Time to hang up one’s hat?
Double-hatting in international arbitration
This paper assesses the concept and impact of double hatting, the practice by which one individual acts simultaneously as an international arbitrator and as a counsel in separate investment arbitration proceedings. The paper lists the criticisms that double hatting most commonly faces and attempts to distinguish fact from fiction by assessing the validity of such claims. It then touches upon important topics frequently discussed in conjunction with double hatting, such as diversity among arbitrators and the general state of investment arbitration. Lastly, arguing firmly against a blanket ban, the paper proposes a range of more nuanced solutions, assessing pre-existing suggestions as well as proposing new ideas.
A. THE CONCEPT
When discussing the history of double hatting, one of the most frequent mentions made is to a speech given by Philippe Sands QC in 2009. In this particular speech at the International Bar Association (IBA)’s annual conference in Madrid, Mr Sands criticised several issues regarding investor-state dispute settlements (ISDS), including the practice of double hatting, which he labelled as being “deplorable” and in relation to which he encouraged an outright ban.
Whilst his stance certainly shone a spotlight on the issue and resulted in closer scrutiny, both the concept and the nomenclature, ‘double hatting’, or ‘dual hatting’, have been in existence since the early 2000s.[1] As early as 2003, Sands and Mackenzie raised similar concerns to those reiterated by him in 2009.[2] Whilst the precise reason for the growth of double hatting since then is unclear, its impact is not: in a recent study[3] of investment arbitration, it was found that “only 42% (450) of ISDS cases did not include any double hatters”.[4] Furthermore, in one fifth of the cases, both counsel and arbitrators had on another ‘hat’ elsewhere: “double hatting-squared”.[5]
Whilst double hatting is generally defined as the practice of arbitrators and legal counsel simultaneously or subsequently taking on both roles in the context of ISDS cases, the exact details are not widely agreed upon. There is no overarching definition of the degree of required overlap in terms of subject matter or institution, nor as to the temporal proximity in cases where the ‘hats’ are not ‘worn’ concurrently.
The ICSID and UNCITRAL’s Draft Code of Conduct for Adjudicators in Investor-State Dispute Settlement (the Draft Code of Conduct),[6] which recently did much to pique the interest of double hat debaters, stated that for some, any concurrent representation creates a conflict of interest and should therefore be prohibited, whilst for others, double hatting is “problematic only in circumstances, where the facts or parties are related”.[7] That is to say that “double hatting does not pose, in theory, an immediate legitimacy problem.”[8]
One can infer from this that double hatting is thus a neutral term, not denoting a conflict per se, and it cannot be said that “[t]he very definition of double hatting conflicts with the idea of impartiality.”[9] This paper shall therefore define double hatting widely, i.e. to include simultaneous appointments regarding the same subject or under the rules of the same institution but also appointments in close temporal proximity, in front of different institutions and regarding different issues.
The precise definition of double hatting is further muddied by the fact that many scholars, as well as the Draft Code of Conduct, refer to the concept as encompassing roles as arbitrator and legal counsel.[10] Other commentators, however, additionally include expert witnesses[11] and even tribunal secretaries within the concept.[12] It is not always clear, in such scenarios, whether the author seeks to assert that all three of these roles may be exercised simultaneously, in conflict with the term “double” or “dual”, or whether they simply refer to a wider pool from which two can be selected. Given this unclarity, this paper shall mainly refer to double hatting in the traditional ‘dual’ sense by including under it the roles of the arbitrator and legal counsel.
B. THE ISSUES
I. The criticism
The most common criticism of double hatting is the issues it can raise in relation to transparency and impartiality, either actual or apparent. Perhaps US Senator Elizabeth Warren summed up critics’ arguments most succinctly when she referred to the incongruity of the “gigantic fines” under ISDS without “independent judges”, questioning how lawyers “looking to maintain or attract high-paying corporate clients” could rule against those exact corporations.[13]
More moderate viewpoints see “double hatting” as “creating an appearance of impropriety that adds to ISDS’s ill odour”.[14] Such seeds of doubt, whether founded or unfounded, are problematic, given that “[i]mpartiality is a fundamental characteristic of arbitration”.[15] At present, however, “the safeguards in place do not preclude reasonable apprehension of bias in regard to double-hatters”,[16] with one commentator stating that “barbers and taxidermists are subject to far greater regulation than [arbitrators].”[17]
The issue is that no matter how little bias exists in practice, the perceptions of many, as reflected by the words of Senator Warren, are an issue in themselves. In this regard, reference is frequently made to the IBA Guidelines on Conflicts of Interest in International Arbitration (the IBA Guidelines), which preclude arbitrators from serving under circumstances, either before or during the appointment, which could give rise to a conflict of interest. This includes circumstances that could give rise to ‘justifiable doubts’ about a conflict of interest from a reasonable third person’s point of view. Under the IBA Guidelines, a doubt is justifiable where “a reasonable and informed third party would reach the conclusion that there was a likelihood that the arbitrator may be influenced by factors other than the merits.”[18] The definition is thus fairly wide and means that even where judges are unbiased, they would need to “be able to convince a reasonable observer that such role bifurcation was maintained.”[19]
However, if leading arbitrators such as Mr Sands and Thomas Buergenthal have doubts about double hatting, “it may be likely that external observers are even more likely to conclude that the arbitrator may be influenced by factors other than the merits of the case.”[20] Furthermore, whilst the IBA Guidelines are not legally binding, they are nonetheless important in light of the public nature of ISDS, given that “conflicts of interest between members of the tribunal and counsel engender a lack of confidence in the investment arbitration system.”[21]
In Bank of Montreal v. Brown[22], the court removed an arbitrator for “acting as counsel against one of the parties concerning an unrelated dispute in a different field of the law”.[23] The court stated that “it may well be that [the arbitrator] considered that he could dissociate the two roles in his mind and that his state of mind was not such as to preclude him of being impartial. But this is irrelevant, as a reasonable apprehension of bias must not be confused with actual bias.”[24]
Some arbitrators do not view double hatting as raising ethical concerns, “particularly where cases do not involve common issues.”[25] Perhaps not exclusively relying on arbitrators’ perception of their ability to be unbiased is wise, however, given that arbitrators may be acting in “[unconscious] disregard of their ethical obligations”.[26] Indeed, one commentator wondered whether arbitrators are “capable of such extraordinary compartmentalisation”[27] where conflicting roles are served concurrently or in close temporal proximity.
The existing self-regulation and high ethical standards which arbitrators must hold themselves to are deemed insufficient in the minds of those critics, who believe more drastic action is required.
II. The concrete conflicts
In a 2010 New York Law Journal publication, the authors divided what critics believe are the main issues with double hatting into two parts: ‘club’ and ‘issue conflicts’. The ‘club conflict’ relates to the fact that there is a ‘revolving door in arbitration’, i.e. many individuals serve in different roles subsequentially, meaning that arbitrators and counsel are likely to appear in front of one another in different roles. It also means that they are reliant on one another for appointments. Parties may thus be apprehensive about double hatters’ treatment of one another, lest their positions be reversed in future. There is an ingrained fear of manus manum lavat, or ‘you scratch my back and I scratch yours”.[28] Empirical studies show the existence of such a ‘club’, specifically in relation to twenty-five frequently appointed ‘power brokers’.[29]
The second issue with double hatting relates to ‘issue conflicts’, which Mr Sands has deemed as “imperilling the entire system of investment arbitration”. [30] Issue conflicts concern an “actual or apparent bias on the part of the arbitrator stemming from (their) previously expressed views on a question that goes to the very outcome of the case to be decided,”[31] i.e. conflict arises from arbitrators’ relationship to the subject matter of the dispute. Other commentators define the issue as additionally encompassing decisions on awards where a co-counsel sat as arbitrator.[32]
‘Issue conflict’ are seen to be particularly problematic in the context of investment arbitration as these frequently concern bilateral investment treaties (BITs), in which there are only a limited number of provisions, leading to similar issues being adjudicated.[33] Where an arbitrator issues a decision, they may be a counsel in relation to the same matter in future. Alternatively, they may have won a claim as counsel and seek to rely on this as an arbitrator.
The problem with ‘issue conflict’ in light of double hatting is clear where the double hatter is working on two cases concurrently or has access to non-public information, given that they will be able to influence the outcome to a greater extent. If neither of these circumstances exist, however, the argument loses some of its strength.
For example, in Telecom Malaysia v Ghana,[34] Emmanuel Gaillard’s role as arbitrator was challenged by Ghana. As a legal counsel in Consortium RFCC v Morocco,[35] another investment arbitration, he was seeking the annulment of an award on which Ghana was relying in Telekom Malaysia. The deciding court required Gaillard to give up one role and he gave up his role as arbitrator. When Ghana challenged his continuing role as arbitrator, this was dismissed. The court stated that, occasionally, “an arbitrator has to decide on a question pertaining to which he has previously, in another case, defended a point of view. Save in exceptional circumstances, there is no reason to assume however that such an arbitrator would decide a question less open-minded than if he had not defended such a point of view before.”
However, can an arbitrator not also be required to “decide on a question pertaining to which he has previously, in another case, defended a point of view” as an arbitrator? That is to say, save for concurrent and confidential cases, does the issue conflict not remain valid where double hatting is abolished? Where legal counsel and arbitrators are limited to their respective roles, they will have expressed opinions or put forward arguments on certain subjects in the past, especially given the narrow range of issues under BITs. For example, an arbitrator may seek to rely on their own previous judgments delivered as an arbitrator, especially as ISDS awards are frequently public.[36] It is difficult to see how a ban on double hatting would solve such an ‘issue conflict’. If anything, a ban on double hatting might mean that parties are “tempted to repeat the same appointments, which might raise concerns regarding the prejudgment of the subject matter of the case”.[37] Therefore, when proposing solutions, ‘issue conflicts’ shall be placed within the narrow framework of concurrent cases and insider information.
III. The wider context
1. Investment arbitration
Whether perceived or actual, conflicts resulting from double hatting “engender a lack of confidence in the investment arbitration system.”[38] The public law nature of international arbitration[39] mean that “concerns over lack of transparency and conflict of interests have featured prominently in the legitimacy crisis surrounding international investment arbitration”.[40] These issues have led to experts such as Juan Fernandez-Armesto to state that restrictions on double hatting may well be “a necessary cost for the survival of investment arbitration.”[41]
This is especially true “in the current political climate.”[42] As seen with Elizabeth Warren’s criticism, concerns about double hatting “has strengthened the many contemporary critiques of the investment arbitration system,” which “will become even more relevant as investment arbitration continues on a growth trajectory.”[43]
Despite investors only having won on the merits in 47% of ISDS cases,[44] ISDS awards have been perceived as excessively pro-investor, pro-investment, and anti-developing state.[45] The “failure to address process legitimacy concerns” such as double hatting “only compounds the sense of crisis.”[46] However, whilst several foreign states have withdrawn from the investment arbitration system or are threatening to do so,[47] “it is not clear that ‘issue conflict’, as opposed to other factors, contributed to the decisions of a handful of countries to [do so]”.[48]
Additionally, other concerns within the wider context of investment arbitration, such as double hatting “leaving awards vulnerable to annulment”,[49] are overstated. One commentator advancing this claim referred to the annulment of a €128 million arbitral award due to undisclosed double-hatting by the experts.[50] However, it seems more probable that it was the non-disclosure rather than the double-hatting that led to the annulment. This case simply reinforces that double hatters should disclose conflicts of interest whenever in doubt. For example, in Vivendi v. Argentina,[51] Argentina objected to Vivendi’s reliance on an award issued in Eureko v. Poland,[52] which was written by one of Vivendi’s lawyers after the Vivendi arbitration had commenced. Nonetheless, the Vivendi tribunal disregarded Argentina’s request by citing Eureko. Other cases where biases were disclosed have resulted in similar decisions.
Furthermore, one commentator postulates that where arbitral awards, written by double-hatters in foreign jurisdictions, are brought to countries recognition and enforcement, judges may be emboldened to refuse to do so on grounds of procedural fairness/due process and public policy.[53] However, in practice, there is also little evidence of this.
Yet the reality is perhaps unimportant, given that the perception of bias is the pivotal issue. What is clear is that investment arbitration is being criticised, and the perception of possible bias among double hatters contributes to this. In order to address the concerns with investment arbitration, it is therefore vital to address the critiques of double hatting.
2. Diversity
Double hatting and diversity are frequently mentioned within the same breath. Diversity has been shown to lead to better decision making”[54] and “better results”.[55] Diversity in ISDS tribunals is particularly important because of the diverse makeup of those affected by public decisions. Whilst “85% of [ICSID] cases are brought […] against a developing country, only about one-third of the arbitrators come from developing countries.” Additionally, a recent survey of ICSID cases from 2012-2019 revealed that out of 1,055 appointments, only 14.4% were women.[56]
However, “the top 25 arbitrators account for 4% of all investment arbitrators, but account for more than 33% of all appointments.”[57] These top 25 arbitrators, frequently referred to as “power brokers,”[58] consist mainly of a “small cadre of established arbitrators caricatured as ‘pale, male and stale.’[59] It is this small group of arbitrators with numerous arbitral appointments that engage in double hatting most frequently.[60] Whilst double hatting occurs among counsel “transitioning from the legal counsel role to the arbitral role, […] they are only in a minority.”[61] There is thus a big difference between “double hatting in the core and in the periphery of the system”, which must be dealt with separately.[62] However, there is disagreement on the best approach. Put simply, there are two groups with discordant ways of framing the issue.
The first group believes that double-hatting, at least in limited form, is what allows younger legal counsel, including women and ethnic minorities, to make the transition from legal counsel to arbitrator.[63] As investment arbitrators are “primarily drawn from the ranks of counsel”,[64] they must continue to practice until receiving “sufficient appointments to make full-time service as arbitrators economically feasible”.[65] In 2019, 31% of female ICSID appointments were first-time appointees,[66] and even after the first appointment, “other appointments may be slow to come, if they come at all.”[67]
This group, therefore, argues that banning double hatting, especially without a transition period, would reinforce “the existing dominance of a relative handful of male arbitrators mostly from Western Europe and North America”,[68] and undermine efforts such as the 2016 ERA Pledge, which aims for more female representation in ISDS. Secondly, they reiterate that limiting double hatting would reduce the overall pool of potential arbitrators and deprive the parties of the ability to select the arbitrator of their choice. Whilst there are over 600 arbitrators to choose from,[69] a ban of double hatting would indeed reduce the pool, given that those otherwise attempting to transition would likely choose to remain in their counsel role, and conversely, “power brokers” would most likely give up their counsel role, given that arbitration is more prestigious and lucrative.[70] Their greater resulting availability, combined with their expertise, may seem them taking on additional appointments and so concentrate the amount of cases decided by the top 25 even further.
As opposed to the first group, which frames double hatting as supporting diversity (though they are likely aware of the downsides also), the second group focuses on double hatting as the reason there is so little diversity among the top arbitrators. In relation to this, they feel that “a restriction on double hatting may diminish the ease with which a small and homogenous group of individuals can dominate the arena.”[71] One commentator stated that forcing people to choose between roles will improve diversity “if they are confident and bold enough to choose the arbitrator course early on in their careers”.[72] Yet herein lies the problem: factors such as ‘imposter syndrome,’ mean that men are more likely than women to put themselves forward for consideration.[73] Belief in oneself or being the loudest in the room, however, does not necessarily make for a better arbitrator. It seems unlikely, therefore, that such a ban on double hatting will improve diversity.
However, when viewed in a more nuanced fashion, perhaps the two viewpoints are not as discordant as they may initially seem. One group feels that double hatting benefits the non-diverse status quo and must, therefore, be restricted, whilst the other maintains that a blanket ban will negatively affect up-and-coming arbitrators. Perhaps, however, both sides would agree to limitations from which new arbitrators would be exempt for a certain number of cases or years.
The suggestions for such limitations are varied. In terms of cases, some propose that “once a counsel obtained their first or even second arbitrator appointment, they should desist from accepting future counsel appointment as they ease into a new role.”[74] However, this does not correspond with reality, given that the average arbitrator only receives one or two ICSID appointments.[75] Perhaps it is better to focus on time limits, in relation to which some propose one year whilst others say two-five years.[76] Given that the average ISDS case takes 3.74 years to be resolved, [77] and that new arbitrators may not be appointed to several cases simultaneously or even within close proximity of one another, a more suitable length seems 10-15 years. Whilst this “could result in high numbers of double hatting cases”,[78] this does not seem that likely, given that double hatting is, and will remain, far more concentrated at the ‘top’. This approach could thus help reduce “the tension between the wish to curb dual hatting and the wish to increase the diversity of those appointed as arbitrators in ISDS cases”,[79] and will be discussed in further detail below.
C. SOLUTIONS
I. ARBITRATOR CHALLENGES
Currently, arbitrator challenges are “the primary mechanism aimed at curbing double-hatting”.[80] Some commentators have suggested that if challenges were [even] more frequent “then nomination of these double-hatters would carry greater risks for all involved and therefore would be avoided.”[81] There are several issues with this proposal. Firstly, it would only work on an individual and not institutional level.[82] Secondly, there has been little success so far. The 2018 ICSID Annual Report reports that of the eighteen annual challenges lodged, sixteen were rejected.[83] This small number of successful challenges, which is likely due to Article 14 of the ICSID Convention requiring a ‘manifest’ lack of moral character, shows not only that challenges are not a viable solution, but also that they are counterproductive, given that they indicate that there is nothing egregious about double hatting. However, given that parties may fear the slowdown in proceedings caused by challenges,[84] those appointing double hatters may think twice about doing so even where they would expect to win any challenge.
II. SELF-REGULATION
Concerns about double hatting have arisen precisely because critics believe “that existing ethical constraints and control mechanisms [e.g. Articles 14 and 40(2) of the ICSID Convention or Article 6(2) of ICSID’s Arbitration Rules] are insufficient to curb the risk or appearance of self-serving behaviour”.[85] This is especially true where arbitrators do not recognise their own unconscious bias. Yet even where they are able to compartmentalise, “[t]he question…is whether they in fact succeed in the eyes of objective observers.”[86] Some commentators find it “unreasonable to expect that all double-hatters will put safeguards in place [and] that self-imposition of safeguards […] will necessarily alleviate the risks of bias.” [87] In the words of a harsher critic, “[s]elf-regulators . . . tend to be self-satisfied”.[88]
Furthermore, “individuals who have ceased to double hat over the past few years do not appear to have done so on the basis of ethical or conflict issues”[89] but rather due to retiring or having gained a sufficiently large caseload as arbitrators.[90] Without evidence of effective self-regulation, it cannot be a viable solution.
III. RESTRICTIONS SIMILAR TO THE JUDICIARY
Commentators state that investment arbitration is becoming more and more institutionalised given its frequent reliance on precedent, public nature, and wide-reaching impact.[91] This has caused a growing number of people to question why arbitration “does not accord with court-like practices.”[92] However, the comparison is artificial and the request unworkable. Judges and arbitrators play fundamentally different roles: judges cannot accept bribes; arbitrators are paid by their clients. Judges have job security; arbitrators work on an appointment-by-appointment basis. Judges can hold parties in contempt of court; arbitrators have no such protections. Additionally, the judiciary falls within the remit of one jurisdiction and national framework, so it is easier to devise equal rules for all. The list goes on. Unless arbitrators can benefit from the same advantages and framework as judges, they should not be subject to the same restrictions.
IV. A BLANKET BAN
One commentator postulates that “an express bar to the practice of double-hatting across institutions seems the most plausible solution to eradicate the same from its roots”,[93] whilst comment 67 of the Draft Code of Conduct states that “[a]n outright ban is easier to implement, by simply prohibiting any participation by an individual falling within the scope of prohibition”.
However, an outright ban “could give rise to a whole new set of problems”,[94] such as the lack of diversity. Indeed, banning “dual hatting would be tantamount to reversing any progress made on gender diversity by the international arbitration community [over] the past years.”[95] Additionally, a blanket ban will be very difficult to arrange unless all institutions come together, i.e. “[c]losing these loopholes would require every arbitration institution to disallow its arbitrators from acting as counsel in any forum”.[96] This means that any individual efforts made so far have only had limited effect.
For example, the International Court of Justice (ICJ) has stated that the simultaneous or sequential exercise of the arbitrator and counsel role is per se incompatible.[97] Whilst this covers all cases rather than just those where the facts, law or parties overlap,[98] it is nonetheless limited to cases within the remit of the ICJ. Similar rules drawn up by the Court of Arbitration for Sport (CAS) back in 2009 are also limited to CAS cases, as per Section 18 of the CAS Code. As an arbitrator could appear as arbitrator elsewhere, the pools of arbitrators could therefore still overlap. Whilst the ‘issue conflict’ might thus be remedied, the ‘club conflict’ would not be.
Certain bilateral treaties go even further than this: pursuant to Article 8.30 of the Comprehensive Economic and Trade Agreement between Canada and the EU (CETA), members of the envisioned arbitral tribunal cannot act as counsel or as party-appointed expert in any pending or new investment dispute under CETA or any other international agreement. Meanwhile, in paragraph 80 of the TTIP investment chapter, the EU includes a ban not only on pending or new ISDS but also on any other dispute under domestic law.[99] Nonetheless, parties to TTIP or CETA could still hire external arbitrators rather than using those from the pool. Additionally, the restrictions under Article 8.30 of CETA are limited in time, whilst counsel appearing before the CETA tribunal can “still arbitrate disputes under other trade agreements.”[100]
Furthermore, a blanket ban is also problematic because double hatting is not an issue per se; it always depends on the facts.[101] For instance, in SGS Société Générale de Surveillance SA v Islamic Republic of Pakistan,[102] the claimant challenged the arbitrator appointed by the respondent because he disclosed he would be acting as co-counsel in GAMI v Mexico.[103] The other two arbitrators deciding the challenge held that this dual role was not a sufficient ground for disqualifying an arbitrator, i.e., something more than the simple existence of ‘two hats’ was needed for a challenge to succeed.[104] Clearly, banning double hatting per se is simply not a solution, meaning a more nuanced approach is required.
V. LIMITED RESTRICTIONS
Article 6 of the Draft Code of Conduct proposes a more nuanced ban. Unlike the examples set out above, this prohibition is not absolute:
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].
(Article 6: Limit on Multiple Roles)
The Draft Code of Conduct’s commentary states that “an outright ban may exclude a greater number of persons than necessary to avoid conflicts of interest” and would affect the freedom of choice so central to ISDS.[105] Perhaps this sentiment is partially due to the fact that unlike the ‘outright bans’ of the institutions listed above (which are nonetheless limited in effect), a blanket ban by ICSID would be far wider in scope, given its prevalence in the field of ISDS.[106]
Whilst this nuanced approach is thus a good starting point, this paper suggests that the clause should be adopted in a specific manner. Firstly, the “disclosure” option should apply to those for whom their first arbitral appointment was within the last 10-15 year, so as not to “unduly harm generational renewal and diversity.”[107] The duty to “refrain from acting” should apply to everyone else and should be valid for two years. In both cases, the clause should only apply where the adjudicators are acting for the same parties or on matters involving a very similar set of facts. As the same treaty and even similar facts are not uncommon in ISDS, anything broader would risk being tantamount to a blanket ban.
Disclosure beyond these specific requirements should be governed by the IBA Guidelines, i.e. matters should be reported where they may affect impartiality from the view of a reasonably-informed third party. The IBA Rules could be included under all arbitral institutions’ frameworks so that there is a binding, overarching duty to disclose,[108] and more generally, arbitration institutions should issue “rule changes aimed at reinforcing ethical requirements or assuring broader disclosures by prospective arbitrators”.[109]
Beyond this, the institutional appointment of arbitrators presents itself as a partial solution.[110] This would enable institutions to choose pre-selected arbitrators based on required expertise and would guarantee employment from the perspective of the arbitrator. Solely the “institutional appointment of even one arbitrator may alleviate the current scenario”,[111] provided the pool contains some experienced arbitrators.[112] Knowing that they are neither dependent on selection by other double hatters or able to appoint them in future might go some way to reduce the “club conflict”. However, when making selections, either for specific cases or for the pool, institutions must be “attentive to issues of generational renewal and diversity in the way they select, and construct the selection criteria and voting rules for selecting those arbitrators/judges.”[113]
Lastly, the dangers of double hatting could further be addressed by limiting unilateral appointment powers and especially by “prohibiting those with unilateral power from serving as party counsel”.[114] Such a diffusion would allow for “high-ranking officials to serve as party counsel while maintaining the integrity of the arbitral process.”[115]
VI. REDUCING DOUBLE HATTING AMONG ‘POWER BROKERS’
Some prominent researchers state that “the greatest impetus for change may come not from regulatory action, but from a few prominent individuals’ decisions to wear a single hat”.[116] A study found that “[i]f about 10 to 15 individuals agreed to stop double hatting, there would both a dramatic drop in the number of cases and, most likely, a rapid delegitimisation of the process.”[117] Whilst this is interesting in theory, an outright ban would cause havoc among the ISDS community and a voluntary decision to step back seems equally unlikely.
However, where the aforementioned solutions do little to curb the double hatting issue, this may be the only solution. In this respect, a “one way” mechanism might be envisaged, under which counsel can act as arbitrators but not the other way around. To determine whether an individual is a counsel or arbitrator, one would assess the majority of their appointments over the past 10 years. Of course, this will likely reinforce successful arbitrators acting as such and lead to a greater concentration of cases among ‘power brokers’, but would also serve to drastically reduce double hatting without limiting the transition of younger counsel to arbitrators.
D. CONCLUSION
Whilst “an organic solution [is] preferable to more drastic action by arbitration institutions”,[118] imminent action must nevertheless be taken in light of the negative attention that ISDS is receiving. However, this should not be in the form of a blanket ban on double hatting, given the threat this poses to diversity among arbitrators. Rather, with appropriate safeguards, it is possible “for double-hatters to have the requisite arbitral impartiality”, thus maintaining the practice of double hatting whilst reducing the criticism surrounding it.[119] Such safeguards should comprise of a restriction on acting for the same parties or in relation to matters covering very similar facts. Alternatively, where arbitrators are still within the 10-15 year transition phase from counsel to arbitrator, such safeguards would merely require disclosure.
Other safeguards may include introducing stricter disclosure obligations regarding conflicts of interest, similar to the IBA Guidelines, into ICSID’s and other institutions’ frameworks. This is especially important given the high bar that otherwise determines conflict, such as Section 14 of the ICSID Convention. Furthermore, institution-appointed arbitrators, even if just one out of three, should be seriously considered, as well as a limitation on the unilateral powers of certain arbitrators. However, such measures should be instituted carefully so as not to inadvertently become “barriers to new and more diverse arbitrators seeking their place in ISDS”,[120] or otherwise reducing party autonomy and the freedom to appoint arbitrators.[121]
Where these proposed safeguards do not result in significant changes, however, more radical solutions must be contemplated. This includes limiting double hatting specifically among the top arbitrators, where it is by far the most prevalent. Again, care must be taken that this does not amount to a blanket ban, given the dangerous effect this may have on diversity.
Overall, these steps should result in gradual change for the better. Whilst it is too early to state that “double-hatting may soon be an evil of the past”,[122] a careful and balanced approach will hopefully see a reduction in double hatting or at the very least a reduction in concerns about impartiality, as well as a renewed belief in the fairness of investment arbitration. Even the simple fact that steps are being taken to remedy concerns may alleviate some tension. Hopefully, the investment arbitration community will realise that accurate or not, perceptions matter, and start working on solutions by creating bridges within the decentralised system. Whether some notable individuals may start ‘hanging up their hats’, however, is something that remains to be seen.
[1] Malcolm Langford, Daniel Behn, and Runar Hilleren Lie, ESIL Reflection the Ethics and Empirics of Double Hatting, ESIL Reflection (2017), https://esil-sedi.eu/post_name-118/ (Langford et al.).
[2] Ruth Mackenzie and Phillipe Sands,‘International Courts and Tribunals and the Independence of the International Judge, 44 Harvard International Law Journal 271 (2003).
[3] Idem.
[4] Joshua Tayar, Safeguarding the Institutional Impartiality of Arbitration in the Face of Double-Hatting, McGill Volume 5 (2018-2019), https://mjdr-rrdm.ca/files/sites/154/2019/07/Tayar_FINAL.pdf (Tayar).
[5] Langford et al.
[6] ICSID and UNCITRAL, Draft Code of Conduct for Adjudicators in Investor-State Dispute Settlement, ICSID Website (2020), https://icsid.worldbank.org/sites/default/files/Draft_Code_Conduct_Adjudicators_ISDS.pdf (last accessed 20 September 2020) (the Draft Code).
[7] Idem, para. 66.
[8] Idem.
[9] Divij Jain, Double Hatting: Challenging the International Investment Dispute Resolution Mechanism, Society For Excellence In Arbitration Law (2020), https://rmlnluseal.home.blog/2020/08/04/double-hatting-challenging-the-international-investment-dispute-resolution-mechanism/ (Jain).
[10] Jain, Tayar, and others.
[11] Nassib G Ziadé, How Many Hats can a Player Wear: Arbitrator, Counsel and Expert?, ICSID Rev 52 2009), https://academic.oup.com/icsidreview/article-abstract/24/1/49/2012008?redirectedFrom=PDF.
[12] Malcolm Langford, Daniel Behn and Runar Hilleren Lie, The Revolving Door In International Investment Arbitration, Journal of International Economic Law (2017), https://www.researchgate.net/publication/327208015_The_revolving_door_in_international_investment_arbitration (Langford et al. 2).
[13]Elizabeth Warren, The Trans-Pacific Partnership clause everyone should oppose, The Washington Post (2015), https://www.washingtonpost.com/opinions/kill-the-dispute-settlement-language-in-the-trans-pacific-partnership/2015/02/25/ec7705a2-bd1e-11e4-b274-e5209a3bc9a9_story.html?utm_term=.6d1fd87c593e.
[14] John R. Crook, Symposium: a Focus on Ethics in International Courts and Tribunals Dual Hats and Arbitrator Diversity: Goals In Tension, AJIL Unbound (2019), https://www.researchgate.net/publication/335854373_Dual_Hats_and_Arbitrator_Diversity_Goals_in_Tension (Crook).
[15] Sport Maska Inc v Zittrer, [1988] 1 SCR 564 at 604, 38 DLR (4th) 221.
[16] Tayar
[17] Richard C Reuben, Constitutional Gravity: A Unitary Theory of Alternative Dispute Resolu-
tion and Public Civil Justice, UCLA Law Review (2000), https://scholarship.law.missouri.edu/cgi/viewcontent.cgi?article=1335&context=facpubs.
[18] IBA Guidelines on Conlicts of Interest in International Arbitration, Part I, §2 (Council of the Int’l Bar Ass’n 2004).
[19] Langford et al.
[20] Idem.
[21] Elsa Sardinha, The Impetus for the Creation of an Appellate Mechanism, ICSID Review - Foreign Investment Law Journal (2017), https://academic.oup.com/icsidreview/article-abstract/32/3/503/4718101 (Sardinha).
[22] Bank of Montreal v Brown, 2006 FC 503, [2006] FCJ No 623.
[23] Tayar
[24] Idem.
[25] Crook
[26] Idem.
[27] Sardinha
[28] Buergenthal
[29] Jain and Langford et al.
[30] Dennis H. Hranitzky and Eduardo Silva Romero, The ‘Double Hat’ Debate In International Arbitration, New York Law Journal (2010), https://www.dechert.com/content/dam/dechert%20files/knowledge/publication/2010/6/the-double-hat-debate-in-international-arbitration/070101031Dechert.pdf (Hranitzky and Romero).
[31] Anthony Sinclair and Matthew Gearing, Partiality and Issue Conlicts, Transnational Dispute Management, Vol. 5, Issue 4 (July 2008).
[32] Margie-Lys Jaime, Counsel Conduct in International Arbitration: An Ethical “No-Man’s Land”?, The International Journal of Arbitration, Mediation and Dispute Management (2019), https://kluwerlawonline.com/journalarticle/Arbitration:+The+International+Journal+of+Arbitration,+Mediation+and+Dispute+Management/85.3/AMDM2019033 (Jaime).
[33] Hranitzky and Romero
[34] Telekom Malaysia Berhad v. The Republic of Ghana PCA Case No. 2003-03.
[35] Consortium RFCC v. Royaume du Maroc, ICSID Case No. ARB/00/6.
[36]ICSID, https://icsid.worldbank.org/services/arbitration/convention/process/confidentiality-transparency (last visited 20 September 2020).
[37] Jaime
[38] Sardinha
[39] Langford et al. 2
[40] Langford et al.
[41] Jaime
[42] Langford et al. 2
[43] Langford et al. 2
[44] Registered cases - treaty-based, ICSID contract-based, or FDI law-based.
[45] Langford et al. 2
[46] Idem.
[47] Hranitzky and Romero
[48] Jaime
[49] Coleman and Bond
[50] Eiser Infrastructure Limited and Energia Solar Luxemburg S.À.R.L. v Kingdom of Spain, ICSID Case No. ARB/13/36 (2020)
[51] Vivendi v. Argentina, ICSID Case No. ARB/97/3 (2005)
[52] Eureko v. Poland, UNCITRAL (2005)
[53] Tayar
[54] Clarissa Coleman and Louise Bond, Two Heads Are Better Than One: Double Hatting and Its Impact On Diversity In International Arbitration, K&L Gates LLP (2020), https://www.klgates.com/two-heads-are-better-than-one-double-hatting-and-its-impact-on-diversity-in-international-arbitration-7-30-2020 (Coleman and Bond).
[55] Vanina Sucharitkul, ICSID and UNCITRAL Draft Code of Conduct’s Potential Ban on Multiple Roles Could Have A Severe Impact on Gender Diversity, Arbitral Women Website (2020), https://www.arbitralwomen.org/icsid-and-uncitral-draft-code-of-conducts-potential-ban-on-multiple-roles-could-have-a-severe-impact-on-gender-diversity/ (Sucharitkul).
[56] Sucharitkul
[57] Langford et al. 2
[58] Langford et al. 2
[59] Michael D. Goldhaber, Madame La Presidente – A Woman Who Sits as President of a Major Arbitral Tribunal is a Rare Creature. Why?, 1(3) TRANSNAT’L DISP. MGMT (2004).
[60] Crook
[61] Idem.
[62] Anthea Roberts, A Possible Approach to Transitional Double Hatting in Investor-State Arbitration, EJIL Talk (2017), https://www.ejiltalk.org/a-possible-approach-to-transitional-double-hatting-in-investor-state-arbitration/ (Roberts).
[63] Sucharitkul
[64] Idem.
[65] Idem.
[66] Idem.
[67] Crook
[68] Sucharitkul
[69] Langford et al.
[70] Hranitzky and Romero
[71] Coleman and Bond
[72] Hranitzky and Romero
[73] Coleman and Bond
[74] Langford et al. 2
[75] Crook
[76] Coleman and Bond
[77] Comment left by Langford on Roberts’ page as referenced above.
[78] Idem.
[79] Crook
[80] Idem.
[81] Tayar
[82] Idem.
[83] Crook
[84] Jain.
[85] Idem.
[86] Tayar.
[87] Idem.
[88] Jan Paulsson, The Idea of Arbitration, Oxford University Press (2013), page 147.
[89] Lacey Yong, Double Hatting Under New Scrutiny, Global Arbitration Review (2017), https://globalarbitrationreview.com/article/1142550/%E2%80%9Cdouble-hatting%E2%80%9D-under-new-scrutiny (Yong).
[90] Yong
[91] Sardinha
[92] Langford et al. 2
[93] Jain
[94] Hranitzky and Romero
[95] Sucharitkul
[96] Tayar.
[97] Idem.
[98] Idem.
[99] Jaime
[100] Tayar
[101] The Draft Code
[102] SGS Société Générale de Surveillance S.A. v. Islamic Republic of Pakistan, ICSID Case No. ARB/01/13.
[103] Gami Investments Inc. v. Mexico, UNCITRAL (2004).
[104] Jaime
[105] Draft Code of Conduct, para. 68.
[106] Langford et al. 2
[107] Roberts
[108] Sucharitkul
[109] Roberts
[110] Crook
[111] Idem.
[112] Idem.
[113] Roberts
[114] Acomb and Jones
[115] Idem.
[116] Crook
[117] Langford et al. 2
[118] Crook
[119] Tayar
[120] Crook
[121] Sucharitkul
[122] Jain