Systemic Bias and the Institution of International Arbitration: a New Approach to Arbitral Decision-making

The popularity and scope of authority of international arbitral tribunals has greatly expanded in the past forty years. International tribunals are now considered capable of deciding a wide range of disputes, even disputes implicating national public policy, which were previously falling in the exclusive domain of State courts. In the area of commercial arbitration, for example, it is now accepted that tribunals have authority to determine not only commercial claims pertaining to the formation, interpretation and performance of commercial contracts, but also statutory claims that may have crucial social implications,¹ such as anti-trust claims, tax claims or claims arising out of securities transactions.²

Meanwhile, in the area of investment treaty arbitration, a significant number of international investment treaties have adopted arbitration as the preferred model of dispute resolution. Accordingly, international arbitral tribunals are now regularly reviewing investor claims concerning government measures, including financial and environmental measures, which would normally fall within the regulatory sovereignty of the host nation.³ Even disputes on sovereign debt bonds are now submitted before international tribunals.⁴

The remarkable rise of international arbitration has challenged the traditional concept of public justice to the point that, very recently, English courts allowed the referral of all aspects of a family dispute, including custody of and contact with the children, to arbitration noting that ‘it is up to parents to agree how their children should be brought up and, if they cannot agree, they should be entitled to choose how their disagreement should be resolved without state intervention’.⁵

However as international arbitral tribunals are increasingly becoming more popular, certain aspects of private decision-making come under scrutiny, often under intense criticism. Although criticism against investment and commercial arbitration arises in different fashion and volume, critical voices coming from both the public domain and academia raise legitimate questions, such as: who are these individuals that act as arbitrators and have the power to decide issues with important implications on national public policy and sovereignty? How do arbitrators decide?

Some years ago, The New York Times notoriously described investment arbitration proceedings under the North American Free Trade Agreement (NAFTA) in the following terms:

Their meetings are secret. Their members are generally unknown. The decisions they reach need not be fully disclosed. Yet the way a small number of international tribunals handles disputes between investors and foreign governments has led to national laws being revoked, justice systems questioned and environmental regulations challenged.⁶

Many scholars raise issues of impartiality of arbitrators, suggesting that arbitral decision-making is biased, apparently favouring investors and multinationals against states and weaker parties such as consumers and employees.⁷ Some countries have cited apparent bias of international arbitration as a reason to withdraw from the 1965 Convention on the Settlement of Investment Disputes between States and Nationals of Other States.⁸

More recently, the 2013 UNCTAD Report on Recent Developments in Investor-State Dispute Settlement stated ‘the continuing trend of investors challenging generally applicable public policies, contradictory decisions issued by tribunals, an increasing number of dissenting opinions, concerns about arbitrators’ potential conflicts of interest all illustrate the problems inherent in the system [of international arbitration]’.⁹

Naturally, thus, arbitral decision-making and the question of whether arbitration is a biased adjudicatory system has become one of the most critical subjects of discussion currently. Interestingly, accusations of bias in arbitral decision-making have prompted different reactions from arbitration practice and arbitration scholarship.

For arbitration practice, the issue of bias in arbitral decision-making is primarily associated with the ethics of the individuals acting as arbitrators. Accordingly, the response of arbitration regulatory bodies has been to strengthen the legal framework concerning the duty of arbitrators to disclose any issue that may give rise to appearance of bias. As a result, there is currently a variety of arbitration laws and in particular institutional codes of ethics regulating the conduct of individual arbitrators in minute detail.¹⁰ Potential arbitrators are advised not to ‘regularly spend considerable time’ with counsel for matters ‘unrelated to professional work commitments or the activities of professional associations or social organizations’.¹¹ In the wake of the information technology revolution, potential arbitrators are further advised by peers to avoid social media and electronic social networks altogether, or limit their friends on Facebook or LinkedIn to truly social contacts.¹² Meanwhile, national courts have been trying to refine the appropriate standards of apparent bias of individual arbitrators arguing whether the right test is that of ‘reasonable appearance of bias’ or ‘reasonable suspicion’ of bias or a ‘real danger’ or ‘real possibility’ of bias.¹³

Overall, the main focus of arbitration practitioners, courts and professional associations has been on whether individual conduct of arbitrators may raise issues of apparent bias. By contrast, for arbitration practice, implicit bias, such as subconscious, cognitive or cultural bias, is extremely difficult to prove and should not be the concern of law. As the English Court of Appeal characteristically noted in Locabail ‘the law does not countenance the questioning of a judge about extraneous influences affecting his mind’.¹⁴ Yet, increased regulation of the conduct of individual arbitrators has failed to appease criticism against the integrity of international arbitration, mainly because it has failed to address issues of implicit bias associated with the culture and the institution of arbitration.

Meanwhile, arbitration scholarship took a different approach to the challenges presented by arbitral decision-making. Influenced by the trend of legal empiricism and the scientific studies on national judiciaries, many scholars embarked on empirical studies on judicial behaviour of arbitrators. Using a variety of scientific methods such as statistics, interviews, and experimental techniques, these studies aimed to capture bias and assess its impact on arbitral decision-making. Yet, and despite their empirical approach, which suggests objective measurements and scientific conclusions, studies on arbitral judicial behaviour have arrived at diverging findings, failing to provide a comprehensive explanation of arbitral decision-making.¹⁵

This failure owes mainly to two reasons: first, the intrinsic methodological difficulties associated with the use of empirical methods for capturing bias, due to the fact that bias is basically an unobservable phenomenon. Secondly, the fascination of these studies with behavioural or attitudinal theories on decision-making. Evidently influenced by similar works on US national judges, these studies depart from the behavioural assumption that arbitral decision-making is driven almost exclusively by extra-legal factors, mainly policy preferences or financial incentives of arbitrators. Based on this theoretical assumption, most of these studies take the same approach: they employ empirical techniques in order to analyse award outcomes and identify voting patterns. According to attitudinal theories, voting patterns betray the policy preferences or financial incentives of arbitrators, and therefore explain how arbitrators decide.

The attitudinal model and the use of empirical methods of analysis have undoubtedly provided useful insight in arbitral decision-making. Yet, they have clear limitations. The correlation between award outcomes, voting patterns and judicial attitude is simplistically linear.¹⁶ Thus, an attitudinal concept of judicial behaviour unwarrantedly reduces arbitral decision-making to a policy or financially driven exercise.

Eventually, as the article argues, such limited theoretical approach fails to provide a comprehensive description of arbitral decision-making, as it leaves important determinants outside the scope of its analysis. One of the most important determinants that the attitudinal theory fails to take into account is the influence of the broader institutional context within which decisions are taken. The critical role of institutions in decision-making has long been acknowledged in legal scholarship. For institutional theories, judicial behaviour and values are shaped by a number of important institutional processes such as the method of selection and appointment of decision makers, their legal and professional training, their distinct professional role and tenure status, their financial and professional dependence on the institution and their sense of obligation towards it, the existence of stare decisis.

Taking an institutional approach, and drawing heavily on studies on national and international judiciaries, the article examines some institutional aspects of international arbitration. As is argued, the multilateral and fluid processes for selecting arbitrators, as well as the lack of tenured arbitrators and the lack of stare decisis underpin the pluralistic, diverse and democratic potential of international arbitration. This is particularly the case if international arbitral tribunals are compared to the highly structured and centrally controlled paradigm of national and international courts.

Of course, a comprehensive study of the institution of international arbitration needs to look beyond the surface of the procedural design, and examine whether international arbitration has developed implicit or informal institutional structures that may also affect arbitral...