The Draft Code of Conduct for Adjudicators in Investor-State Dispute Settlement: An Important Step Forward in the Reform Process?

On May 1, 2020, the secretariats of ICSID and UNCITRAL released the first draft of the Code of Conduct for Adjudicators in Investor-State Dispute Settlement (ISDS). States, international organisations and other stakeholders have until October 15 to send comments on the draft.

I had the privilege to work extensively on the draft Code as a scholar in residence at ICSID from September 2019 to March 2020, while on Sabbatical leave from my own law school. In this short contribution, I first put the code in the larger context of the ISDS reform process, then explain the main provisions and general content of the draft, and finally provide some comments and ways forward.

The Code in Context

Readers of this blog already know that ISDS is undergoing a profound process of reform (which has been the subject of several excellent posts in this blog, see generally here). The process of ISDS reform is taking place within Working Group III (WGIII) of UNCITRAL. States delegates have requested UNCITRAL to explore numerous topics for potential reform. The reform process includes proposals for both bold systemic changes, such as the creation of permanent multilateral court for investment disputes or of an appeal mechanism, and for small incremental changes that would address only specific concerns (for a helpful explanation see here). In the edifice that graphically represents the Multilateral ISDS Reform Agenda, the Code of Conduct for Adjudicators is both part of the foundations of the building and one of the substantive issues identified by stakeholders as needed to be discussed when addressing reform.

Indeed, it is peculiar that a Code of Conduct for Adjudicators does not already exist, given the complexity of the disputes addressed by ISDS proceedings and the complexity of ISDS as a dispute resolution mechanism, which involves parties, counsel and adjudicators from diverse jurisdictions, admitted to the practice of law under diverse rules and who have undertaken different legal training and education (for some background see here). Yet, as of now, there are no common ethical rules and standards of conduct applicable generally in ISDS proceedings, and the reform process seems to be a good time to finally discuss and introduce a common Code.

In 2015 and 2016, the UNCITRAL Commission began considering proposals for future work on a code of conduct for arbitrators in investment arbitration. In 2017, the Commission then considered more specifically future work on ethics in international arbitration. Delegates at WGIII expressed broad agreement on the importance of a code of conduct for ISDS tribunal members, especially as a measure to enhance confidence in the independence and impartiality of ISDS adjudicators. After preparing some background work and collecting Member States’ comments, delegates at WGIII’s thirty-eighth session in October 2019 suggested that the Secretariats of ICSID and UNCITRAL cooperate in preparing model provisions for a code of conduct for adjudicators. For this purpose, several important issues were highlighted as possible content for the code, including independence and impartiality of adjudicators, issues of integrity, diligence and efficiency, principles related to confidentiality and competence, and obligations related to disclosure.

The result of this common effort is now available on the websites of the two secretariats (ICSID and UNCITRAL) for review and comments.

What’s in the Code?

The draft Code addresses many key ethical and contested issues identified by WGIII delegates and more generally by ISDS’s critics, and provides policy makers with numerous choices on how to best regulate adjudicators’ behaviour through a Code of Conduct.

The draft Code consists of 12 articles, with helpful commentaries explaining the rationale and context for each provision. Overall, the Code can be divided into three sections. Articles 1-3 provide a first introductory section. Article 1 defines essential terms, Article 2 addresses the applicability of the Code and Article 3 offers an overview of adjudicators’ obligations. A second section, which includes Articles 4-11, provides more substantive obligations: Articles 4 to 9 expand on the provisions enumerated in Article 3, and Articles 10 and 11 regulate pre-appointment interviews and fees. Finally, in a third section, Article 12 addresses the fundamental issue of enforcement of the Code.

The Code applies to all kinds of adjudicators, including arbitrators, ad hoc committee members, candidates to become adjudicators, and judges and appeal judges in (possibly future) permanent bodies or appeal mechanisms. This drafting guarantees that the Code can be applied in future disputes, even if systemic changes are made during the reform process. The Code also applies to adjudicators’ assistants. Separate codification will be required to regulate the conduct of counsel and other ISDS actors, possibly at a later stage.

Substantively, the Code contains many notable provisions.

Several provisions relate to (mostly) non-controversial and essential ethical requirements and are broadly consistent with those found in existing codes of conduct such as those of CETA and CPTPP (for more, see here). These include the duties of integrity, fairness, competence, diligence, civility and efficiency and the obligation to respect confidentiality. It is important for these duties, which are largely supported by different stakeholders, to be included and spelled out in the Code. The Code also includes the key duty of adjudicators to be at all times independent and impartial and to avoid conflicts of interest. The inclusion of these principles in the Code is also non-controversial as they are widely shared principles, though differences may exist as to specific definitions and the elaboration of particular principles.

The draft also addresses three core and sensitive issues that have generated much debate in the ISDS critique and have been highlighted as problematic in the context of ISDS reform: repeat appointments, issue conflict and double hatting. Provisions for these issues are drafted with bracketed options so that States delegates, who requested their inclusion in the draft, can better discuss the many existing regulatory options. We should expect substantial debate on each of the options when the time for discussions and decisions comes.

Article 5 addresses repeat appointments and issue conflict as they relate to the possible existence of conflicts of interest. Repeat appointments raise the concern that an adjudicator who is repeatedly appointed by the same counsel, client, party or ‘side’ may develop a dependence or affinity with the nominating party, or become biased in its favour. As bias may be unconscious, the concern is difficult to address. Repeat appointments raise complex definitional and policy matters, also in relation to concerns related to the lack of diversity in ISDS and a barrier to entry to new players. Article 5 also addresses issue conflict, another complex topic that has attracted criticism. Issue conflict is based on the apprehension that an adjudicator who has expressed views in publications, speeches or other awards and judgments has settled views on a legal issue and thus may not have an open mind.

In both cases, the Code requires extensive disclosure as the main regulatory tool. Adjudicators must be pro-active and must make a reasonable effort to become aware of interests, relationships or matters that can create a conflict that could be perceived as affecting their independence and impartiality (for a nuanced take on conflicts of interest, see here). Adjudicators also have a continuous duty of disclosure and should err in favour of disclosure in case of doubt. Yet disclosures that would be trivial are not required. Article 5 is drafted so as to give several choices to policy-makers on how extensive disclosure obligations should be in the final version of the Code. For example, disclosure could be limited (or not) to activities that occurred during a specified number of prior years, it can be extended to relationships with subsidiaries, parent companies and agencies related to the parties, as well as to any third party that has a direct or indirect financial interest in the outcome of the case. As drafted, the provision also envisages the possible disclosure of the adjudicator’s participation in all other ISDS and/or other international proceedings or related domestic arbitrations. The rationale for such comprehensive disclosure requirement – which includes work as counsel, adjudicator, expert or other function in other international matters, as well as the disclosure of lists of publication and public speeches – is that it would allow for a full assessment of any possible conflicts of interest of the adjudicators by the parties, so that parties can be fully satisfied with their choice, or, alternatively, raise their concerns and decide to challenge the adjudicator.

Another significant issue addressed by the draft Code is double hatting, which has also attracted significant criticism. Double hatting refers to the practice of an adjudicator to simultaneously act as (and thus wear the hats of) counsel, expert, adjudicator or in other roles in other ISDS or other international proceedings. Double hatting is not a technical term (which makes it harder to regulate as there is no unanimously agreed definition of what double hatting is) so that Article 6 is entitled “Limit on Multiple Roles.” Article 6 states that:

Adjudicators shall [refrain from acting]/[disclose that they act] as counsel, expert witness, judge, agent or in any other relevant role at the same time as they are [within X years of] acting on matters that involve the same parties, [the same facts] [and/or] [the same treaty].

Regulating double hatting is complex for many reasons, including the difficulty of agreeing to a common definition, the frequency of the practice, and the divergent views on how to best regulate it (for an interesting prospective see here). Article 6 is formulated as to give policy makers a range of options from creating a complete ban on double hatting, including any work as counsel, expert, agent and other possible roles, to only requiring disclosure of any work on other cases. The draft also includes the possibility of introducing a time element for disclosure, so that only involvement in cases in the prior “X” years should be disclosed. The draft also provides a range of options to define what kinds of matters may lead to a double hatting, for example those involving the same parties, facts, or treaty. Given the interest in this issue, Article 6 will surely be extensively debated among delegates at UNCITRAL.

Where to now?

After the written commentary phase has concluded in October 2020, the Code will be discussed at future meetings of WG III. Because the reform agenda is so full, and because of the disruption brought about by Covid19, a date for discussion of the Code is not yet public.

Though there is broad support for the approval of a Code of Conduct generally, we can certainly expect substantive discussion on details and several specific issues.

First, reaching agreement on sensitive issues such as double hatting, multiple appointments and issue conflict will not be easy. There is no agreement on what some of the terms mean, and how to regulate them. Because so many possible regulatory choices exist, we can expect protracted discussion on the advantages and disadvantages of each. Irrespective of what the conclusion will be, engaging in the discussion will be important and will certainly prove fruitful for the advancement of reform. Yet, it is also important that an agreement is reached and that these practices, which have been the subject of much analysis, are regulated once and for all. Ideally, clear and mandatory provisions that require full disclosure and that qualify and limit the pervasiveness of double hatting will be adopted.

Second, any discussion on the adoption of the Code will also require consideration of other related policy issues. Finding the right balance between interlinked ethical priorities, concerns over unconscious bias and appearance of bias, the interest in enhancing diversity, party’s autonomy to make appointments and the ISDS reform process generally all require in-depth discussion and foresight. In this context, a successful reform process should take into consideration ways to enhance diversity and ensuring a system based that enhance and support the rule of law.

Finally, but certainly not last, enforcement and implementation remain key issues for the success of the Code. Article 12 relies on voluntary compliance by adjudicators. Enforcement is envisaged through challenges procedures in each arbitral institution, which are different from one another. While these mechanisms take into consideration the status quo, they will likely not be sufficient, and more and diverse mechanisms can be established. For example, the creation of a centralised and common enforcement mechanism might be considered. Much will also depend on how the Code is implemented. Will the Code be incorporated in a new, multilateral Treaty, following the example of the Mauritius Transparency Convention? Can it be included as an annex to existing investment treaties? Will it be part of arbitration rules? (for more on implementation see here). How the Code is implemented and enforced will certainly have an effect on its success. Moreover, the creation of a permanent court, other new institutions, or an enforcement centre will affect available options.

So, to go back to the question asked in the title of this post: yes, the publication of the draft Code of Conduct is an important step forward in the ISDS reform process. The devil, however, will be in the details and what states delegates agree on when discussions and negotiations really begin.