlaps with existing guidelines such as the IBA Guidelines in a way that may sow confusion. As a result, I suggest that the Draft Code in its present form constitutes regulatory overreach, which in the age of the tactical challenge may create many more problems than it ultimately solves.

A. ICSID/UNCITRAL Code, Draft Article 3

According to the major arbitral rules and statutes around the world, the principles of independence and impartiality are focused, understandably, on circumstances that give rise to or evidence a risk of dependence or partiality. Article 3 concerning Independence and Impartiality will now embellish that focus on factual circumstances by inviting debates on whether a particular arbitrator might be influenced by various emotions, amongst them the fear of criticism.3See ICSID and UNCITRAL, Draft Code of Conduct for Adjudicators in International Investment Disputes, Version 2 dated Apr. 19, 2021, art. 3(1) (“Adjudicators shall be independent and impartial, and shall take reasonable steps to avoid bias, conflict of interest, impropriety, or apprehension of bias.”); id. art. 3(2) (“In particular, Adjudicators shall not: (a) be influenced by self-interest, fear of criticism, outside pressure, political considerations, or public clamor…”).X3

There is no doubt that the strength of character to withstand a fear of criticism is usually a virtue we should welcome in international arbitrators. But requiring it, in a code of conduct would elevate a virtue into a duty, thereby opening a door to a whole new category of challenges to arbitrators and their decisions. This raises a number of evidentiary questions: How do you prove a fear of criticism? How do you disprove it? Is some concern that your decision will invite criticism necessarily a bad thing? Are those arbitrators who pay no regard to what others think necessarily the best adjudicators? This kind of appreciation of an arbitrator’s character and approach should not be a matter subject to regulation. Considering these questions in the context of far more challenging factual situations, such as the complex scenario presented in Halliburton v. Chubb, raises questions as to how challenges under this provision will play out in practice. Indeed it likely will be as difficult to set forth evidence showing a fear of criticism and may add fuel to the fire surrounding arbitrator challenge debates. Experienced regulators in other fields tend to advise that in formulating regulations less is more, that those who regulate should beware the law of unintended consequences, and in this instance, the unintended consequences of regulating in this way are not so difficult to see.

B. ICSID/UNCITRAL Code, Draft Article 6

Article 6 of the Draft Code titled “Other Duties” does not stop at embellishing the duties of independence and impartiality, as it goes further and identifies other ancillary duties.4See id. art. 6(1) (“Adjudicators shall: (a) display high standards of integrity, fairness, and competence; (b) make best efforts to maintain and enhance the knowledge, skills, and qualities necessary to fulfil their duties; and (c) treat all participants in the proceeding with civility.”).X4 While it is certainly to be hoped that your arbitrators will display high standards of civility and competence, a regulatory code of conduct should not be a place to list virtues. Elevating a virtue into a duty creates an additional basis to challenge an arbitrator and his decision. In the age of the tactical challenge, what will parties bent on disruption make of duty to display high standards of competence? If such a party claims that an arbitrator has made a mistake of law, will that allow it to contend that the arbitrator has failed to display that required standard of competence? In this way, will such a duty therefore become a back door to a right of an appeal for a mistake of law? Can we be confident that such a provision will not be used and abused to challenge an arbitrator simply because she has arrived at a decision that a challenging party considers to be wrong? There can be no doubt that the questions this provision gives rise to will quickly become more than simply theoretical questions, and these are precisely the kinds of questions that those who regulate should be considering in the process.

C. ICSID/UNCITRAL Code, Draft Article 10

Arguably one of the most important provisions of the Draft Code, draft Article 10 titled “Disclosure Obligations”5See id. art. 10(1) (“Adjudicators shall disclose any interest, relationship or matter that may, in the eyes of the parties, give rise to doubts as to their independence or impartiality, or demonstrate bias, conflict of interest, impropriety or an appearance of bias. To this end, they shall make reasonable efforts to become aware of such interest, relationship, or matter.”); id. art. 10(2) (“Adjudicators shall make disclosures in accordance with paragraph (1) and shall include the following information: (a) Any financial, business, professional, or personal relationship within [the past five years] with… (b) Any financial or personal interest in: (i) the proceeding or its outcome; and (ii) any administrative, domestic court or other international proceeding involving substantially the same factual background and involving at least one of the same parties or their subsidiary, affiliate, or parent entity as are involved in the IID proceeding; and (c) All IID [and non IID] proceedings in which the Adjudicator has been involved in the past [5/10] years or is currently involved in as counsel, expert witness, or Adjudicator.”).X5 deals with an arbitrator’s disclosure obligations on which many modern challenge situations turn, as was the case in Halliburton v. Chubb already discussed. Specifically draft Article 10(2)(c) addresses a circumstance that might be particularly significant in the context of investment treaty arbitration, and requires the disclosure of all other treaty arbitrations or indeed any other arbitrations that an arbitrator has been involved in over the last five or ten years in an effort to address the problem of double-hatting. One can debate however whether the disclosure requirement for such an extended period of time casts too wide a net, or is even remotely realistic. Over that time period arbitrators may have been involved in hundreds of cases, including commercial arbitrations, many of which are far less likely to be relevant in any event to the issues that arise again and again in treaty arbitrations.

The portions of draft Article 10 that are of a more general nature and address issues that are not unique to investment-treaty arbitration, appear to be duplicative of grounds already covered by other instruments, in particular the IBA Guidelines which have already proven to be useful in cases such as Halliburton v. Chubb, and raises many questions. Is this general standard of disclosure intended to be the same as the IBA Guidelines? Is it intended to be a departure? If so, does that mean that the IBA Guidelines are no longer considered as representing an international consensus on disclosure standards? What effect will that have on future attempts to present these IBA Guidelines as an international standard of best practice to courts that might otherwise favor more parochial solutions? Again, these are questions that those who regulate should be asking themselves. And I would respectfully suggest those driving forward this Draft Code of Conduct might ask themselves again.

As the Halliburton saga and the ICSID and UNCITRAL Draft Code of Conduct illustrate, arbitrator challenges are complex and can give rise to a number of issues that have a material impact on the conduct of proceedings. I have sought to illustrate some of the unintended pitfalls that can accompany expansive regulation in the form of the presently evolving ICSID and UNCITRAL Draft Code of Conduct. As international arbitration has become an increasingly popular and successful method for resolving disputes, new interest groups are understandably involving themselves in the way in which the process works and creating new and more expansive regulation in an effort to standardize the process. Of course, while there can be positives drawn from these changes, there may be unintended consequences for overregulating the conduct of arbitration proceedings. Just as underregulation may no longer be adequate for the modern arbitral processes, so the resulting overregulation can quickly become the road to arbitral hell.

As the question of how to strike the right balance of regulation is not a new one, there are lessons to be learned from examining the waxing and waning cycles of regulation in other fields. As one example, the literature for the state of company regulation that was introduced in the financial sector following the financial crisis of 2008 may prove instructive. In a study published by Aizenman in 2011, he discusses the paradox of underregulation, and how prolonged periods of economic tranquility reducing demand for regulation, and inducing underregulation can itself contribute to financial calamity. On the other side of the regulation paradox, he warns of the tendency to underregulate in good times and the risks associated with overshooting the adjustment needed following a crisis. And he tells us that “these considerations suggest the need to strive towards Goldilocks,” his golden rule of prudential regulation.1J. Aizenman, Financial Crisis and the Paradox of Underregulation and Overregulation, in Lessons from East Asia and the Global Financial Crisis, ABCDE World Bank Conference Volume, 217-218 (Annual Bank Conference on Development Economics) (Justin Yifu Lin and Boris Pleskovic, eds., 2011) (observing “the tendency to underregulate in ‘good times’ and the risk associated with overshooting the adjustment needed following a financial crisis. Both underregulation and overregulation may reflect the paradox of financial regulation: the success of the prudential regulator or a prolonged period of economic tranquility lead to complacency, reducing the demand for the regulator’s services, inducing underregulation, which leads to a financial calamity. … The demand for regulation declines during prolonged good times, increasing the ultimate cost of eventual crises. The other side of the regulation paradox is the hazard of overregulating… These considerations suggest the need to strive toward a golden rule of Goldilocks prudential regulations.”).X1

When we regulate for our system of arbitration, let us be objective so that we can create clear, readily applicable standards that do not require an attempt to read a challenged arbitrator’s mind. Let us be incremental. Given the acceptance of the IBA Guidelines, new efforts should build off that baseline, not duplicate it. In an effort to strike the right balance, let us subject all new regulations to a cost-benefit analysis by asking the essential question: “What is the marginal benefit of the new regulation and does that benefit outweigh the inevitable unintended consequences of all new regulations?” Let us not react to the perception of underregulation in the world of arbitration by now moving to the other regulatory extreme. In a phrase, let us follow Goldilocks’ golden regulatory rule.
Thank you very much for your attention.