This is a transcript of a speech delivered by Hilary Heilbron QC from Brick Court Chambers, London, at the 14th Annual Clayton Utz / University of Sydney International Arbitration Lecture in Sydney on 25 November 2015.

It is indeed a great privilege and a real honour to be asked to give the 14th Clayton Utz/Sydney University International Arbitration lecture this evening in one of my favourite cities. I am extremely humbled to follow in the footsteps of such eminent past speakers.

In those fourteen years we have seen many changes in international arbitration as it moves to adapt to the increased global demand. Changes both in relation to international commercial arbitration and also in investor-state arbitration. But at its core remain the same principles and expectations which attracted parties to this mode of dispute resolution many centuries ago.

My topic this evening addresses some of these expectations and enquires whether international commercial arbitration is becoming too unpredictable, particularly in the sense of outcomes, and asks whether this is a price parties are nonetheless prepared to pay in order to obtain a bespoke method of dispute resolution. In particular I want to consider the impact of this exponential growth of international arbitration in recent years with the resultant involvement of many new players in the field, both as arbitrators and practitioners emanating from many different jurisdictions.

International commercial arbitration, a consensual process, is chosen because disputants want a private, flexible, speedy and effective means of resolving their disputes with significant input from the parties and the relative ease of enforcement under the New York Convention.

Nonetheless, however much we, as lawyers, strive to perfect the arbitral process, what parties and businesses actually want is to avoid the process altogether or at least to settle the dispute early, once arbitration has been commenced, often to preserve business relationships and commercial reputations. There are also the practical matters which disputes raise for corporations, such as how large a reserve should be made in the accounts; how the likelihood of success or failure will impinge on other elements of the existing and future business; whether managerial heads should roll; and what will be the cost.

Such realities and the desire for reasonably predictable outcomes are something that we, as practitioners, often lose sight of in our endeavours to make international arbitration as a process as efficient and effective as possible. Importantly, these business considerations militate strongly in favour of legal advisers being able to advise their clients, within reasonable parameters, of the chances of success or failure and how much they may be liable to have to pay or are likely to receive.

It is clear that a lawyer's subjective assessment of the factual and legal merits of a case can only be part of the answer. For the question is not how the lawyer will himself or herself perceive the case, but how the lawyer can predict how the tribunal will perceive the case having heard both sides.

In other words, although the outcome of any arbitration – or for that matter any case - depends on many factors, including the evidence and skill and persuasion of the legal representatives, this exercise also requires an assessment of how the particular tribunal will assess and react to the evidence and submissions.

Predictability

I shall start by discussing more precisely what I mean by predictability. Prediction pervades every aspect of our life, yet we rarely stop and dwell on it. It is a statement of what will or may happen in the future, but it cannot be a certainty. It usually is dependent on an act of a third party or some unknown or uncertain future event. It takes many forms.

  1. The global economy is dependent on a variety of predictions from predictions as to international global cycles, to predicting currency fluctuations and other economic indices.
  2. In the professional arena, doctors predict chances of recovery; engineers whether structures will withstand certain loads, and lawyers outcomes of cases.
  3. In the professional arena, doctors predict chances of recovery; engineers whether structures will withstand certain loads, and lawyers outcomes of cases.
  4. At the other end of the scale weather forecasters predict the weather,writers of horoscopes predict what will happen to you next week.

All these predictions are based on the individual's assessment of the facts and his or her experience of similar situations - and sometimes, prejudices. A critical component to forecasting will often be consideration of evidence of past behaviour, statistics or other historical information.

Legal opinions whether as to merits or damages are no different.[1] Legal advisers do not always get it right. What is clear is that even allowing for the fact that clients do not always take advice, there are a large number of lawyers, particularly on the losing side, who get it wrong. Or as Voltaire put it: "The history of human opinion is scarcely anything more than the history of human errors."

There have been a variety of studies which show that expertise and experience do not necessarily make someone a better reader of the evidence. Philip Tetlock in his book "Expert Political Judgment: How Good is it? How can we know?"[2] found that distinguished experts of political science and economics not only fare no better than less specialised predictors in reading emerging situations, but also that the more over-confident the expert, the less likely he or she was to be right. It is also clear that some people are better forecasters than others and some are "superforecasters" as Philip Tetlock so describes them in his latest book.[3]

The question therefore is whether in the arbitral context the difficult task of predicting or forecasting outcomes is becoming even more difficult? I propose to address the subject under three main headings:

  1. First, the increasingly wide diversity in the potential make-up of tribunals and how tribunals are chosen;
  2. Second, the discretionary powers of tribunals in its widest sense; and
  3. Third, the decision making process of tribunals and the dynamics at play in such process.

There is, of course, considerable overlap between these three elements.

Tribunals

It is self-evident that different tribunals will react in different ways. As Roy Cohn, a famous US attorney, once said "I don't want to know what the law is, I want to know who the judge is."

Where commercial disputes are determined before national courts, at least in jurisdictions such as yours and mine and many others - although unfortunately this is not always the case - a party and its legal team can expect judges of a certain quality, specialism, experience, and integrity to hear their commercial disputes. The lawyers practicing before them will know the judges either personally or by reputation and can read their judgments.

That does not mean that the result before national courts is predictable – far from it – but save in countries where the judiciary is corrupt, of a low standard or a mouthpiece of its government, there is likely to be a surer foundation on which to assess how a single judge or appellate court will react.

In contrast to national courts, the variety of decision makers in international arbitration is potentially infinite. Whereas historically there was a smaller body of well-known and experienced arbitrators, variously pejoratively described as the "mafia" or the "elite", and perceived to be a few well-known arbitrators from the West trying disputes between Western countries, today we find tribunals composed of arbitrators coming from an increasingly wide trans-global, multi-ethnic and diverse gender, pool. This wider diversity is to be hugely welcomed, but in its wake it inevitably makes outcomes less predictable.

Statistics as to new players in the arbitral field, both arbitrators and lawyers, are largely anecdotal, but there is some statistical support from LCIA and ICC figures for an increase in the number of first time arbitrators and of arbitrators coming from a wider range of countries, including third world countries.[4] As for the arbitral lawyers, anyone attending arbitral conferences could not but notice their increase in size and frequency, as well as the presence of many unfamiliar legal faces from firms of lawyers from every corner of the globe openly admitting that they want to get involved in arbitration or sit as arbitrators.

The typical three person arbitral tribunal provides a huge potential for every kind of diversity. Let me list ten examples.

  1. The members of the tribunal may vary as to their experience of arbitration and its process or even dispute resolution in general. Sometimes an arbitrator may be completely new to the field and be known only to the party or the party's legal representative appointing him or her.
  2. The members of the tribunal may come from different cultures and have different languages as their mother tongue.
  3. They may be qualified in different laws, whether civil or common law: and not necessarily that of the law of the seat or the contract.
  4. They may vary as to industry specialism. They may not even be lawyers, but could be accountants or engineers.
  5. They may vary as to their ability as lawyers and arbitrators of fact.
  6. They may have different views as to their roles and their attentiveness.
  7. There will be a mixture of personalities from the quiet and thoughtful to the loud and forceful.
  8. They may have different availability contributing to the speed with which the arbitration can be conducted, partly ameliorated by new institutional rules on disclosure of availability now demanded, for example, by the ICC and LCIA.
  9. Moreover, the composition of the tribunal is likely to mean that the three members may not only have not sat with each other before, but may not even know each other or anything about each other before sitting together as a panel, save what is on their curriculum vitae.

Thus assessment of how a tribunal as an entity or homogenous whole will react may at times be very difficult to predict. Importantly, international arbitration is a global form of dispute resolution. Tribunals are increasingly likely to include new players from less sophisticated jurisdictions and emerging markets such that the percentage of tribunals where all or the majority of the arbitrators are well known and experienced is likely to decrease over time. Any discussion of the topic cannot therefore be confined to the more cosy world of familiar faces and reputations with whom many of this audience are familiar and in which arena we mostly tend to operate. The landscape is much wider.

What remains significant however is that given that the past behaviour of decision makers is clearly very relevant to predicting outcomes, this will often be a very limited tool in arbitration where there is little knowledge as to how a particular three person tribunal will react. Yet prediction has an added imperative in arbitration over litigation because of the finality of the decision, save in very limited circumstances.

Such limitations are one of the reasons why parties desire to have as much control over the constitution of the tribunal as they can.

Choosing a Tribunal

Gary Born states a common view that[5]:

"...one of the defining characteristics and attractions of the arbitral process is the parties' ability to select a tribunal that will decide their dispute in a competent, expert manner, rather than relying on random selection of a national court judge."

But while this may be the ultimate aim, I question whether this is always the case. For the choice of an international arbitral tribunal may be far more random than had been originally envisaged by the parties.

The very recent 2015 Queen Mary/White & Case Survey[6] indicated that 38% of respondents regarded the ability to choose arbitrators as one of the three most valuable characteristics of international arbitration. Yet in its 2010 survey 50% of those surveyed were disappointed with the arbitrator's performance in various respects. Even allowing for sour grapes of the losing party, this is a high percentage.

There is no uniform method by which parties can choose their tribunal in international commercial arbitration. The variety is infinite. Although rare, the parties can even specifically choose named arbitrators.

Often, not enough thought is given to how a tribunal is to be formed. Parties at the end of long and hard negotiations, are reluctant in the euphoria of a successful deal to devote much thought to what might happen if it all goes wrong and many such clauses are incomplete or pathological clauses.

Sole arbitrators are either the result of a joint choice or an institution choice and I will not dwell on those as, on the whole, the larger cases are dealt with by three person tribunals. Similarly I will not discuss court appointments made to fill vacancies or meet defaults. However, for tripartite tribunals, there are many variants.

By far the most common method, according to the Queen Mary/White & Case study in 2012, about 75%, is for the parties each to nominate its own arbitrator and the co-arbitrators then to nominate the chair or presiding arbitrator.[7] The most common alternatives are for each party to nominate its own arbitrator and an institution nominates the chair or for an institution to nominate and appoint all three members of the tribunal.[8] Inherent in all three methods is the assumption that each party will have an equal stake in forming the composition of the tribunal.

Parties and their legal advisers go to considerable lengths to choose their appointees or to suggest criteria if the appointment is to be institution based. Short of knowing the individual concerned, parties look to directories and other information about potential candidates. Sometimes a party will even hire investigators at huge expense looking for an individual's legal predilections to determine whether he or she is likely to be sympathetic to and supportive of its case, short of impartiality, so far as this information is available. On other occasions parties may interview the potential nominees within clearly defined guidelines, such as not discussing the merits of the case or the candidate's views, but asking about such matters as the potential arbitrator's "expertise, experience, ability, availability".[9]

By so nominating its arbitrator a party may feel it has some input into the tribunal's composition, but the extent to which this can determine the outcome is probably illusory. At best it will ensure that it has one decision maker of a certain quality and experience whom it believes may be sympathetic to its case. But one out of three is not great odds for prediction.

I say that the extent of one party's input may be illusory for three reasons:

First, once appointed an arbitrator has to be independent and impartial. He or she cannot be the advocate for the party. In fact he or she should forget which side made the appointment. In their new 2014 Rules the CPR have introduced a screened selection procedure which parties can opt into whereby the arbitrator never knows which party originally selected him or her.[10]

Some argue that party appointed arbitrators should go further and ensure that the case of their appointing party is properly put or that they should assist with cultural issues, whilst maintaining their impartiality – a so-called middle ground.[11] However if the party appointed arbitrator over-steps the mark and is not impartial, then in any event his or her views are likely to be cancelled out by the other side's arbitrator.

Of course, there are many cases where both sides appoint equally experienced arbitrators: but there is no guarantee, particularly as the field becomes larger. Moreover, a respondent may be at an advantage, because at the time its arbitrator is nominated, it is likely to know the name of the arbitrator the claimant has nominated, such name usually appearing in the Request for Arbitration.

Second, the arbitration clause is predicated on the fact that the parties to the arbitration will be the named parties to the contract, but this, of course, is not always the case. Frequently non-signatories are brought in as claimants and respondents and issues may or may not arise as to jurisdiction. Nonetheless they remain claimants and respondents for the purpose of the arbitration and the award and more particularly for the purpose of nominating an arbitrator. Joint parties, whether or not non-signatories, may have to make a joint choice which means compromise or, if they do not represent collectively two separate sides, the institution appointing the tribunal, thus diluting each party's direct input into the choice of its own arbitrator.[12]

Third, and critically, the assumption of equal input to the composition of the tribunal ignores the importance of the chair or presiding arbitrator. It is this third feature which I believe is not given sufficient prominence. There appears to be little research or academic writing on the impact of the chair or presiding arbitrator on arbitral decisions. But the chair or presiding arbitrator is crucial to the arbitration. Let me give five examples:

  1. It is the chair who usually dictates the pace of the arbitration as he or she does most of the drafting of the procedural and other orders and, more especially, the award or awards.
  2. Whatever the extent of prior deliberations, the award will usually be in the words of the chair and chairs are often given considerable discretion as to the way the award is written. Moreover, depending on the conscientiousness of the co-arbitrators, the chair's work may not necessarily be fully cross checked.
  3. The co-arbitrators may delegate day to day procedural decisions to the chair alone with the agreement of the parties and in any event in practice the chair tends to have the last word in procedural decisions unless fundamental. [13] This can be subject to revision by the full tribunal.[14]
  4. It is the chair who sets the whole tone of the arbitration and conducts the hearings.
  5. Finally, it is the chair or presiding arbitrator who manages the co-arbitrators. An industrious chair and two less industrious, or even lazy, co-arbitrators - or more particularly inexperienced or novice co-arbitrators who will often pay deference to the chair, can mean the chair has a disproportionate influence or impact on decisions.

However, despite the importance of the position and the presiding arbitrator's pivotal role, the procedure for selecting a chair of a tribunal where an institution is not involved, is in my experience haphazard and unstructured with no universal practice as to what process is to be used for what type of case or situation. The only apparently accepted view is that, unless otherwise agreed by the parties, the presiding arbitrator should be a different nationality from the parties.

The process is usually dictated by what the party-nominated co-arbitrators think appropriate, and each may have a different view on this, unless the parties have previously indicated a preferred approach, which is less common than one might expect. Often the choice is determined by the need to choose a chair by a certain deadline in the arbitration agreement or to enable the tribunal to be constituted so as to make interim rulings.

One of the questions which remains unclear is whether in choosing a chair, the co-arbitrators should act as their appointing parties' agent or act as principals in their own right. The accepted view is that parties should not have a veto on appointments other than for good faith reasons such as conflicts or lack of experience. But to what extent should they influence the choice made by the co-arbitrators?

The IBA Guidelines on Party Representation permit communication between the parties and party-nominated arbitrators on the choice of presiding arbitrator, although the 2014 LCIA Rules require notification to the Registrar before this is done. On the other hand, the IBA Guidelines permit direct communication between the parties and candidates for presiding arbitrator only where all parties agree.

Where an institution is involved and agreement is not possible then the default position is usually for the institution to choose the chair.

According to ICC statistics, in about 70% of cases agreement is reached on a presiding arbitrator, but the issue in the context of predictability is not the fact of agreement, but who is chosen and by what method. As many of you here will likewise have experienced, it can include any of the following. It should not, of course include one arbitrator communicating with the party nominating him or her without the co-arbitrator being aware of this fact.

  1. The arbitrators, having cleared conflicts and ensured that the chair is not the same nationality as the parties, just choose a chair between themselves without resort to the parties. This may mean a narrow base of candidates.
  2. The party appointed arbitrators discuss requirements for presiding arbitrator with their respective appointing parties either by e-mail or on the phone or in person and then make a choice.
  3. The third scenario involves using candidates' names and this brings up the issue of lists whether or not a prior discussion has taken place.
    1. Do the arbitrators jointly produce a short list and then ask each party to indicate its order of preference without this being disclosed to the other side? If one side gets its second choice and the other its fifth is this satisfactory?
    2. Do the party appointed arbitrators just present the parties with a list and ask if there are any principled objections?
    3. Do they ask the parties to try to agree a short list?
    4. Do the arbitrators ask each side independently to produce a list, but this is even less likely to produce a candidate on both lists and if a party is chosen from one list and not the other, then one side would be justifiably dissatisfied.
    5. How many rounds should the party appointed arbitrators permit.
    6. Should the party appointed arbitrators indicate that if agreement cannot be reached they will impose someone?

Importantly there is also often an imbalance where one co-arbitrator is new to the field and may be swayed, by lack of knowledge or otherwise, by the other co-arbitrator's choice as to arbitrator thus in practice diminishing the input of one party. The reality is that the chair is often a compromise, or if no agreement can be reached because of tactical manoeuvres or otherwise, then the decision is left to an institution or occasionally to the court.

It would appear that the method of choosing a third arbitrator is a hangover from when there were umpires, when the umpire did not descend into the decision making process unless and until the party appointed arbitrators had disagreed[15]. 19th century case law [16] makes it clear that although the two arbitrators were expected to use their judgment to concur in and choose, rather than leave to chance, the identification of a third person as the umpire, the parties could agree to let co-arbitrators choose the umpire by any method, even in one case by drawing lots out of a hat, provided the candidates were fit and acceptable.

Thus, whether the chair is appointed by the party appointed arbitrators or the institution, the actual parties, or realistically their external counsel in most cases, may have no or only limited input into the central role of the chair and no or only limited knowledge of his or her experience and capabilities.

If past behaviour of the tribunal as a whole and its individual components is a key indicator of prediction then this is an important element in the prediction puzzle which lawyers have to grapple with. There is a limit to what a legal adviser can glean from a perusal of directories as to how a three person tribunal will gel and react. It seems to me that this lack of uniformity of procedure as to appointment of chairs of panels or presiding arbitrators is a lacuna to which the arbitral community should give more prominence and I shall come back to it when looking at the way forward later in this lecture.

Discretion

Let me move now from diversity to the second element of prediction, discretion. Roscoe Pound described judicial decision making as "a sort of judicial slot machine". So how does unpredictability impinge on the arbitral and decision-making process? One of arbitration's great assets is its flexibility to adapt the process to meet the case in question untrammelled by prescriptive national court rules.

In the context of international arbitration flexibility is often a synonym for discretion. Most institutional rules give tribunals wide powers as to the way in which they conduct the arbitration, subject to the agreement of the parties and any mandatory national laws of the seat.

The increasing number of such rules is reflective of the expansion of international arbitration and they are often geared to their own particular culture and users. Their universal purpose is to provide an efficient, speedy, cost effective and fair procedure.

But not all arbitrations are conducted under institutional rules: many are ad hoc and thus even the limited framework provided by institutional rules is absent. Guidelines such as the IBA Guidelines on the Taking of Evidence and Party Representation provide additional attempts at a more uniform procedural approach crossing borders, but their interpretation at the end of the day will vary from tribunal to tribunal. Experienced arbitrators will be familiar with how international arbitration is conducted, but we have all seen litigation lawyers who are unprepared to leave behind their national civil law rules or US trial lawyers still finding it difficult to differentiate between an arbitration and a jury trial.

Discretion can be exercised in various contexts.

First, there is what one can term basic procedural discretion, most of which decisions will have no real impact on the end result such as extensions of time. Then there are those discretionary decisions which are more significant in terms of prognosis of outcomes, a sort of substantive discretion, such as whether to allow expert evidence; disclosure; consolidation and joinder of parties permitted by some institutional rules; preliminary issues and bifurcation; the power to make peremptory orders and the powers to continue in the absence of a respondent; and the granting of conservatory and interim measures.

Finally, there are those substantive decisions which may be critical to the ultimate outcome or result and have a discretionary element, such as determining the "appropriate" law to govern the substantive merits of the dispute if the parties have failed to agree or the interpretation of imprecise contractual and statutory language.

Discretion can be widely or narrowly interpreted. Take for example the IBA Rules of Evidence and many institutional rules which leave it to the arbitrator to determine the admissibility, relevance, materiality and weight of the oral and documentary evidence. A US lawyer, particularly a trial lawyer who has not sat as an arbitrator before is likely to give a much wider interpretation to such rules than an experienced civil arbitral practitioner. The result of such diversity of legal cultures is that arbitrators often tend to "transnationalise" their approach to such discretionary decisions.

In the context of the exercise of any of these discretionary decisions and in relation to the ultimate conclusory decision on the factual and/or legal merits, the inter-play and dynamics of the members of the tribunal is therefore critical.

Decision Making – dynamics

So let me turn to my third element of prognosis: the dynamics of tribunals. I can only touch on this as it is a huge subject. The assumption that judges or arbitrators act purely rationally and logically and make rational choices has been questioned, as it really explains only part of the judicial decision-making process.

While most studies undertaken to date are aimed towards counsel's better understanding of how to present a case before a judicial or arbitral tribunal, these factors also impinge on the ability of lawyers to advise their clients as to outcomes. Studies of judicial and arbitral behaviour based on cognitive science and behavioural psychology[17] show that judicial decision making comprises both the conscious deliberative process and the intuitive sub-conscious element and there is nothing to indicate that arbitral decision making is any different.

As Benjamin Cardozo said: while "we [as judges] may try to see things as objectively as we please...we can never see them with any eyes except our own." All arbitrators bring to the deliberation table their knowledge and experience of their own cultures and general experiences of life and social inter-action and with it the unconscious and subconscious influences on decision-making and behavioural tendencies of individual arbitrators. These studies show the presence of:

  1. First, heuristics or mental shortcuts that we all make to resolve complex situations which in turn lead to mistakes. Examples are over-reliance on hindsight in determining probability and, in calculating damages, anchoring from a suggested figure;[18]
  2. Second, the propensity for decision makers to create narratives or stories to supplement the evidence with inferences; and
  3. Third, the various cognitive biases and prejudices we all have, but do not necessarily appreciate or acknowledge. As Luther Burbank said: "It is well for people who think to change their minds occasionally in order to keep them clean. For those who do think, it is best at least to rearrange their prejudices once in a while."

Of particular relevance to international arbitration is cultural bias. Parties will often appoint someone as an arbitrator whom they believe understands their own culture, though the presiding arbitrator is usually culturally neutral. Individuals with different backgrounds and life experiences will interpret the same issues or evidence differently based on different systems of beliefs which in turn can affect outcomes. Thus where there are balancing acts to be made, different arbitrators will attach different weight to different matters. It is difficult to eradicate totally the norms in one's own legal system.

There has also been considerable psychological research on the impact of group decision making compared with individual decision making in the context of US juries and judges' decisions.[19] The conclusions were largely inconclusive depending much on the various factors involved in each experiment. But two themes did emerge.

First, where the error of one decision-maker is obvious and could easily be demonstrated such as calculation of quantum, the other member or members of the group would point it out. However, where the issue was complex or involved a value judgment the group tended to produce a more extreme outcome than that which would have reflected the members' pre-deliberation preferences. The explanation was that people strive to perceive themselves and to be perceived by others favourably and if they see the view moving in one direction will adopt that view, but take a more extreme position such that the initial view is strengthened as a result of group deliberation. On the other hand, it can be argued that diversity counters prejudice.

It is true that as Benjamin Franklin said: "If everyone is thinking alike, then no one is thinking". But even thinking arbitrators may reach different results depending on the make-up of the tribunal: the nature of the dialogue, the cross-cultures, their inner and sub-conscious beliefs and prejudices; and the inter-relationship and familiarity between the members of the panel which in turn emanates from the characteristics and culture of each arbitrator.

Such dynamics add further uncertainties as to the ultimate outcome. So where does that leave us?

Suggestions

The rise of international arbitration, like the sea's incoming tide, appears unstoppable, yet those involved in disputes which surf its waves may ponder to discern that outcomes may be increasingly unpredictable. Do we ride the ocean of international arbitration regardless or do we pause to see if we can predict more successfully where the ship will anchor? I venture to suggest that the balance of advantages outweighs its disadvantages, but that does not mean that it cannot be improved. Legal prediction will remain an art not a science.

So how can we reduce, if at all, the potential for increasing uncertainty already inherent in predicting outcomes in international arbitration given the increased diversity of tribunals and lawyers. Let me examine briefly three options.

  1. First, the introduction of a more robust appellate system to provide more uniformity. Currently appeals are almost universally precluded save where there is procedural unfairness or the absence of jurisdiction. England, exceptionally, allows a limited appeal on points of law.[20]

    Ambrose Pierce described an appeal in law as "put[ting] the dice into the box for another throw." Over 75% of respondents to the 2015 Queen Mary/White & Case survey were against introducing an appeal mechanism. I believe that an appeal is inimical to the whole process based on finality and the remarkable success of the New York Convention. If parties want to go through appellate procedures then they should opt for national courts.
  2. Second, the provision of more reported redacted decisions by institutions. Given that substantive legal issues will usually involve national law[21] such decisions are likely to be of benefit in predicting outcomes most often in relation to challenges to appointment of arbitrators or procedural issues where how a previous tribunal has acted may prove useful guidance, particularly where some form of transnational law is applied.

    But, as Professor Gabrielle Kaufman-Kohler found in research for her Freshfield's lecture in 2006, even where past awards were put to arbitrators they tended to do what they wanted with past cases and there is no clear practice. It remains doubtful therefore how significant the absence of precedent really is in predicting outcomes, as opposed to providing transparency, particularly with the increase in numbers of institutions all providing redacted decisions on rule specific issues.
  3. Third, improvements in the method of appointing tribunals and in particular the presiding arbitrator.

    I am not a protagonist for an increase in institutional appointments of tribunals, nor does this appear from studies to be what parties actually want. By and large the procedure for appointment of party nominated arbitrators works well and there are sufficient protections in place to ensure that the arbitral nominating and appointment process is both fair and just.

    However, the same cannot always be said for the chair or presiding arbitrator. Again I would not advocate that parties automatically relinquish the practice of the co-arbitrators choosing the chair in favour of institutional choice, but there is not a sufficiently universal acceptance of the need for prior consultation with the parties or the method by which the appointment of presiding arbitrator should be carried out. I will suggest three ways this can be achieved, but there are doubtless others.
    1. First, parties can be more vociferous in demanding input into the choice of chair at the outset of the dispute or with mutual agreement interviewing the prospective candidates in accordance with accepted guidelines.
    2. Alternatively, they can include in their arbitration clause a phrase after the statement that the co-arbitrators shall jointly choose the presiding arbitrator such as "subject to prior consultation with the parties".
    3. Thirdly, the international arbitral community can and, in my view, should adopt a more uniform approach. It has shown itself on many occasions to be able to provide uniformity of practice, such as the use of Redfern Schedules or the provision of detailed first Procedural Orders. It does not need more Guidelines. Even experienced arbitrators do not invariably adopt a uniform practice and inexperienced arbitrators or practitioners do not suggest it.

It should be common practice for co-arbitrators, whoever initiates the list, to consult with the party or parties who appointed them as to a chair or presiding arbitrator, save in exceptional circumstances e.g. where there is extreme urgency to constitute a tribunal. A chair whom both parties buy into may not necessarily produce a more cohesive tribunal, nor guarantee to improve the certainty of outcomes, but given the chair's pivotal role, it will at least remove some elements of uncertainty from the process and enable parties to have a more equal hand in the ultimate composition of the tribunal. It arguably could making choosing a chair more difficult, but default procedures exist if this happens.

Open-minded tribunals of experienced and knowledgeable arbitrators who deliberate conscientiously are the aspiration; mixed cultures are inevitable and to be welcomed; sub-conscious beliefs will still pervade decisions and counsel will try to attune their arguments to the tribunal they have.

Conclusion

So let me conclude: Prediction of outcomes in international arbitration is difficult. Does the increased diversity of tribunals with its concomitant effect on deliberations and the exercise of discretion increase this? The answer is probably yes, although it depends ultimately on the make-up of the tribunal in question.

But at the same time there is no doubt, as evidenced by the huge explosion in the numbers of arbitrations worldwide that parties will opt to take that risk in favour of a tailored approach to the resolution of their disputes and the other benefits of arbitration. It is a price they are prepared to pay.

Footnotes

[1] Tribunals only award on average less than 40% of the amount claimed according to the recent survey carried out by PCW principally of investment treaty awards: "International Arbitration Awards – closing the gap between claimants and respondents", 2015.
[2] Princetown University Press: 2005.
[3] "Superforecasting: The Art and Science of Prediction" Philip Tetlock; Dan Gardiner: Random House 2015.
[4] LCIA figures indicate that in 2012 and 2013 an average of approximately 19% of arbitrators were appointed by the LCIA for the first time, although this does not indicate whether or not they had sat elsewhere before.

A comparison of the ICC statistics shows that whereas in 2004 arbitrators appointed or confirmed came from 61 countries. In 2014 this had risen to 79 countries, an increase of a third on the 2004 figure, including several third world countries. [5] International Commercial Arbitration, Second Edition page 1686.
[6] October 2015.
[7] However, informal statistics from the ICDR obtained by the author indicate that this may be on the high side in relation to party appointed arbitrators (though not in relation to the presiding arbitrator), but this may be because by choosing ICDR the parties have bought into the list and rank procedure provided by Article 12(6) of the International Dispute Resolution Procedures 2014.
[8] Where an institution is involved, as opposed to an ad hoc arbitration, the institution can act as a safety net for wholly unsuitable arbitrators. Some rules, like SIAC and ICC require the institution to affirm or approve the nomination (Article 13, ICC Rules 2012; Article 5 LCIA Rules 2014; Article 6.3 SIAC Rules ), but others, like the UNCITRAL Rules and ACICA provide for direct designation or appointment by the parties (Article 9(1) UNCITRAL Rules; Article 10 ACICA Rules 2011). ICDR provides a list procedure (Article 12 International Dispute Resolution Procedures (Including Mediation and Arbitration Rules).
[9] Article 8 of the IBA Guidelines on Party Representation in International Arbitration; See also Chartered Institute of Arbitrators Practice Guideline 16: The Interviewing of Prospective Arbitrators.
[10] Rule 5.4 CPR Rules 2014 (International Institute for Conflict Prevention and Resolution).
[11] Practical Guidelines for Interviewing, Selecting and Challenging Appointed Arbitrators in International Commercial Arbitration: Doak Bishop & Lucy Reed.
[12] Eg. Article 8.1 LCIA Rules 2014.
[13] Eg. Article 31 UNCITRAL Rules.
[14] Eg. ACICA Rules Article 17(9).
[15] E.g. section 9(1) of the English Arbitration Act 1950 in its unamended form.
[16] Re Cassell (1829) 9 B & C 624; European and American SS Co v Crosskey (1860) 8 CB (NS) 397; In the Matter of Arbitration Between Tunno and Bird; (1833) 5 Barnewall and Adolphus 488 110 E.R. 870.
[17] "Cultural Bias in Judicial Decision Making", Masua Sagiv Boston College Journal of Law & Social Justice Vol 35/Issue 2, May 2015; "Unconscious Influences on Judicial Decision-Making: The Illusion of Objectivity", Honourable John F. Irwin and Daniel L. Real; "Understanding Judicial Decision-Making: The Importance of Constraints upon Non-Rational Deliberations", John N. Drobak and Douglas C. North; "Arbitrator Decision-Making: Unconscious Psychological Influences and What you can do about them", Edna Sussman.
[18] The Kaplan Lecture 2012, "Arbitral Decision-making: Art, Science or Sport?" Lucy Reed; 2012.
[19] See the summary in "Judicial Decision-making: A Behavioural Perspective", Doron Techman and Eyal Zamir.
[20] Section 69 Arbitration Act 1996.
[21] "Arbitral Precedent: Dream, Necessity, or Excuse?" Gabrielle Kaufmann-Kohler.

Clayton Utz communications are intended to provide commentary and general information. They should not be relied upon as legal advice. Formal legal advice should be sought in particular transactions or on matters of interest arising from this bulletin. Persons listed may not be admitted in all states and territories.