The adoption of universal rules of conduct for adjudicators in international investment disputes (IIDs) (including judges, arbitrators and annulment committee members) has been an elusive goal for decades. This is hardly surprising considering the fact that these decision-makers, just like the disputing parties, come from diverse backgrounds in legal tradition and education, areas of expertise and practice, and professional development. One universal expectation and mandate of all arbitrators is that of neutrality. Namely, IID adjudicators are expected to be independent of outside influence and impartial towards the disputing parties, and the subject matter of the dispute.

Beyond these overarching standards, which are often used interchangeably in practice, depending on the applicable law and rules in each case, there are no universal rules of conduct providing specific norms that must be complied with by international arbitrators in order to uphold the duty of independence and impartiality. One guardrail against conflicts of interest in international arbitration is the ongoing duty for arbitrators to disclose any and all circumstances that may give rise to justifiable doubts as to their independence or impartiality in the eyes of the parties. However, the disclosure obligation is also subject to the same uncertainty in terms of the applicable standards – What exactly should an arbitrable disclose and how to anticipate what the parties may consider problematic? 

Absent binding rules, the IBA Guidelines on Conflicts of Interest in International Arbitration (IBA Guidelines) have served as a compass for parties, arbitrators and institutions alike. In their useful application lists, the IBA Guidelines have categorized the various facts and circumstances that could or should be disclosed [i]. While the IBA Guidelines put the general standards into more specific terms, as a soft law instrument, they were still applied and interpreted within the broader, fragmented system of rules, and ultimately did not lead to international harmonization in practice.

The existing framework governing the standards for the conduct of adjudicators is particularly criticized in the context of investment arbitration. Arbitrators deciding investor-State disputes have been under increased scrutiny in the discussions of governments under the auspices of the UNCITRAL Working group III (ISDS Reform). State representatives have voiced particularconcerns with the practice of double hatting (arbitrators simultaneously acting as counsel and experts), the repeat appointments of arbitrators by the same parties or law firms, and the lack of accountability to the communities that may be affected by the decisions of the tribunal on a particular contested policy of the State.  After extensive discussions and written submissions from States, practitioners and experts, thesecretariats of UNCITRAL and ICSID undertook the ambitious task of developing adraft Code of Conduct that should apply to all investment adjudicators (including arbitrators and judges of a possible standing court). 

To date, three versions of the CoC have beenpublished for comments and modified based on the feedback provided by the States, arbitrators and academia. As this article is being written, do not seem to be any closer to a universal set of binding rules that will govern the conduct and accountability of the arbitrators. One notabletrend is the consistent movement towards flexible compromise solutions in each new version, leaving more to the discretion of the parties, which is in contrast with the initial high standards and ambitions to adopt a strict and harmonized set of rules.

This article explores the evolution of the draft CoC through the existing three versions in order to determine whether the lessons from the preceding process are indicators of the likelihood of the successful adoption of a universal set of rules for arbitrator conduct in the future.

The Evolution of the Terminology of the Draft CoC

The three versions of the draft CoC were developed and published between May of 2020 and October 2021. The first version waspreceded by several rounds ofdiscussions in the UNCITRAL WGIII dedicated to this topic in particular in which the delegates recognized the issue of the arbitrator’s independence and impartiality as an area of concern that merits reform at the international level. Considering the serious criticisms of the mechanisms enforcing the existing standards, the first version of the draft CoC reflected a very strict approach, leaving very little to the discretion of the parties. Thus, the arbitrators were subject to a strict and expansive disclosure obligation, as well as the prohibition of double hatting and other high standards of conduct and performance. The comments from the delegations, arbitrators and other interested commentators pointed out the possible practical difficulties with enforcing the strict provisions at the international level, and the potential harmful effect on the diversity of the pool of available adjudicators who could be appointed under such terms.

On the basis of these comments and the ongoing discussions within the UNCITRAL WGIII, thesecond and third versions of the CoC contained much more moderated and flexible language, providing more specificity on certain definitions, but less strict on the standards applicable to the adjudicators themselves. As a result, and as it will be demonstrated in some notable examples below, the draft CoC is moving away from the envisioned rigid set of universal and binding rules with each new version that is released. 

The first draft CoC applied to adjudicators in investor-State dispute settlement (ISDS), broadlydefined as  “a mechanism to resolve disputes involving a foreign investor and a State or a Regional Economic Integration Organization (REIO), or any constituent subdivision of the State or an agency of the State or the REIO, whether arising under an investment treaty, domestic law or an agreement by the parties to the dispute”. The second version narrowed the scope of application to international investment disputes (IID)  bydefining it as “International Investment Dispute (IID) means a dispute arising pursuant to the investment promotion and protection provisions in an international treaty.

This particular change in approach is notable because it already narrows the scope of application of the draft CoC, to the exclusion of IID proceedings based on contracts or domestic investment protection laws. Such a solution runs against the goal of ensuring harmonized rules for all Investor-State disputes and opens the door to further fragmentation, even after the CoC is adopted. Even if the disputing parties in contract-based or statute-based ISDS cases opt to apply the CoC, this will still be subject to case-by-case determinations and not necessarily binding in nature.

On the other hand, the move from ISDS to IID opens the door to the application of the final CoC to State-to-State arbitration. Considering the different nature and dynamics in State-to-State disputes, it is uncertain to what extent the IID CoC may beadequate or desirable in this context. Furthermore, and as indicated by the United Kingdom’s comments to the second version of the draft CoC, State-to-State disputes are not within the mandate of the WGIII. Nevertheless, the change in scope of application was reflected in the title of the second draft of the CoC and retained in the third version as well.

The Evolution of the Provisions on Double Hatting

In the preliminary discussions preceding the development of the first draft CoC, a vast majority of the delegations at the UNCITRAL WGIII expressed seriousconcerns with the practice of adjudicators accepting appointments as counsel and experts, while acting as arbitrators in other investment proceedings. This practice was recognized as a common occurrence in ISDS which hinders the adjudicators’ ability to remain impartial if they are acting in different roles in potentially related disputes, representing different interests [ii]. The issue of repeat appointments was also recognized as a significant issue that has the potential of affecting the independence and impartiality of the adjudicators or creating the perception of bias on the side of the parties.

Since these issues are considered particularly problematic in ISDS cases, they are not explicitly addressed by the existing rule or soft law instruments. Therefore, they were given particular attention in the discussions leading up to the development of the draft CoC, and in all three versions of the CoC itself.

The first draft CoC provided a strict and narrow provision on double hatting requiring adjudicators to: “refrain from” accepting appointments in other roles, with the option to make the prohibition limited in time or the subject matter. This reflected the attitude of the delegations at the WGIII, most of whom were in favour of a total prohibition of double hatting. However, following the comments from the delegations and other relevant stakeholders (including arbitrators) on the first draft CoC, it became apparent that the views were more nuanced and that an outright ban on double hatting would create significant issues to experienced and up-and-coming adjudicators alike [iii]. Therefore, in thesecond draft CoC, the provision was modified to prohibit double hatting unless the parties agree otherwise and provided a further option to limit the prohibition to cases involving the same subject matter and parties in the dispute.

Thethird version of the CoC provides three options to address double hatting that encompasses-

1. the total prohibition from version one,

2. the modified prohibition from versions 2 and 3 added a third option

3. to allow double hatting with the obligation of full disclosure of any simultaneous appointments during the course of the IID proceedings, in cases “involving the same or related parties, the same measures, or [substantially] the same legal issues as are at issue in the IID.

The prohibition of double hatting in Options 1 and 2 are opt-out provisions (the ban on double hatting will stand unless the disputing parties agree otherwise), while option 3 requires the full disclosure of any concurrent appointments that the parties can accept and proceed with the appointment, thus waiving the right to subsequently challenge the arbitrator on the grounds that were disclosed. Therefore, the provisions on double hatting have gone from a prohibition (with some possible qualifications) to a more targeted approach banning only the types of concurrent appointments that are closely related to the parties or the underlying dispute, and thus present a risk of real or apparent of conflict of interest. Furthermore, the disputing parties are able to opt-out of this provision and agree to appoint arbitrators who are acting in other concurrent positions. 

This is a more pragmatic approach, in contrast with the vocal calls for the abolition of double-hatting in IID voiced during the preliminary discussions in the WGIII, and in the comments tothe second version of the draft CoC. While there are still governmentsadvocating a complete ban of double-hatting, other States, practitioners and professional organizations have cautioned against the numerous negative implications of such an approach. The biggest concern is the prospect of shutting the door for new and diverse arbitrators from pursuing appointments in IID, if they are unable to place their practice as counsel on halt. In addition to being harmful to diversity, this would also perpetuate the repeat appointments of established players in international arbitration, which is already identified as a separateconcern related to the independence and impartiality of adjudicators.

There appears to be an emerging consensus preference for a middle ground solution, outlined in option 2 that provides for a targeted prohibition of concurrent appointments as counsel or expert in other IID cases involving the same contested “measure, substantially same legal issues, the same parties, affiliates disputing parties or its subsidiary, affiliate, parent entity, State agency, or State-owned enterprise or the same treaty”. All three options remain open for discussion at the upcoming WGIII session and it remains to be seen if the double hatting provision will get its final contours in the near future.

The Evolution of the Disclosure Provisions

The disclosure of circumstances and relationships that may give rise to doubts of independence and impartiality is not new in international arbitration. Most national laws and institutional rules place an ongoing obligation of disclosure on arbitrators to ensure that the parties and arbitrators are aware of any potential conflict of interest and that such issues are addressed in a timely manner, to protect the integrity of the tribunal and the final award. TheIBA Guidelines also encourage arbitrators to err on the side of disclosure whenever in doubt.

The delegations in the WGIII have recognized disclosures as a key component of ensuring independence and impartiality of the adjudicators in ISDS, and there were many proponents of a maximum duty of disclosure, including matters that are otherwise in the public domain. In light of this high threshold, thefirst version of the draft CoC provided a long list of items the candidates and appointed adjudicators are expected to disclose, including “[A]ll ISDS [and other [international] arbitration] cases in which the candidate or adjudicator has been or is currently involved as counsel, arbitrator, annulment committee member, expert, [conciliator and mediator]; and (d) A list of all publications by the adjudicator or candidate [and their relevant public speeches].” The sweeping nature of this proposed provision prompted a strongreaction from some States and prominent arbitrators alike. All stakeholders recognized the disproportionate burden the disclosure obligation placed on the adjudicators, especially those with decades of experience. Some arbitrators went as far as to call the provisions offensive and discriminatory against professional arbitrators.

In light of the numerous comments received on this provision, the second draft was much more measured, excluding the requirement to disclose publications and speeches, and narrowing thedisclosure of previous appointments to “[A]ll 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. In the most recent version, the reference to “non-IID” proceedings in the bracketed text was replaced with “[and all related proceedings]”, further narrowing thedisclosure requirement to cases that may raise justifiable doubts in the eyes of the parties as to the independence and impartiality of the candidate or appointed adjudicator. This approach is meant to ensure the proper disclosures, while at the same time curtailing any frivolous or opportunistic challenges of adjudicators for the failure to disclose a broad range of cases that may not be objectively relevant, but nevertheless may be considered a breach of the CoC.  

This is reinforced by the notable inclusion of explicit language stating that neither the fact of non-disclosure or disclosure does not automatically constitute a breach of the CoC.  Furthermore, the drafters included a provision allowing for the waiver of any conflict identified following a disclosure by a candidate or appointed adjudicator. Such provisions should encourage the pro-disclosure attitude and help filter disclosure-related grounds for challenges.


There is no doubt that universal, precise and binding rules of conduct for adjudicators in investment disputes are necessary to protect the integrity of the decision-makers and the legitimacy of the final decisions – both in the eyes of the parties and the public at large. While the idea of a unified set of rules is appealing, putting the desired standards into strict provisions has proven to be very difficult to date, which is demonstrated by the most recent valiant efforts of UNCITRAL and ICSID to develop a draft code of conduct for adjudicators in IID. Although they started from a common baseline of expectations and parameters provided by the delegates in the WGIII, seeing the proposed rules “on paper” opened up more doors for discussion instead of leading to a conclusion on the desired language of the draft CoC.

The draft CoC has evolved both in its scope of application and the stringency of the proposed terms, in response to the identified possible complications that may arise in practice from the strict application of the intended standards. Interestingly enough, it was not only the arbitrators and other practitioners that raised concerns about the possible negative effects of imposing burdensome obligations on adjudicators in IID. States also recognized that broad obligations may also limit the pool of available candidates and increase the rate of challenges against adjudicators as parties may find additional grounds for a challenge in the extensive obligations imposed by the CoC.

Therefore, despite the ambitious, and justified aspiration towards imposing the highest standard of conduct for adjudicators in IID, the CoC is evolving into a more pragmatic and moderate set of rules [iv]. Considering the complex and fragmented IID framework, a uniform solution that could apply across the board escapes the imagination, but the draft CoC does address some of the unique challenges that arise in investment disputes for the first time.

Until there are better conditions for a more unified approach, the CoC should help to establish a baseline of expectations, both on the side of the arbitrators and the disputing parties, leaving less room for discretion. As the discussions on the draft CoC continue in the months and perhaps years to come, it remains to be seen what solutions will emerge out of this evolving instrument. 

About the Author

Ms. Fahira Brodlija is an International Commercial and Investment Arbitrator and is affiliated to the International University of Sarajevo.


The views and opinions expressed by the authors are personal.

Editorial Team

Managing Editor: Naman Anand

Editors-in-Chief: Aakaansha Arya and Jhalak Srivastav

Associate Editor: Kshitij Pandey

Junior Editor: CH Sriniwas

Preferred Method of Citation

Fahira Brodilija “The Draft Code of Conduct for Adjudicators in International Investment Disputes – Evolution or Revolution?” (9 November, 2021)



[i] The IBA Guidelines provide 1. seven general Standards and a Practical Application of the General standards. The practical Application is divided into a list of situations that may occur in practice, categorized into the Red (waivable and non-waivable), Orange and Green lists. The Red List refers to circumstances that are likely to give rise to doubts of independence and impartiality of the arbitrators and thus must/should be disclosed. The Orange List refers to situations that may or give rise to such doubts in the eyes of the parties and thus their disclosure is advisable. The Green Last provides examples of circumstances that are not likely to give rise to conflicts of interest.

[ii] Malcolm Langford, Daniel Behn, and Runar Hilleren Lie, ‘The Ethics and Empirics of Double Hatting’ 6:7 ESIL Reflection (2017).

[iii] Some commentators favoring a limited prohibition noted that this approach will raise fewer obstacles to entry for a large number of qualified adjudicators, thus encouraging diversity; it will protect the autonomy of the will of the parties; and it would reduce the likelihood of repeat appointments.

[iv] Following two rounds of comments, Version 3 of the Draft Code of Conduct seeks to create a “balanced, realistic and workable” document. It seeks to address practical issues that may arise, to avoid unknown or ambiguous concepts, and to use concise drafting. Version 3 of the Draft Code of Conduct.

error: Content is protected !!