Imagine this situation. You are waiting to be interviewed by two senior lawyers from a major city law firm. This may well be it. The career breakthrough you have been waiting for. After years of diligent legal training and professional experience, publishing articles, building a network, and speaking at conferences, you are well on track to securing this highly coveted legal job.
The interview starts. The senior lawyers ask open-ended questions about your background. The atmosphere is congenial, even jovial at times. Then they zero in on some legal views you have expressed in the past. The questions become more specific. They want to know what you would say in this or that hypothetical scenario. The tone remains cordial. After some back and forth, the senior lawyers look at each other, excuse themselves and withdraw from the room.
Now you are a little nervous. While the atmosphere felt right, you are not sure you gave the answers they expected. But there it is: your fate is sealed. The wait continues. Suddenly the door opens. The two lawyers are smiling. They approach you and enthusiastically shake your hand: “Congratulations. The job is yours.” You are no longer a candidate. Now you are an arbitrator in the case where these two senior lawyers act as counsel. You are thrilled and could frankly jump for joy, but you preserve a solemn façade and merely nod with dignified contentment.
Welcome to the world of unilateral party-appointed arbitrators. There are, of course, variations on this theme. The interview may take place by phone or even email. The interviewers may already know the candidate well and dispense with the need for an interview altogether. The candidate may be so in-demand that the interviewers may be the ones jumping for joy if their offer of appointment is accepted.
Whatever the exact variation on this theme, there will be a constant. You, the candidate-turned-arbitrator, will be imbued with a sense of gratitude towards the appointing party. Even if you are already a busy arbitrator, feeling further in-demand is universally appreciated. As a matter of basic human nature, you will feel inclined to reciprocate the courtesy in some fashion or other. This might also help you get additional work in the future.
But there is a caveat. Under the applicable arbitration rules, which are in line with most major rules, you must be impartial. That is, you must treat the other party to the arbitration, whom you have never met or maybe even heard of before, just like the party who hired you. This creates an inherent conflict between your natural predisposition as a matter of human nature and your duty as arbitrator.
The duty of impartiality seeks to contain, but crucially cannot suppress, your predisposition towards your appointing party. A 2016 scientific study (Sergio Puig and Anton Strezhnev) shows that subduing this predisposition may simply be beyond your grasp, even if you are fair-minded. It operates even at an unconscious level. Also, if it is hard for you to detect your own bias, it will be even harder for others. Impartiality is, after all, a subjective state of mind. The likelihood that your bias will go undetected might tempt you to act on it, unconsciously or otherwise.
As a result, the ever-present risk of bias is built into this appointment method. This risk shall be referred to as soft bias. The built-in soft bias in the system has the potential to turn into actual bias at any point. Just like greenhouse gases in the atmosphere, you may not be able to see soft bias, but it is everywhere in our arbitral biosphere.
Soft bias made its official entry into the archives of investment case law when, in Supervision v. Costa Rica (2017), arbitrator Joseph Klock penned a dissenting opinion, making the extraordinary admission that the two party-appointed arbitrators – himself and Eduardo Silva Romero – were working under an “inherent” and “uncomfortable aura of conflict” permeating the proceedings. Soft bias is real.
The effects of soft bias are far-reaching, percolating through the entire fabric of international arbitration. Most ominously, there is mounting evidence that, to a high degree, soft bias turns into actual bias. This undermines not just the impartiality of arbitrators, but also the equality and consistency of the system as a whole. Soft bias thus rocks the very foundations of international arbitration.
Evidence of soft-turned-actual bias is by its nature not easy to obtain. Nevertheless, this evidence is substantial and keeps piling up.
First, the 2016 Puig and Strezhnev study shows this process can be unconscious and thus can happen even to the best-intentioned arbitrator.
Second, the 2017 Berwin Leighton Paisner survey (the BLP survey) shows that 55 per cent of arbitrators in the poll witnessed party-appointed arbitrators being biased towards the appointing party. Plus 70 per cent of arbitration counsel thought they witnessed this as well.
Third, virtually all dissents in international arbitration are authored by party-appointed arbitrators in favour of their appointing party. In support of that, there is the 2009 Albert Jan van den Berg study in relation to investment arbitration, and Alan Redfern made a similar point in relation to commercial arbitration in 2004, relying on ICC data.
Fourth, some of the most authoritative figures in the field have publicly spoken out against the problem of actual bias. These include, inter alia, Jan Paulsson, Sundaresh Menon, Nassib Ziadé and Juan Fernández-Armesto. Alexis Mourre, current President of the ICC Court, has noted that “many party-appointed arbitrators tend to favour their appointing party.” These voices are based upon decades of experience practising at the highest echelons of the profession. Hence, these are expert opinions.
Fifth, a renowned arbitrator once candidly admitted: “If the solution is clear, I go for that solution; if the solution is not clear, I go for the party who appointed me.” A principled basis for unprincipled bias. This shows that sometimes well-meaning arbitrators will be consciously biased if they are not sure what to do (“in case of doubt break the impartiality window”).
Lastly, the problem of party-appointed arbitrator bias is one of the reasons behind UNCITRAL’s drive to reform investment arbitration. That this issue features high on the UNCITRAL agenda is proof, if not of bias itself, then of the credibility of the evidence of actual systemic bias.
The effects of soft bias morphed into actual bias run deep and reverberate across the entire edifice of international arbitration. This bias has a negative effect on the following.
First, by definition, it affects impartiality. This is the effect most often mentioned in the literature. The focus here is on each party-appointed arbitrator individually. This bias will cause a party-appointed arbitrator to be partial towards that party.
Second, it may also have an impact on equality. Here the focus is not on each arbitrator individually, but on the tribunal as a whole. There is no reason to believe, as it is often said, that two biased party-appointed arbitrators will “cancel each other out”. This view assumes a lot: that both party-appointed arbitrators will be biased and that they will be biased to an equal degree. Yet experience does not bear this out. Moreover, it is bizarre to craft a justice system where more bias is prescribed as the antidote against possible inequality. The opposite is the case: bias is more likely to lead to unequal treatment.
Third, bias also has an impact on consistency. Here the focus is the broadest yet: it is on the system of international arbitration as a whole. When two-thirds of arbitration tribunals are under the influence of soft bias, the corpus of law they craft may not be the most consistent. Lack of consistency is one of the key reasons behind plans, spearheaded by UNCITRAL, to create a permanent investment court. Bias is at the root of this problem.
Fourth, this bias may even spill over onto the tribunal chairs, who often have a keen grasp of the dynamics at play. Familiar with the system, tribunal chairs may give face-saving concessions to an arbitrator whose appointing party is about to lose the case. The chair may be more likely to make these concessions, even unconsciously, if that party-appointed arbitrator is well-connected. The point is that the assumption that chairs are entirely insulated from the effects of soft bias is one that remains to be tested.
If soft bias is ubiquitous like greenhouse gases, these are some of the (more or less likely) effects on the system.
Here is where phase I of the climate change paradox manifests itself: in come the deniers.
The deniers gainsay that soft bias exists or that it is a problem.
Among the leading exponents of the deniers are Charles Brower and Charles Rosenberg, authors of the Nightingale article (2013). In that article, they make two overarching points: (1) biased arbitrators would never be appointed or effective; and (2) the current system works well, with problems limited to just “a handful of bad apples”.
Both points are incorrect. The first point ignores the possibility of unconscious bias and wrongly assumes that bias is an all-or-nothing attribute that can be easily detected amid sophisticated arbitrators. Rather, bias is more likely to be subtle and to appear in disguise. The second point ignores that the system, whatever its current state, could work far better. The problem is not limited to a “handful of bad apples”: these are but the most extreme manifestations of the ubiquitous soft bias problem. The “bad apples” are the tip of the problem; Brower and Rosenberg have missed the rest of the iceberg.
The problem of soft bias calls for a solution. Users of international arbitration deserve the best possible version of the system – not its current version. In addition to addressing the problem of soft bias, there are other arguments supporting a case for reform.
First, reform is what users actually want. A 2010 Queen Mary survey – cited in the Nightingale article – shows that most users (66 per cent) want party-appointed arbitrators to be impartial (“open-minded and fair”). They also still want party-appointed arbitrators. Thus, if a system could ensure both party-appointment and impartial arbitrators, users would logically support it.
Second, the scale of the reform required is not significant. There is no need to rewrite the arbitration rules. Small changes to the rules of appointment would suffice to address the big problems outlined above.
But there are also some obstacles militating against reform.
First, while there is a growing consensus that something should be done about soft bias, there is no consensus about what should be done exactly. Some commentators propose that arbitral institutions make all the appointments, although most users (69 per cent) in the 2017 BLP survey were against it.
Second, the current system protects vested interests. Counsel to the parties are de facto appointing authorities; a considerable power. That is why their advice to clients about the merits of the current appointment method may itself be less than impartial. Also, arbitrators have built their careers on this system.
Third, there is a sense of familiarity and comfort with the status quo.
This represents phase II of the climate change paradox in arbitration: failure to take collective action to redress the problem of soft bias.
There are, however, viable solutions within reach. The question of party-appointed arbitrators is often wrongly presented as a binary choice: the current system of party appointment versus no party appointment. This is a fallacy. Within the system of party appointment, there are options. Two are proposed here.
First proposal: blind appointments. That is, not telling party-appointed arbitrators what party appointed them. This is a most promising solution. It adresses all the soft bias problems while preserving the benefits of party appointments. The parties can still feel in control of the process, since they are involved, and can trust the system for that reason. In short, blind appointments ensure the advantages of party appointment without its disadvantages.
Two complaints have been made against blind appointments. The first is that it would not allow the party-appointed arbitrator to fulfil the role of ensuring that the appointing party’s arguments are “adequately considered”. But this role is obsolete: all arbitrators must adequately consider all arguments, as Sundaresh Menon persuasively argued in 2017. The second is that actually party-appointed arbitrators could guess who appointed them. While this would still be better than the status quo, this illustrates an important point: under a blind appointment system, arbitrators should not know what party appointed them nor the method used to appoint them, and there should be an array of methods available to the parties. This should put to rest any concerns about arbitrators guessing which party appointed them.
Two arbitral institutions have already adopted a blind appointment system. The first to do so was the Conflict Prevention & Resolution Institute in 2014, for which they won the GAR Innovation Award in 2016. Tellingly, Charles Rosenberg, co-author of the Nightingale article, endorsed this system because it eliminates “the risk of arbitrator bias in favour of the appointing party” (itself evidence that he believes there is a soft bias problem). More recently, in 2018, the Mauritius arbitration centre also embraced this system on an opt-in basis.
These two institutions are showing the way. Other major institutions should follow suit. Blind appointments should become the norm.
Second proposal: sequential appointments. Under this system, each party would prepare a list with three or more candidates – all of whom should be free of conflicts – and the other side would pick one candidate from this list. This would preserve the advantages of the party appointment system with two additional benefits: both parties would be involved in the appointment of both party-appointed arbitrators – enhancing legitimacy – and it would minimise the risk of challenges (as there would be no reason for them).
Finally, phase III of the climate change paradox: failure to take action in time may lead to a mass extinction event. In the field of investment disputes, plans to create a permanent multilateral investment court are well underway.
Justice is typically portrayed as a blindfolded lady holding a set of scales. It is in an iconic image with an ancient pedigree. It signifies the impartiality of the adjudicator, with a state of mind free of any predisposition to any party. For international arbitration, this image will not work.
The current system of unilateral party appointments gives rise to soft bias, predisposing arbitrators in favour of the appointing party. This has profound ramifications. It has negative effects on the impartiality of arbitrators, the equal treatment of the parties and the consistency of the system.
But there is reason for optimism. Strangely, that blindfolded lady may provide the key to the answer. Make party appointments blind. Arbitrators do not need to know who appointed them or by what method. Parties do not need arbitrators to know this either. Turn it into a default rule. Create an array of appointment methods, such as sequential appointments, which can be combined with blind appointments. Users want this reform. They deserve it too.
The party appointment debate mirrors the climate change paradox. First, there is a failure to accept the problem, with eyes gradually opening in the face of ever mounting evidence. Second, there is a failure to take collective action to tackle the problem, a combination of lack of consensus, vested interest and sheer comfort with the status quo. Finally, this failure may lead to a mass extinction event (the replacement of arbitration with a permanent multilateral court).
Polar bears and impartial arbitrators may have more in common than might have been previously thought.