Apparent bias and conflicts of interest in the arbitral process
Speech delivered at the ADR Address 2019, Sydney, 3 October 2019

I commenced my life as a barrister and finished my life as a judge, ensconced in the sanctity of the litigious process. Although out-of-court settlement of disputes was routine at the Bar and later warmly received by myself and other overburdened judges, a formalised system of Alternative Dispute Resolution – including mediation and arbitration – was peripheral to my work. I was concerned with the running of cases and later, as a judge, with the determination of cases in courts of law.

Given that background, it is appropriate that I commence tonight by reasserting the essence of the function of the civil judicial process in society. Rather than placing my own ‘spin’ on this, let me quote from Professor Hazel Genn QC 2 :

“The civil justice system provides the legal architecture for the economy to operate effectively, for agreements to be honoured and for the power of government to be scrutinised and limited.”

Professor Genn then describes the relationship between the law and the courts as follows:3

“The civil law maps out the boundaries of social and economic behaviour, while the civil courts resolve disputes when they arise. In this way, the civil courts publicly reaffirm norms and behavioural standards for private citizens, businesses and public bodies.”

Notwithstanding the centrality of the judicial system in and to ‘civil society’, given the frequency with which mention is made of ADR in chambers, in the streets surrounding the law courts and in law schools, one might be forgiven for thinking we are in the midst of an ADR epidemic of tsunamic proportions.

For the traditionally minded, this is sometimes seen as a dangerous intrusion into the hallowed halls of litigation and the judicial process – together the keeper of the ‘rule of the law’. The title of Chief Justice Bathurst’s address last year: “ADR, ODR and AI-DR or Do We Even Need Courts Anymore” 4 acknowledged the threat presented by this intrusion.

Sometimes the ADR intruder is pitched as an archcompetitor of the courts as the institutional protector of the rule of law. Modern protagonists of ADR contend for a free market where choice is the underlying precept. As legal historians well know, however, ADR, including arbitration, is as old as conflict itself. 5

Two trends of the last half century are responsible for bringing to the fore presence and, if you like, the intrusion of arbitration, into dispute resolution. One is the globalisation of commerce and trade, about which nothing further need be said. The other is the increasing institutionalisation of arbitration.

The almost universal uptake of the Convention on the Recognition and Enforcement of Foreign Arbitral Awards (the New York Convention) and the implementation around the world of the United Nations Commission on International Trade Law (UNCITRAL) Model Law on International Commercial Arbitration (the Model Law) have been major forces in this process. To that can be added the establishment of the various arbitral bodies and institutes, including, in the Australian context, the Australian Centre for International Commercial Arbitration (ACICA). And that is to say nothing of the national and international journals on arbitration, innumerable academic studies and university courses on ADR and arbitration in particular, and deluge of international student moots.

Central to the principles embodied in the New York Convention are the recognition and enforcement of agreements to arbitrate 6 and of arbitral awards. 7 In the Australian and the UK contexts these obligations, that is, recognition and enforcement, have a litigious history. These include whether a particular agreement constitutes an agreement to arbitrate and the reach of an arbitral clause. It is sufficient for present purposes to refer, by way of example, to Rinehart v Hancock Prospecting Pty Ltd. 8

However, those questions are not the subject of tonight’s lecture. Rather, the focus of my address is the question of apparent bias and conflicts of interest in the arbitral process. I will make reference to the stop light categories in Part II of the International Bar Association (IBA) Guidelines, which identify categories of actual and potential conflicts, in lists running from red, where an arbitrator should not accept an appointment, through to green, where no question arises as to whether the arbitrator should accept the appointment to sit.

Having set the framework for discussion, and notwithstanding my opening remarks, I should indicate that I do not view the various forms of dispute resolution, whether curial or arbitral, as being in competition – although I am acutely aware that the supporters of the latter are vocal as to the benefits of arbitration as compared to court processes.

Rather, as has been frequently emphasised, 9 they are properly viewed as parallel forms of dispute resolution. The two modes of dispute resolution also exist in a symbiotic relationship, most significantly in respect of the recognition and enforcement of arbitral awards.

Conflicts of interest

A significant issue that arises in commercial arbitration is the potential for an arbitrator to have a conflict of interest, raising questions as to his or her independence and impartiality. 10

The Model Law addresses this issue. 11 Article 12(1) of the Model Law imposes on an arbitrator a continuing obligation of disclosure of any circumstance likely to give rise to justifiable doubts as to his or her impartiality or independence. 12 Article 12(2) provides for the making of a challenge to an arbitrator’s appointment 13 , but, relevantly for the purposes of this paper, “only if circumstances exist that give rise to justifiable doubts as to [the arbitrator’s] impartiality or independence…”. A challenge to the independence or impartiality of an arbitrator may be made directly and may also form the basis of a challenge to an arbitral award. 14

It is important to note in the Australian context that the International Arbitration Act 1974 (Cth) was amended in 2010. Section 18A, introduced as part of those amendments, provides that for the purposes of Arts 12(1) and (2) of the Model Law “there are justifiable doubts as to the impartiality or independence of an arbitrator only if there is a real danger of bias on the part of the arbitrator in conducting the arbitration”. 15

Section 18A picks up the language of those English authorities, including R v Gough 16 (Gough) and Locabail (UK) Ltd v Bayfield Properties Ltd and Another , 17 which applied the ‘real danger’ test. In Gough , Lord Goff said:

“… I think it unnecessary, in formulating the appropriate test, to require that the court should look at the matter through the eyes of a reasonable man, because the court in cases such as these personifies the reasonable man; and in any event the court has first to ascertain the relevant circumstances from the available evidence, knowledge of which would not necessarily be available to an observer in court at the relevant time. Finally, for the avoidance of doubt, I prefer to state the test in terms of real danger rather than real likelihood, to ensure that the court is thinking in terms of possibility rather than probability of bias.” 18

Lord Goff said that the question to ask was whether “there was a real danger of bias … in the sense that [the arbitrator] might unfairly regard (or have unfairly regarded) with favour, or disfavour, the case of a party to the issue under consideration by him” 19 .

While subject to some academic debate, s 18A, it would seem, creates a higher threshold or hurdle for a successful challenge to an arbitrator’s appointment on the grounds of bias than do Arts 12(1) and (2). This correspondingly is likely to reduce the possibility of derailing an arbitration by challenging the appointment of an arbitrator.

I have to this point referred to the principal tests for determining bias under various arbitral laws and instruments by way of introduction to the issues and cases I wish to discuss. Before proceeding, however, I wish to refer to the High Court’s decision in Ebner v Official Trustee in Bankruptcy 20 ( Ebner ) which, on the test there stated, requires that a judge be disqualified from adjudicating a dispute

“… if a fair-minded lay observer might reasonably apprehend that the judge might not bring an impartial mind to the resolution of the question the judge is required to decide”. 21

I have referred to Ebner for two reasons. First, as the High Court pointed out, the test for apprehended bias does not require a prediction as to how the judge will approach or determine the case. 22 The question is “one of possibility (real and not remote)”.

Secondly, the observation in Ebner as to the two-step application of the test is also useful. The first step requires “the identification of what it is said might lead a judge … to decide a case other than on its legal and factual merits”. 23 The second step requires that there must be “an articulation of the logical connection between the matter [said to constitute the perceived bias] and the feared deviation from the course of deciding the case on its merits”. 24

The Ebner approach ensures a robust and pragmatic assessment of potential or apprehended bias as is demonstrated in its result. There, the High Court held that the judge’s ownership of shares in a public company did not give rise to an apprehension of bias where the decision of the court would not affect the value of the shares.

Also of interest was the Court’s acceptance, as a reasonable but non-exclusive categorisation, of the circumstances in which there may be an apprehension of bias: interest, conduct, association and extraneous information. 25 These may seem, and often are, obvious, although in a modern context questions of interest and association can give rise to difficult questions, including in respect of “ideological” associations. 26

That brings me to conflicts of interest. The circumstances in which there are likely to be questions of apprehended bias are various and, again, in many cases, obvious. In this regard the stop light lists of relationships in the IBA Guidelines serve as a useful port of call. Nonetheless, disputes still arise and there are a myriad of cases that can be examined. I will refer to two UK cases by way of example.

In A&AT Corporation and Lucent Technologies Inc v Saudi Cable Company , 27 the nominated arbitrator had failed to disclose that he was a director of a company that had lost a bidding war for the project subject of the arbitral proceeding. A&AT, the successful bidder, challenged his appointment. Whilst that challenge was being litigated the tribunal made several interim rulings, all against A&AT. The UK Court of Appeal ruled against the challenge on the basis that as a non-executive director of the unsuccessful company which might indirectly profit from his rulings against A&AT, there was no “real danger” of bias.

In Cofely Limited v Bingham & Knowles Limited , 28 ( Cofely ) the relevant test for apprehended bias was expressed in the terms of art 12 of the Model Law. 29 Of concern in that case was the repeat appointment of the arbitrator. That fell into the IBA Guidelines orange list calling for disclosure where an arbitrator had received, or had been nominated for two or more appointments over a three year period by one of the parties. The purpose of disclosure under this Guideline is to protect against an arbitrator’s material financial dependence on a particular party. 30

The arbitrator in Cofely had failed to disclose that he had derived 18% of his appointments and 25% of his income as arbitrator or adjudicator over a three year period from one of the parties to the arbitration. The repeat appointments were not the only problem. When information was sought about how many appointments and nominations he had had from the opposing party the arbitrator had responded in an evasive and aggressive manner, adding considerable angst to the dispute.

What was particularly concerning to the Court was a methodology that the opposing party had developed to steer the appointment process towards its favoured arbitrator. It did this in two ways. First, it sought an arbitrator with a very narrow skill set. Secondly, it maintained a black list of arbitrators. The concern, as expressed by Hamblen J was that:

“It means that the arbitrator/adjudicator’s conduct of the reference may lead to him/her falling out of favour and being placed on that list and thereby effectively excluded from further appointments involving [that party]. That is … important for anyone whose appointments and income are dependent on … related cases [of that party] to a material extent”. 31

Position where the various rules and guidelines are silent

What is the position, however, where the arbitral rules or guidelines make no provision or have no apparent application to the circumstances that arise in a particular case?

An interesting question arose in two cases from the International Centre for Settlement of Investment Disputes (ICSID). The question in issue was not whether a party could successfully challenge the arbitral tribunal, or an individual member of the tribunal, but whether the tribunal could refuse the right of appearance of a party’s legal representative. The outcome was different in each case. More relevantly, there was a significant difference in the reasoning in each case.

In the first, Hrvatska Elektroprivreda dd v Republic of Slovenia , 32 (Hrvatska), the Tribunal President was a door tenant in the same chambers as the English counsel retained to appear at the hearing. This was not disclosed at the time that the Tribunal was constituted, the respondent considering it unnecessary to do so. Shortly before the hearing proper, the claimant became aware of the “association” of the Tribunal president and counsel. By reference to the IBA guidelines, the relationship fell into the “orange category” of potential for conflict. The Tribunal barred the right of counsel to appear.

In the second case, The Rompetrol Group NV v Romania , 33 (Rompetrol) the conflict was said to arise in circumstances where counsel had previously been an employee of the law firm of which a tribunal member was a partner, a circumstance which does not fall into any of the IBA lists. The Tribunal refused the application.

Two fundamental principles were in issue in each case: a party’s right to a fair and impartial arbitral tribunal; and the right of a party to an adequate opportunity to present its case to the tribunal 34 . Integral to this latter right is the right to choose one’s legal representative. Taken together, these principles are sometimes referred to as the ‘Magna Carta’ of International Commercial Arbitration. 35

In each case, if the application were granted, the respondent might contend that it was denied its right to the full presentation of its case under art 18 of the ICSID Arbitration Rules. If denied, the claimant might contend that there had been a breach of the requirement to adjudicate the award fairly as required by r 6. Either way, any award was potentially contestable.

In Hrvatska , the Tribunal’s starting point was its “obligation as guardian of the legitimacy of the arbitral process to make every effort to ensure that the Award [was] soundly based and not affected by procedural imperfection”. The Tribunal considered, by reference to the IBA Guidelines, that English barristers chambers, in their modern collective emanation, should be treated no differently from a law firm and that there would be some who would consider that favouritism might be shown to counsel from the same chambers as the arbitrator. The Tribunal’s conclusion therefore was that the circumstances were such as could lead a reasonable observer to form a justifiable doubt as to the impartiality of the arbitrator. As there was no relevant express power to bar counsel from appearing, the Tribunal searched for and found an inherent power to do so.

In Rompetrol , the Tribunal applied the test in Porter v Magill , 36 namely “whether [a] fair-minded and informed observer, having considered the facts, would conclude that there was a real possibility that the tribunal was biased”. 37 The Tribunal concluded that there was no basis to intervene. On the facts, that was hardly surprising, the relationship being far more tenuous than in Hrvatska .

Whilst differing outcomes are not unusual in adjudicative deliberations, the two decisions raise a question of principle. Does an arbitral tribunal have the power to control the appearance or conduct of counsel?

In Hrvatska , the Tribunal, being able to locate neither any express nor implied power, considered that, as a judicial formation governed by public international law, it was vested with inherent power to take measures to preserve the integrity of its proceedings.

The Tribunal also found support for the existence of an inherent power in art 44 of the ICSID Convention which provides that a tribunal has a discretion to decide any question of procedure not expressly covered by the Arbitration Rules. Two things might be noted about this part of the reasoning. First, if the matter was covered by art 44, the Tribunal would have an express, albeit discretionary, power to make the order. However, and this is the second point, it is unlikely that choice of counsel, integral to the right to adequately present one’s case, can be properly characterised as a matter of procedure.

The reasoning in Rompetrol is not so clear as to the existence of an inherent power. The fact that the Tribunal adjudicated on the issue would imply that it considered it had power, although it too was unable to locate any express power. The Tribunal acknowledged that in “special circumstances” 38 the two basic principles (namely,the right to present one’s case and the right to a fair and impartial adjudication) could come into conflict. Should that occur, the Tribunal considered that there was a duty “to find a way of bringing [the two rights] into balance”. 39 It warned, however, against any practice of challenging counsel as “a handy alternative to raising a challenge against the tribunal itself”. 40

The Tribunal also sought to narrow the impact of Hvratska by suggesting it was better understood as “an ad hoc sanction for the failure to make proper disclosure in good time than as a holding of more general scope”. 41 Whilst that characterisation does not sit comfortably with the reasoning in Hrvatska , it is unhelpful in any event as the Tribunal’s order would still need to be sourced in an available power.

At this point, it is necessary to point out that, at one level, both these decisions can be confined to legal history as they predate the provisions of the IBA Guidelines on Party Representation. Guideline 5 states that a person should not accept representation of a party after the Arbitral Tribunal has been constituted where there is a conflict of interest between the representative and the arbitrator. Guideline 6 provides that where there is a breach of Guideline 5 “the Arbitral Tribunal may … take measures … including exclusion of the new Party Representative” 42 .

Nonetheless both cases remain relevant, at least to demonstrate the types of relationships that might give rise to an arguable conflict of interest. This is especially so on the facts of Hvratska . The cases also remain relevant on the question of “power” when there is no express power to act. In addition, the Guidelines do not have any mandatory operation, although they are almost universally adopted.

This leads me to the question of whether an arbitral tribunal is vested with inherent powers. For present purposes this question needs to be refined or at least contextualised: does a tribunal whose existence depends upon the agreement of contracting parties have inherent power to make orders to protect its integrity?

The traditional view has been that arbitral tribunals do not have inherent powers. 43 Professor Chester Brown has considered the question in the context of the more formal structures of International Courts and Tribunals, 44 including the International Court of Justice. He observes that, although the existence of inherent powers in those bodies has doctrinal support, their source, nature and scope remains obscure. 45 As he notes, for the most part, the existence of an inherent power is assumed. Although directed to international courts, Professor Brown’s analysis has broader application given that these courts are themselves “the product of the express will of states as expressed in international agreement, and, as such, their jurisdiction is based on the consent of the parties”. 46

The International Court of Justice (ICJ) has no doubt that it is vested with inherent powers. In its 1974 Nuclear Tests decisions 47 the ICJ described itself as a “judicial organ established by the consent of States”. Having given itself that characterisation, it stated that it had inherent powers “in order that its basic judicial functions may be safeguarded” emphasising that it “possesse[d] an inherent jurisdiction enabling it to take such action as may be required, on the one hand to ensure the exercise of its jurisdiction over the merits … and on the other, to provide for the orderly settlement of all matters in dispute”.

A commercial arbitral body cannot aspire to such lofty heights. Nonetheless, arbitral tribunals have long exercised powers on the basis that they have inherent powers. Most notably, in commercial arbitration the “competence-competence” principle (which provides that arbitral tribunals are competent to determine their own competence) is usually conceived of as an inherent power. There may be other approaches. For example, does the fact that parties have entered into an arbitral agreement mean that, as a matter of a good faith bargain, there is an implied contractual obligation to ensure that the two fundamental tenets of arbitration are observed?

Finally, it is reasonable to ask whether arbitration has become so regulated, often in response to situations as arose in Hvratska , that such questions will not, or are unlikely to, arise. The answer is not necessarily straightforward. Take Guideline 26 of the IBA Guidelines on Party Representation, which provides certain remedies for misconduct. Under that Guideline, the Tribunal may admonish the party representative; draw inferences in assessing the evidence or legal arguments advanced by the party representative; make costs orders and take any other appropriate steps to preserve the fairness and integrity of the proceedings. It is arguable that this last provision would include barring counsel from further appearing. However, as already noted, these are guidelines only. Does a guideline in and of itself provide a basis for the exercise of a power where there is no express provision to do so?

Let me finish this section of the paper with one more example. In a Finnish case, 48 counsel for a party was also that party’s main proposed witness, having negotiated the contract the subject of the arbitration. The opposing party objected to counsel’s appearance. The arbitral tribunal ruled that counsel could continue to appear, although his evidence had to be taken before any other evidence so as not to give that party an unfair advantage.

Litigation funding

It was inevitable that third-party funding, now commonplace in litigation, would raise its head in arbitration. It is as inevitable that the presence of a third-party funder in arbitral proceedings will give rise to questions of conflicts of interest. Let me provide three possibilities. By analogy, each of the scenarios to which I refer would fall within the stop light lists in Part II of the IBA Guidelines.

The first and most obvious case of conflict is where there is a direct and undisclosed relationship between the arbitrator and the funder. Depending on the nature of that relationship, this could fall into one or other of the red or orange IBA list.

A second possibility would arise where there is a relationship between the law firm of which the arbitrator is a partner and a third-party funder. Again, by analogy this would fall into the red or orange IBA list. 49 There could be a number of variations on this. What, for instance, if a partner of the arbitrator acts for the third-party funder, for example, as a commercial adviser on the form of funding contracts generally used by the funder or who acts for the funder in litigation or arbitral proceedings to which the funder is a party?

A third is where the litigation funder has been directly involved in the appointment of the particular arbitrator and has appointed this arbitrator on previous occasions. This would fall into cl 3.1.3. and 3.1.4 of the IBA orange list if the arbitrator had been appointed at least twice by the funder in the previous three years.

Whilst these scenarios can be considered by reference to the IBA lists, it is a matter of significant doubt whether a third-party litigation funder is an “affiliate” within those lists 50 . If a litigation funder is not an affiliate, where is the requirement for disclosure either of the funding arrangement or of the relationship between the funder and the arbitrator?

There are other difficulties. Let me return to the first example where the arbitrator has a direct relationship with the funder by the holding of shares in the publicly listed funder company. Should the question of that possible conflict of interest be resolved in the same way as it was in Ebner such that there would be no conflict if the arbitration would have no impact on the value of the shareholding? In that case, one might ask whether there would be a need for disclosure at all.

And what of the commercial intelligence that suggests funders only finance cases with a high likelihood of success? Could it be suggested that this would or could constitute a subtle influence on the arbitrator’s decision? 51 That, of course, would be a difficult one to argue and would not, I suggest, pass the “real danger” test. Nonetheless, it does introduce an interesting backdrop to this ever-expanding form of dispute support service.

I will mention one further matter which is not without significance. Everything I have said thus far on this topic presupposes some form of disclosure or knowledge by other means that a third-party funder is involved. Many third-party funding agreements require confidentiality at least in relation to the terms of the agreement. In the UK there is a voluntary Code of Conduct for Litigation Funders. 52 The Code, however, does not expressly apply to arbitrations.

There does not appear to be any case law involving the types of conflicts to which I have referred in arbitrations involving third-party funders 53 . Although my research on this topic has been limited, it is undoubtedly a “problem in waiting” and parties should be aware of the possible issues that could arise.

Conclusion

Although arbitral institutions and associated bodies have responded to particular problems as they have arisen and will likely do so in the future, the world of arbitration, including international arbitration, will continue to throw up a myriad of issues that will capture the attention of lawyers and academics alike. In particular, conflicts of interest will present a smorgasbord of problems. Attempts at uniform laws, rules and norms have had a large degree of success in the international sphere. Nonetheless, the introduction of s 18A in the International Arbitration Act 1974 (Cth), although arguably welcome, indicates that national bodies are and will be prepared to takes their own course as they see fit. The impact of national bodies being prepared to do so will be a matter for ongoing assessment and interest.