chevron-down Created with Sketch Beta.


Reflections on Arbitral Assumptions

Neil Kaplan


  • There is a noticeable discrepancy between what arbitration counsel advocate at conferences and their actual practices, such as endorsing narrow document production but demanding extensive schedules.
  • Arbitration clauses can be designed to address common criticisms by including provisions like limiting the number of arbitrators based on claim value, restricting disclosure, setting submission lengths, excluding punitive damages, and expediting procedures.
  • Suggestions to enhance efficiency include shorter, more succinct awards, early opening hearings to engage the tribunal early, and built-in mediation pauses after early openings to encourage settlement discussions.
  • The high cost of arbitration can be mitigated by considering parties' conduct when awarding costs, capping recoverable costs, and using sealed offers to incentivize reasonable settlements, thereby controlling litigation expenses.
Reflections on Arbitral Assumptions
Andrii Lutsyk/ Ascent Xmedia via Getty Images

There are a number of things that I would like to discuss today that do not conveniently fall under one catchy title. So it will be a bit of a potpourri, but I hope none the worse for that. The common thread is my attempt to critically examine the world of international arbitration from the perspective of an experienced arbitrator, who has been working in arbitration for more than 40 years in 4 different jurisdictions.

The Gap Between Theory and Practice

There is often a mismatch between what counsel say at conferences and the way they conduct themselves in an arbitration.

For example, most counsel seem to agree that the process of document production should be narrow. Yet, in practice, we often get voluminous schedules for document production. Similarly, counsel often complain at conferences that arbitration cases have become too formal and legalistic. However, the same counsel will object to any request from the other side while referring to the tribunal’s procedural orders and relevant procedural rules—as if the arbitration proceedings were court proceedings and subject to statutory rules of court. One last example, although there are many more, counsel often complain that submissions should be brief and material. However, in their own cases, they often argue every possible argument. This approach makes the pleadings longer and drives up expenses. Submissions running to hundreds and sometimes a thousand pages are not that uncommon. They cause slower and longer awards and ramp up cost.

Arbitration clauses to minimize the complaints. Another unfortunate feature of arbitration today is the way in which so much criticism is heaped on the arbitrators, the institutions, the judges dealing with arbitration, and sometimes even arbitration counsel themselves. But how often have you seen an arbitration clause drafted in such a way so as to minimize the complaints of which we hear so much subsequently?

There are many sophisticated users of international arbitration who have access to able and experienced in-house and external counsel. So let me ask you how many times you have seen an arbitration clause with any of the following points covered?

(1)   If the claim is under $10 million, a sole arbitrator; over $10 million, three arbitrators. It is a far more cost-effective approach and avoids delays in smaller cases. You do not know what sort of dispute will materialize.

(2)   No reasons or limited reasons.

(3)   Limited disclosure. Why not disclose just what you rely upon with a power to the arbitrator to order a specific document if it is shown to be crucial to the claim or defense? Common-law discovery does not exist in anything like same way in civil law systems, but can you say less justice there?

(4)   Length of submissions and number of rounds of submissions.

(5)   Use of tribunal-appointed experts or some combination of that with party-appointed experts.

(6)   Clarification with regard to legal professional privilege, in particular in relation to in-house counsel.

(7)   Confidentiality.

(8)   Exclusion of punitive damages.

(9)   Hearing within a specified time.

(10) Award within a reasonable time. The International Chamber of Commerce has a two-month limit for a sole arbitrator and three months for a three-person tribunal. The sanction is reduction of fee!

(11) Agreed expedited procedures unless the tribunal thinks them inappropriate. There has been a lot of discussion recently about expedited hearings. Why shouldn’t they be the norm unless the parties opt in to a more relaxed procedure, in which case any delay is down to them?

(12) Cap on or recoverable costs.

(13) Agreement that the tribunal can assist in relation to a settlement.

I am sure you can think of other topics that could helpfully be addressed in the arbitration agreement which would have the benefit of making the arbitration more effective and less expensive.

Reasonable opportunity to present one’s case. The arbitration agreement is not the only place where the parties may contribute to the overall effectiveness of the arbitration. By being reasonable while presenting their case, counsel and parties may easily reduce the time and costs of the proceedings.

In this regard, I want to mention a very important judgement of the Singapore Court of Appeal given by the chief justice, Sundaresh Menon. The case is CNMC v. Jaguar, which was an application to set aside an award of a tribunal. The chief justice addressed Article 18 of the UNCITRAL Model Law, which, as you know, states that a party must be given a full opportunity to present its case. The issue was whether the word “full” added anything or did it just mean “reasonable”? The chief justice concluded that the drafters really did mean a reasonable opportunity. This judgement is really worth reading as it is also relevant to the issue of due process paranoia. If tribunals have to give a reasonable opportunity to parties to present their case, they should resist attempts by a party to request an unreasonable opportunity. Hopefully, this judgement will be a partial cure for this brand of paranoia. I should add that some jurisdictions saw this coming and adopted the Model Law by deleting the word “full.”


I now turn to the topic of reasons. The 1985 UNCITRAL Model Law in Article 31(2) mandated reasons, and this was followed by the English Arbitration Act 1996. It is important to point out that both instruments permit parties to opt out of the need for reasons to be given. I have, however, never come across this.

There appear to be three main reasons why it was thought that reasons should be mandated. The first was so that both parties knew why they won or lost. The second is that the supervisory court had sufficient material to consider the award and its reasoning in the case of any challenge. The third is that when dealing with standard form contracts, the drafters of those contracts, say, the International Federation of Consulting Engineers (FIDIC), should know that their clauses are being differently interpreted so that they can consider revisions if need be.

I have found that, over the years, awards have become longer and longer. As long ago as 1994, the late Judge Howard M. Holtzmann of the U.S./Iran Claims Tribunal remarked on “the growing tendency to write Awards that are overly long and excessively detailed.” I think even he would be surprised at how far that tendency had grown since he wrote those words.

It is important to bear in mind that an award is not the same as a judgment. Most awards, at least in commercial arbitrations, never see the light of day and have no precedential effect.

Succinct awards. Let me be clear at the outset. I am not advocating the abandonment of reason unless the parties so agree. What I am advocating is shorter, more succinct awards.

For starters, the procedural history should be consigned to an annex. It is rarely relevant unless a due process issue has arisen. If included in the main body of the award, it just clogs it up and makes it harder to follow. Most readers will skip it anyway.

I also accept that, in difficult cases, the correct conclusion only appears during the writing process and the conclusion may be the opposite to that taken away at the end of the hearing. If you start to write an award leading in one direction and you cannot put it down on paper, it often transpires that the opposite conclusion is the correct one, and when you do that on paper, it flows effortlessly. However, I do believe that there are many awards or parts of awards that can be written far more succinctly without in any way affecting the cogency of the reasoning, and I would encourage this.

Early Openings

I am pleased to note that the early opening proposals that I made in 2014 have been widely used to great effect. In a nutshell, the early opening is a hearing that is fixed early on in the case, generally after the first round of written submissions and witness statements. Counsel are given the opportunity to open their case, rather in the old-fashioned way, drawing the tribunal’s attention to the key points in their case, documents, and authorities.

The early opening has a number of advantages. First, it ensures that the whole tribunal reads into the case at a far earlier stage than with the classic timeline. It gets the tribunal together and involved at a far earlier stage.

Second, it enables the tribunal to understand the case going forward and informs its subsequent case preparations. This is particularly true when it comes to expert evidence, on which the tribunal arguably needs more assistance.

Most importantly, the early opening allows the tribunal to have a meaningful dialogue with counsel about the issues in dispute, peripheral points, outstanding disclosure issues, unnecessary evidence, and gaps in the evidence. It facilitates the tribunal putting points to the parties, which they will then have time to consider and respond to. From this dialogue, there is a chance that at least part of the case may be settled or points of disagreement minimized.

Overall, the early opening is a useful tool to ensure a speedier resolution of the dispute and, I would suggest, better awards.

Built-In Mediation Pauses

A topic often discussed is whether the tribunal should ever suggest that the parties attempt a mediation. I am usually reluctant to do so because it is hard to accept that warring parties represented by competent lawyers have not thought about mediation, whether formal or informal.

What I believe would be extremely helpful is if the procedural timetable had built into it a pause in hostilities during which it was understood that there was a window of opportunity to, at least, think about mediation.

It seems to me that the ideal time for that would be immediately after the early opening. Each side would have heard the other side’s case being opened and would have heard any questions or comments made by the tribunal. What better time could there be for the parties to sit down and give mediation one last chance before hostilities begin in earnest again?

Accordingly, I suggest that immediately after the early opening, there be built into the timetable, let us say, a six-week gap where nothing is to happen. The parties will know, having been told at the first meeting, that this period has been set aside for them to think about mediation and try it if they both wished to do so. If either party refuses to engage in mediation, the tribunal should not be told which party refused, only that a mediation has not taken place.

I have recently noted that my colleague Edna Sussman, a former chair of the New York International Arbitration Center, has had exactly the same idea—have the tribunal set a mediation window. I agree with her analysis and hope she will agree that after the early opening is an ideal time.


Finally, I now turn to the contentious issue of costs. We now see arbitrations in which the cost on both sides comes to many millions of dollars. The claim for costs takes on a life of its own. Parties continually complain about the cost of arbitration, but to some extent, the remedy lies within their own hands. We know that the costs of the tribunal and the institution are a small part of the total cost of an arbitration. By far the largest element is the legal and experts’ costs.

I have always been fascinated at how the issue of costs has diverged as between England and the USA. The British rule has been in England for already over 700 years and was transplanted in most of the common-law world. But this rule never seeded in the USA. As long ago as 1796, the Supreme Court rejected the notion.

Whether it was because it was thought that the prospect of paying the other side’s costs was a disincentive to commencing proceedings or was just the American way of not kicking a man while he was down, it has survived because I understand it is still the basic rule in U.S. litigation. But, interestingly, the rule that costs follow the event is found in the AAA International Rules as of course in many other rules.

I am not suggesting that we abolish the rule that costs follow the event. But I do think the time has come to look at it critically and debate whether it needs any change.

How arbitrators can address the issue of costs. However, in the meantime, there is plenty that arbitrators can do. Firstly, I think at the very first meeting with the parties and hopefully with senior management from both sides, the tribunal should make clear that it will be taking into account both sides’ overall behaviur in the arbitration when considering costs. So if the winning side behaved unreasonably throughout the arbitration, this could be taken into account. If a claimant sought 10 million but recovered only 100,000, this too would be taken into account. Similarly, if the claimant made 10 different claims and won on only one of them, this too could be taken into account. If the claimant overlawyered its case, this too could be considered.

I think it important that these issues be canvassed at the outset so neither party can express surprise if this actually happens.

Other ideas that have been canvassed in relation to costs include costs capping by the tribunal or costs budgeting in conjunction with the tribunal, as well as basing recoverable legal costs on an ad valorem scale—rather like the way the International Chamber of Commerce and Singapore International Arbitration Centre and other institutions remunerate arbitrators.

Sealed offers. Finally, I would like to make a plea for more use of sealed offers. If a claimant is seeking 10 million and the respondent thinks it is only worth 6 million, it can write a without prejudice letter to the claimant offering, say, 6.5 million on terms that if accepted within 14 days, that sum will be paid together with reasonable costs up to date of the offer. If accepted, the case ends. If not accepted and the claimant wins 6.6 million, then subject to the discretion of the tribunal, the claimant gets its costs.

If, however, the claimant wins only 6.4 million, it will receive that sum but will have to pay the respondent’s costs from the date of the refusal of the offer. It will get its costs prior to the offer being made.

In English litigation, this is similar to the payment into court provisions and, in the U.S., has some counterpart in Rule 68 offers of judgment. But in arbitration, it is best dealt with by a sealed letter not to be opened by the tribunal until costs fall to be considered. This procedure was considered in and approved by the case of Calderbank v. Calderbank, [1975] 3 All ER 333.


So, in conclusion, there is still much to be done to keep arbitration efficient and reflective of the needs of its often vocal users. I think the use of virtual hearings has given a great boost to more effective case management because it is so easy to set up an urgent hearing, thus preventing disputes from festering.

So, in the light of that, it is possible to endorse the view that every cloud has a silver lining!