ESIL Reflection: The Ethics and Empirics of Double Hatting
Vol 6, Issue 7 – TURN TO EMPIRICISM Series
Editorial board: Samantha Besson, Jean d’Aspremont (Editor-in-Chief), Jan Klabbers and Christian Tams
Is it time to end the practice of double hatting in international adjudication and especially investment arbitration? In his closing speech at the 2015 European Society of International Law conference, Philippe Sands took aim at some of the association’s members. The international legal profession, he maintained, bore some responsibility for the legitimacy crisis in international law. The crux of his concern was the ethics of appointments. To a reasonable observer, it might appear that international lawyers were prioritising their material and political interests over independence and impartiality.
Sands named four specific practices. First, select lawyers and law firms were ‘capturing’ international investment arbitration and charging excessive fees. Second, International Court of Justice (ICJ) judges were acting as arbitrators – seemingly the ‘only’ international court to allow this practice. Third, some judges and arbitrators were too close to states, participating in the appointment processes of state counsel or leaking confidential information to governments. Fourth, he labelled as ‘deplorable’ another practice of double hatting in which individuals act simultaneously as arbitrators and legal counsel in international investment arbitration.
This final warning about double hatting is neither new nor isolated. Already in 2003, Sands and Mackenzie raised such concerns; and warnings have come from other senior figures. These admonitions should be taken seriously. Concerns over lack of transparency and conflict of interests have featured prominently in the legitimacy crisis surrounding international investment arbitration, but also in critiques of the ICJ.
To be sure, raising concerns over the practices of senior and influential figures is uncomfortable. As Sands puts it, such critique is “delicate and embarrassing” but unavoidable as it goes to the “heart of the system.” In this regard, he asserts that senior international lawyers carry a special duty – it is up to them to raise the alarm as such a task is too ‘daunting’ for junior lawyers and scholars. This assumption of responsibility is magnanimous but it is doubtful whether the new generation of international lawyers, which includes ourselves, should indulge in either silence or obsequence. Even if it carries reputational risks or threatens chances of future appointments, reflexive critique is a necessary task in both professional practice and academic scholarship. However, any critique must be informed. It should be empirically based and normatively contingent – open to the possibility of justifiable forms of double hatting.
In this Reflection, we examine the practice of double hatting in the specific context of international investment arbitration. We ask three questions: how widespread is the practice; when is it a problem; and what can be done? In doing so, we introduce and develop new empirical findings but also make links between investment arbitration and broader practices in international adjudication.
How Widespread Is Double Hatting?
Despite a decade-long expression of concern with double hatting, it has never been measured. It remains the subject of anecdote and casual observation. Even Sands was unsure as to how far he could draw on “isolated” incidents to undergird his general critique.
With our new PITAD database, we can empirically measure the extent of double hatting in one prominent form of international adjudication: investment arbitration. Our dataset includes all known treaty-based arbitrations, ICSID contract and foreign direct investment law-based arbitrations, and ICSID annulment committee proceedings (1077 cases in total through January 1, 2017). In an article just published in the Journal of International Economic Law (JIEL), we drew on this database to quantitatively analyze the networks of investment arbitrators, legal counsel, expert witnesses, and tribunal secretaries, identifying the ‘power brokers’ in the system and those arbitrators most engaged in the practice of double hatting.
Here, we draw on but develop further the analysis of double hatting with a focus on its scope as well as its nature. We examine the extent of double hatting, the identity of those that double hat, and whether there is a change over time.
We measure the formal extent of double hatting in international investment arbitration in two principal ways. First, an ‘arbitrator-focused’ approach counts all individual cases in which at least one arbitrator on the panel is simultaneously acting as legal counsel in at least one other individual international investment arbitration case. Second, a ‘counsel-focused’ approach counts individual cases where at least one legal counsel (on either side of the dispute) is simultaneously acting as an arbitrator in at least one other individual international investment arbitration case (the inverse of the first measure).
Table 1 sets out the figures. We have found that a total of 47% of cases (509 in total) involve at least one arbitrator simultaneously acting as legal counsel. This is dramatically higher than expected. Moreover, in 190 of the cases in this arbitrator-focused category, there are also legal counsel double hatting elsewhere as arbitrators – deepening the extent of the revolving door. In other words, in a fifth of cases both the bench and bar are playing the opposite role somewhere else: double hatting-squared. Turning to the counsel-focused category, we find that there are another 11% of cases (118 in total) where legal counsel (but none of the arbitrators) are double hatting elsewhere as arbitrators.
Table 1. The Extent of Double Hatting (No. of Cases)
|Type of Double Hatting||No. Cases||%|
|Cases with only arbitrators double hatting||319|
|Cases with both arbitrators and legal counsel double hatting||190|
|No double hatting||450||42%|
The difference between these categories sheds an important light on the nature of double hatting. The number of arbitrator-only double hatting cases (319 in total) is three times that of legal counsel-only cases (118). This is largely attributable to the fact that the pool of double hatting legal counsel is much more diverse and fragmented than the pool of double hatting arbitrators. Double hatting is a practice that is dominated by a small group of arbitrators with numerous arbitral appointments but a comparatively smaller amount of simultaneous legal counsel work. If we examine the top 25 most frequent double hatters according the index we created in our JIEL article, we find that these individuals account for 412 arbitral appointments but only act as legal counsel in 224 cases.
The next question concerns the identity of this group. Which arbitrators are engaged in this practice? Is it a constant practice or a phenomenon that only arises in a career transition between roles? And is it a random selection of individuals or are there specific features which characterize this group?
In operationalizing these questions, we have developed a double hatting index. The method was previously reported in our JIEL article. If an individual is involved with a minimum of two international investment arbitration cases in a minimum of two different roles, then a yearly score of (2) is assigned to that individual for a given year. The score is rather conservative and does not attempt to describe the intensity of double hatting within a calendar year. Hence the maximum score an individual can receive for a year is (2).
Figure 1 lists the top 10 according to those criteria. This index is revealing about the nature of double hatting. First, it shows that the practice is dominated by many of the arguably most powerful and influential arbitrators in the system – that we have also previously identified with social network analysis. Second, there are many powerful arbitrators who do not feature at all on this list or even among the top 100 on our double hatting index. This includes a number that have spoken out against the practice, such as Philippe Sands, W. Michael Reisman, Sir Franklin Berman, and Judge Thomas Buergenthal. This suggests a degree of agency and choice in whether to double hat or not. Third, if we examine the top 25 double hatters according to our index, we do find a group of prominent legal counsel that appear to be merely transitioning from the legal counsel role to the arbitral role – but they are only in a minority.
Figure 1. The Double Hatting Index (Top 10)
Finally, it is important to ask if the degree of double hatting has changed over time? Indeed, anecdotes to the effect that it has decreased are not uncommon in the international investment arbitration community – but no evidence has been offered. Figure 2 shows the proportion of cases affected each year by double hatting. The purple blocks show the proportion of arbitrator-focused double hatting cases. Interestingly, double hatting is a relatively late phenomenon that primarily begins in the early 2000s – the origins of which should be investigated further. For present purposes, it is notable that it continues to account for a high proportion of cases from 2003 to 2016, and some of the highest proportions are in the period 2011 to 2015. The share of double hatting cases is lower in 2016 but it is not yet clear whether this represents a new trend – a similar one-off reduction occurred in 2006.
Figure 2. Double Hatting over Time
It also interesting to examine who has ceased double hatting in recent times. In the last three to four years, some individuals have largely ceased taking new cases as either arbitrator or legal counsel. This seems mostly due to the onset of retirement, a sufficiently large caseload of arbitrator appointments (precluding by necessity further work as legal counsel), or appointment to the ICJ (which does not permit work as legal counsel simultaneously). Therefore, it does not appear that the individuals ceasing to double hat over the past few years are making this choice as a response to ethical or conflict issues.
When Is Double Hatting a Problem?
Our measurement of the double hatting phenomenon clearly identifies a practice that continues to persist among a limited group of highly influential and respected individuals. But is it really a problem?
It is possible to argue that double hatting in everyday commercial arbitration is not problematic. It might be sometimes even beneficial. As both Shapiro and Stone Sweet have theorized, when two parties in a conflict first turn to third-party resolution, trust in the dispute resolver’s neutrality and expertise, their “impartiality and wisdom”, is paramount. In the long-standing field of domestic and international arbitration, this premise of trust has been prominent. Parties to disputes typically seek to draw arbitrators from a legal community where their expertise, reputation (and arguably preferences) are well-established. Thus, it is not uncommon for a single individual to play the roles of arbitrator and legal counsel over a sustained period.
Given that arbitrations are, by their nature and structure, one-off disputes that do not provide tenured appointments, international arbitral practice has unsurprisingly taken a fairly relaxed attitude to counsel appointment for arbitrators. Historically at least, an arbitrator’s primary occupation was not one of the ‘professional, full-time arbitrator’. Thus, considering the unique nature of arbitration-based adjudication, double hatting does not pose in theory an immediate legitimacy problem. Indeed, critiques of the appointments system in international arbitration tend to focus on related but different issues. These include the exploitation by insiders of information asymmetries in order to monopolize or allocate appointments or the existence of quid pro quo arrangements between counsel and arbitrators for future appointments.
However, international investment arbitration is a different kettle of fish. It is arguably more public than private. Investment arbitration involves: international treaties; states as respondents (often in their role as sovereign); issues of domestic public policy; the interests of multiple stakeholders; and, in the case of ICSID and the Permanent Court of Arbitration (PCA), public institutions appointing arbitrators. Moreover, international investment arbitration since the early 2000s has taken on a more juridical form as precedent has formed an important part of arbitral reasoning. Such a context fits well with what Shapiro and Stone Sweet label as socially complex forms of adjudication, which place higher demands on dispute resolvers. The outcomes to any case are less likely to fall within the range of accepted outcomes for both disputing parties (and directly interested third parties), and so an adjudicator must work much harder to demonstrate neutrality.
It is thus no surprise that as investment arbitration has become more public and complex that it must address, like courts, broader concerns with legitimacy – of which double hatting is just one. Moreover, investment arbitration is now constituted by a de facto judiciary – a cadre of full time arbitrators that dominate the field. The investment arbitrator of today resembles nothing of the traditional and genteel picture of the moonlighting arbitrator in the occasional dispute. Investment arbitration looks increasing like a court-like system, and it is being challenged to why it does not accord with court-like practices.
The above discussion indicates why double hatting might have emerged and is now under scrutiny. The important question to ask though is whether it is really a problem – is there a legitimacy problem? Even if double hatting is not currently prohibited in international investment arbitration, should we seek to end it?
The key reason for the skepticism to the practice can be summarized in this dilemma. Philippe Sands asks: Can a lawyer that “spends a morning drafting an arbitral award that addresses a contentious legal issue” divorce themselves from their role in addressing the same or similar legal issue in the “afternoon” as counsel in a different case. And, even if they can, would they be able to convince a reasonable observer that such role bifurcation was maintained. The concern remains: an arbitrator’s ability to shape legal precedent and evaluation of facts (or access confidential information) may assist them as legal counsel elsewhere. Another concern would be the inverse of the first: an arbitrator acting as legal counsel is colored by their arbitral role or uses his or her pleadings in one case for the purpose of being picked up and used by the same individual in their work as an arbitrator in another case.
While double hatting is not formally prohibited, this example from Sands corresponds with rules concerning conflict of interests set out in soft law. The IBA Guidelines on Conflicts of Interest in International Arbitration provide that:
(a) An arbitrator shall decline to accept an appointment or, if the arbitration has already been commenced, refuse to continue to act as an arbitrator, if he or she has any doubt as to his or her ability to be impartial or independent; (b) The same principle applies if facts or circumstances exist, or have arisen since the appointment, which, from the point of view of a reasonable third person having knowledge of the relevant facts and circumstances, would give rise to justifiable doubts as to the arbitrator’s impartiality or independence […]; (c) Doubts are justifiable if a reasonable third person, having knowledge of the relevant facts and circumstances, would reach the conclusion that there is a likelihood that the arbitrator may be influenced by factors other than the merits of the case as presented by the parties in reaching his or her decision.
The classic example of role bifurcation creating concerns involved French arbitrator Gaillard and the case of Telkom Malaysia v. Ghana. He was acting as the claimant-appointed arbitrator at the same time that he was the claimant’s legal counsel in an ICSID annulment proceeding in a different case (RFCC v. Morocco). Even though Gaillard disclosed that he was acting as counsel in this case, the respondent state (Ghana) in the arbitration where he was acting as arbitrator lodged multiple challenges against him in the Hague District Court (Netherlands was the seat of the arbitration) alleging that his role as arbitrator was incompatible with his simultaneous role as legal counsel in the ICSID annulment, especially due to the fact that both cases involved similar legal issues and that Ghana was relying on the case of RFCC v. Morocco in its submissions. Ghana argued inter alia, that Gaillard, “who in his capacity of counsel opposes a specific notion or approach, cannot be unbiased in his judgment of that same notion or approach in a case in which he acts as arbitrator.” The Hague District Court ordered Gaillard to choose whether to continue as arbitrator or legal counsel, but not both. Gaillard subsequently resigned from this position as legal counsel in the ICSID annulment proceedings. Even if Gaillard was able to separate his different roles in different cases, this example highlights, at a minimum, that an appearance or perception of bias or conflict was reasonable, even if there was no showing of actual bias. Ultimately, this challenge to Gaillard was resolved through his resignation in one of his roles.
While not every case of double hatting may fall into this category, the emphasis in both the IBA Guidelines and Hague case on the justifiable doubts of a reasonable third person should give the arbitral community pause. If leading arbitrators in the field have doubts over the practice, 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.
Nonetheless, it is not uncommon to hear arguments that support the practice. First, there is the claim that, practically speaking, there is a small pool of investment arbitrators that can sit in these types of arbitrations. Thus, limiting qualified individuals from sitting as arbitrator due to work as legal counsel would undermine the quality and rigor of the adjudicators acting in the current system. Empirically speaking, however, this argument would only hold for the 1990s. Since then, the potential pool of experienced investment arbitrators has expanded significantly – more than 600 having served in at least one case. While there may be a shortage of qualified adjudicators in other fields, investment arbitration appointments are viewed as highly prestigious. Investment arbitration is a buyer’s market not a seller’s market.
Second, some point to the need for transitional arrangements as lawyers move from the legal counsel role to the full-time arbitrator role. From an economic perspective, it may be fair to permit legal counsel seeking to become full-time arbitrators space to carry on their legal counsel practice until there is some guarantee that arbitral appointments will come in the future. This argument, however, probably only holds for a short period of time: once a counsel obtained their first or even second arbitrator appointment, they could desist from accepting future counsel appointments as they ease into a new role.
In our view, the defences of double hatting are largely thin and unconvincing in the contemporary world of investment arbitration. Today, there is a highly competitive arbitration market with a large pool of experienced arbitrators and legal counsel. Moreover, international investment arbitration is highly judicialized and public in character and dominated by a small group of arbitrators that constitute a de facto international investment judiciary. This suggests that the norms and expectations of standing courts rather than ad hoc arbitration should apply. Finally, empirically, double hatting is more a norm than transition – it is practiced foremost by experienced arbitrators rather than younger counsel in transition. Although it may be possible to defend double hatting when it truly is a graduated and transparent transition between roles.
There are also pragmatic reasons for ending double hatting. Perceptions of illegitimacy have consequences for the system. These must be acknowledged even if one is unconvinced by the normative concerns – the perception of bias or conflict remain. This alone should be enough for the international investment arbitration community to reconsider the practice.
What Can Be Done?
Given that double hatting in international investment arbitration can give rise to justifiable doubts about the impartiality and independence of arbitrators; or at the very least is a practice that undermines the sociological legitimacy of the international legal system, what can be done?
The fundamental challenge with any reform proposal is that the regime is highly polycentric. There is no overarching coordinating institution that could facilitate and enforce particular conflict of interest rules. For instance, the IBA Guidelines are just that: guidelines. They do not create clear legal obligations on arbitrators that can be sanctioned. In our view, four possible approaches might be adopted on a scale from short-term to long-term.
The first approach is self-regulation. If about 10 to 15 individuals agreed to stop double hatting, there would be both a dramatic drop in the number of cases and, most likely, a rapid delegitimatisation of the practice. As effective ‘gatekeepers’ in international arbitration, central arbitrators exercise remarkable power in shaping the norms and processes of appointment. We would be interested to know if the leading actors would consider such a move.
The second approach is enhanced transparency. Increased and sustained attention on the practice may reduce the reputational incentives for double hatting. Until now, double hatting has probably carried positive reputational benefits. In 1997, in international commercial arbitration, Dezalay and Garth observed the enhanced symbolic power of individuals that crossed institutional lines – becoming the “virtuosos and prima donnas” of arbitration. A higher number of cases signals to both markets and onlookers a particular form of power. Yet, an increase in transparency around the practice could turn the reputational effects negative. While this would require more public information about appointments, there is no guarantee that information in itself affects how the arbitration market functions. Transparency initiatives would need to affect concretely decision-making processes, highlighting normative concerns in order to shift reputational incentives. Thus, we plan to annually report our double-hatting index in order to show general trends as well as which individuals are double hatting more or less in the period.
The third approach is litigation strategy. If legal counsel for claimants and respondents raise the issue more regularly in litigation, as in the Gaillard case above, then nomination of double hatting arbitrators carries greater risks for all involved. The fact that an arbitrator is currently acting as legal counsel may become a potential liability that requires thought before nomination. Moreover, an increase in such challenges could help clarify when double hatting becomes a significant than just a possible ethical concern.
The final approach is institutional reform. This could include a double hatting ban in new international investment agreements (as for example in the recent CETA) or a reform or modification of the institutional rules governing the vast majority of these arbitrations. This may be difficult, but the adoption of model texts and vanguard initiatives from certain arbitration houses or ICSID (as we have seen in regards to efforts to promote women in the arbitral role) could produce a snowballing effect that catalyzes a change in the culture.
Cite as: Malcolm Langford, Daniel Behn, and Runar Hilleren Lie, ‘The Ethics and Empirics of Double Hatting’ 6:7 ESIL Reflection (2017).
* Professor, Faculty of Law, University of Oslo and Co-Director, Centre on Law and Social Transformation, University of Bergen and CMI. Email: email@example.com.
** Visiting Assistant Professor of Law, Penn State Law, Pennsylvania State University and Senior Research Fellow, PluriCourts Centre of Excellence, University of Oslo. Email: firstname.lastname@example.org.
*** Research Assistant, PluriCourts Centre of Excellence, University of Oslo. Email: email@example.com.
 Phillipe Sands, ‘Developments in Geopolitics: The End(s) of Judicialization?’ 2015 ESIL Conference Closing Speech, 12 September 2015. Reprinted in EJILTalk!
 Ruth Mackenzie and Phillipe Sands, ‘International Courts and Tribunals and the Independence of the International Judge,’ 44 Harvard Int’l Law J. 271 (2003). See also Phillipe Sands, ‘Conflict and Conflicts in Investment Treaty Arbitration: Ethical Standards for Counsel,’ in Arthur Rovine (ed.), Contemporary Issues in International Arbitration and Mediation: The Fordham Papers (New York: Brill, 2012), 28-49.
 See e.g. Thomas Buergenthal, ‘The Proliferation of Disputes, Dispute Settlement Procedures and Respect for the Rule of Law,’ 3(5) Transnational Dispute Management (2006).
 See e.g. Chiara Giorgetti, ‘Who Decides Who in International Investment Arbitration,’ 35(2) U. Penn. J. Int’l Law 431 (2014); Sergio Puig, ‘Blinding International Justice,’ 56(3) Virginia J. Int’l Law 647 (2016); Catherine Rogers, ‘Gamblers, Loan Sharks and Third-Party Funders,’ in Catherine Rogers (ed.), Ethics in International Arbitration (Oxford: OUP, 2014); Pia Eberhardt and Cecilia Olivet, ‘Profiting from Injustice: How Law Firms, Arbitrators, and Financiers are Fueling an Investment Arbitration Boom,’ Corporate Europe Observatory (November 2012).
 See e.g. Eric Posner and Miguel de Figueiredo, ‘Is the International Court of Justice Biased?’ 34 Leg. Stud. 599 (2005).
 Sands, supra n 1.
 For other critics, see Judith Levine, ‘Dealing with Arbitrator “Issue Conflicts” in International Arbitration,’ 61 Disp. Res. J. 60 (2006); Joseph Brubaker, ‘The Judge Who Knew Too Much: Issue Conflicts in International Adjudication,’ 26(1) Berkeley J. Int’l Law 111 (2008). Double hatting as an expert witness at the same time as acting as arbitrator or legal counsel may also be problematic, but requires a more nuanced ethical discussion.
 Sands, supra n 1.
 PluriCourts Investment Treaty and Arbitration Database (PITAD). See introduction in Malcolm Langford and Daniel Behn, ‘Managing Backlash: The Evolving Investment Treaty Arbitrator?’ European Journal of International Law (forthcoming, 2018).
 International Centre for Settlement of Investment Disputes.
 Malcolm Langford, Daniel Behn and Runar Lie, ‘The Revolving Door in International Investment Arbitration’, 20(2) Journal of International Economic Law 301 (2017), https://doi.org/10.1093/jiel/jgx018 (accessed July 15, 2017). It is part of a data-focused special issue: see Wolfgang Alschner, Joost Pauwelyn and Sergio Puig, ‘The Data-Driven Future of International Economic Law,’ 20(2) Journal of International Economic Law 1 (2017).
 These figures include 38 additional cases not reported in our JIEL article.
 Langford, Behn, and Lie, supra n 11, sec. V.
 Cases are included in the evaluation from their constitution date and until they are either discontinued, settled or finally resolved through a decision in the form of an award. Cases that are pending as of January 1, 2017 have a concluding date as of January 1, 2017 for the purposes of our analysis.
 One arbitrator has been excluded from the originally reported version in JIEL as the inclusion of the annulment proceeding was deemed to overstate the degree of double hatting.
 Langford, Behn, and Lie, supra n 11, sec. IV.
 Martin Shapiro, Courts: A Comparative and Political Analysis (Chicago: Univ. Chicago Press, 1981); Alec Stone Sweet, ‘Judicialization and the Construction of Governance,’ 32 Comp. Polit. Studies 147 (1999) . Quote from Stone Sweet.
 See Catherine Rogers, ‘The Vocation of the International Arbitrator,’ 20 Am. U. Int’l Law Rev. 957 (2004), 968-9; and more recently Magdalene D’Silva, ‘Dealing in Power: Gatekeepers in Arbitrator Appointment in International Commercial Arbitration’, 5 Journal of International Dispute Settlement 605 (2014).
 Thomas Buergenthal, ‘The Proliferation of Disputes, Dispute Settlement Procedures and Respect for the Rule of Law,’ 3(5) Transnational Dispute Management (2006).
 We note that when international commercial arbitration takes on a more public character the same concerns can arise.
 Shapiro, supra n 18; Stone Sweet, supra n 18.
 See Sands, above n 2, 28-49.
 Ibid, 31-32.
 Ibid, Part I, Article 2.
 Telekom Malaysia Berhad v. The Republic of Ghana, PCA Case No. 2003-03, UNCITRAL, Settled.
 Consortium R.F.C.C. v. Kingdom of Morocco, ICSID Case No. ARB/00/6, Decision on Annulment (18 January 2006).
 Republic of Ghana v. Telekom Malaysia Berhad, Hague District Court, Challenge No. 13/2004, Petition No. HA/RK 2004.667, 18 October 2004; Challenge 17/2004, Petition No. HA/RK/2004/778, 5 November 2004.
 Ibid, Challenge No. 13/2004, Petition No. HA/RK 2004.667, 3.
 For a much more recent example of double hatting giving rise to accusations of conflict of interest, see Luke Eric Peterson, ‘As Damages Phase Unfolds in Pakistan Mining Case, a Challenge is Lodged against Stanimir Alexandrov – Citing his Client’s Alleged Interest in a Rarely-Used Valuation Method under Scrutiny,’ Investment Arbitration Reporter (11 July 2017).
 It might be possible for some arbitrators to argue that there is no actual or perceived risk for arbitrators as they act for both claimants and respondents – making strategic action difficult. But there is only a very small group of double hatters that are regularly appointed as arbitrator by both claimants and respondents (and as presiding arbitrator) and that likewise act as counsel for both types of parties.
 On this point, Barton Legum, a prominent counsel and arbitrator based in Paris, has noted that: “the pool of qualified arbitrators is already ‘vanishingly small’ and that it would be problematic for the users of the arbitration system if efforts were made to exclude all practicing BIT counsel from this pool.” Luke Eric Peterson, ‘Arbitrator Decries “Revolving Door” Roles of Lawyers in Investment Treaty Arbitration,’ Investment Arbitration Reporter (25 February 2010).
 Ibid, 2.
 Susan Franck et al., ‘International Arbitration: Demographics, Precision and Justice,’ ICCA Congress Series No. 18: Legitimacy: Myths, Realities, Challenges 33 (2015).
 Legum states: “I have a lot of sympathy for those who say you need arbitrators with skill and experience. How on earth does somebody get established as an arbitrator if he or she never gets a chance to start? So I think inevitably there has to be some overlap. But there may be a stage in one’s career when it becomes sensible to do one thing or the other.” Peterson, supra n 32, 2.
 D’Silva, supra n 19.
 Yves Dezalay and Bryant Garth, Dealing in Virtue: International Commercial Arbitration and the Construction of a Transnational Legal Order (Chicago: Uni. Chicago Press, 1996), 109.
 See most significantly, Catherine Roger’s Arbitrator Intelligence project, www.arbitratorintelligence.org (accessed July 15, 2017).
 D’Silva, supra n 19.
 Article 8.30(1) of the Canada-EU Comprehensive Economic and Trade Agreement stipulates that members of the Tribunal: ‘[…] shall be independent […] In addition, upon appointment, they shall refrain from acting as counsel or as party-appointed expert or witness in any pending or new investment dispute under this or any other international agreement.’
 For a detailed assessment of the rules governing international investment arbitration and their relation to double hatting, see Nathalie Bernasconi-Osterwalder, Lise Johnson and Fiona Marshall, Arbitrator Independence and Impartiality: Examining the Dual Role of Arbitrator and Counsel (Winnipeg: IISD, 2010).
 We note, however, that such initiatives on their own are likely to have only limited effects: see Taylor St. John et. al., ‘Glass Ceilings and Arbitral Dealings: Explaining the Gender Gap in Investment Arbitration,’ Gender on the International Bench, PluriCourts-iCourts Workshop, Oslo, March 23-24, 2017.