The expert’s role – indeed, his or her overriding responsibility – is to assist the tribunal on matters within his or her expertise. It is critical that experts, parties to an arbitration, counsel and tribunal members should recognise this obligation, given that only approximately 10 per cent of experts are appointed directly by the tribunal (with the remaining 90 per cent being appointed by parties to the arbitration).
To fulfil this obligation, an expert’s evidence must be useful to the tribunal. The expert is appointed to offer his or her professional opinion on matters within his or her expertise, in contrast to a factual witness, who presents his or her recollection of the facts. We suggest that the key indicators of the usefulness of an expert’s evidence are independence and impartiality. It is also vital that an expert acts with integrity and that there is clarity in both written and oral evidence. A lack of these characteristics impairs the reliability of the expert’s opinion and will erode the tribunal’s trust in that expert.
International arbitration is a deliberately flexible process in which the parties often have significant influence over the implementation of the rules and procedures to be followed. However, regardless of the rules and procedures governing a particular case, the role of the expert has broad, and important, underlying similarities in any international arbitration.
In this chapter, we discuss the role of the expert (with a focus on damages experts) in an arbitration. In particular, we discuss (1) guidelines for the conduct of the expert; (2) the importance of effective instructions; and (3) the execution of the expert role in an arbitration.
Guidelines for the conduct of the expert
Flexibility and control throughout the process are two of the characteristics that attract parties to international arbitration. Given the multiple jurisdictions and arbitral institutions around the world, there is no single definitive set of rules that govern the conduct of the expert. However, an examination of several sets of rules reveals a number of important common themes.
Duty to the tribunal
It is almost universally accepted, and often dictated by the relevant rules, that the primary and overriding duty of an expert is to assist the tribunal on matters within his or her expertise. This is the case whether an expert has been appointed by a party to an arbitration or by the tribunal itself.
For example, the Chartered Institute of Arbitrators states that ‘[a]n expert’s duty, in giving evidence in the Arbitration, is to assist the Arbitral Tribunal to decide the issues in respect of which expert evidence is adduced.’
The way in which this obligation can be satisfied is through the provision of useful evidence to the tribunal. In particular, the evidence provided should demonstrate:
- independence and impartiality, meaning that the opinions reached are not influenced by any parties to the dispute, but are based upon an objective consideration of the evidence available to the expert;
- integrity, meaning that the expert informs the tribunal of any limitations in the conclusions he or she has reached, or the effect that relevant alternative assumptions would have on those conclusions; and
- clarity, meaning that the expert’s instructions, assumptions, reasoning and basis for his or her conclusions should be set out clearly and in a manner that the tribunal will be able to follow.
Although independence can be considered part of the expert’s overriding duty to the tribunal, its importance is explicitly referenced in many arbitration rules, including those that do not emphasise the expert’s overriding duty to the tribunal. For example:
- the International Bar Association rules on the taking of evidence state that the expert’s report should state ‘his or her independence from the Parties, their legal advisors and the Arbitral Tribunal’ and include ‘an affirmation of his or her genuine belief in the opinions expressed in the Expert Report’;
- the Chartered Institute of Arbitrators states that the expert’s report must include a declaration that it reflects his or her ‘own, impartial, objective, unbiased opinion which has not been influenced by the pressures of the dispute resolution process or by any party to the arbitration’; and
- the London Court of International Arbitration states that any ‘expert shall be and remain impartial and independent of the parties; and shall sign a written declaration to such effect’.
The appointment of an expert by a party can create a tension between the expert’s obligation to the tribunal and the interests of his or her instructing party. It is the expert’s responsibility to ensure that his or her evidence prioritises assistance to the tribunal over and above serving his or her instructing party’s interests.
In practical terms, this means that the expert should not simply argue the most beneficial possible case from the perspective of his or her instructing party (indeed, he or she should not ‘argue’ the case at all). Rather, the expert should consider the available evidence objectively and arrive at an independent view of the issues at hand.
The rationale behind this is clear. By definition, experts express their opinions on technical areas that may be outside the expertise of the tribunal and, potentially, the parties to the arbitration. If experts did not conduct themselves in an independent manner, the tribunal would be faced with a nearly impossible task of deciding between competing and potentially disingenuous evidence on technical issues outside of its own expertise. Unfortunately, this situation has been known to arise.
That is not to say that experts should or will always agree. As with any decision-making process, particularly one involving the application of professional judgement in areas that may have an element of subjectivity, there is the potential for reasonable parties to disagree. Moreover, it is not only differences in professional judgement that can lead to different conclusions being reached by experts. For example:
- there are often factual disputes in a case. An expert may be instructed to make a particular factual assumption consistent with his or her appointing party’s case. The outcome of a factual dispute will not be within the expertise of the expert and, ultimately, will be determined by the tribunal;
- particularly in the case of a quantum expert, the expert may be reliant upon other (contested) expert evidence. For example, a valuation expert may be dependent upon expert evidence as to the potential growth in a particular market or the technical limitations of a newly developed technology. The valuation expert would typically be instructed to rely upon the evidence of that other expert appointed by his or her appointing party but, ultimately, it would be for the tribunal to determine which of the competing ‘upstream’ expert opinions should be incorporated into the valuation;
- the time available to the expert to complete his or her analysis and form an opinion. One expert may have been appointed earlier in the process than the other, and may, therefore, have had more time to digest the facts of the matter (including disclosure), perform research, prepare analysis and consider his or her views; and
- there may be information asymmetry between the experts, particularly at the start of an arbitration, which could contribute to differences between the conclusions reached. A well-managed disclosure process can help to mitigate this risk, as can meetings between experts. Even so, it is not uncommon for one expert to have had greater access to relevant information.
We discuss, later in this chapter, ways in which the expert should recognise the above factors in his or her expert report, as well as the potential assistance that a joint expert report may offer the tribunal in understanding the implications and effects of different instructions or assumptions.
Experts may feel that the confidential nature of arbitral proceedings offers them some protection from public criticisms regarding their independence. However, maintaining independence of experts (and arbitrators) within international arbitration is critical to the success of the whole arbitration industry. The seriousness with which independence should be taken is illustrated by the United Arab Emirates’ introduction of a law, effective from October 2016, which states:
Any person who issues a decision, gives an opinion, submits a report, addresses a case or proves an incident for the benefit or against a person, failing to maintain the requirements of integrity and impartiality, in his capacity as an arbitrator, expert, translator or investigator . . . shall be sentenced to temporary imprisonment.
While the aims of this legislation are laudable, its implementation carries significant risks. It is not unknown for counsel to levy fairly indiscriminate allegations of bias against experts retained by their opponents; the risk of imprisonment, particularly without proper cause, may have a chilling effect on the willingness of experts and others to be appointed in the region. It remains to be seen whether this amendment to the UAE Federal Penal Code will be used in practice, or whether it will be in place in the long term, but the risk of custodial sentences (temporary imprisonment is defined as three to 15 years) sends a very strong message regarding the importance of independence in arbitration.
The importance of effective instructions on an expert’s opinion
When an expert is appointed by a party to the dispute, as happens in the majority of cases, he or she must take instructions from that party (usually through the appointing party’s legal team). In these circumstances, the questions that the expert is answering are often set by one of the parties, rather than by the tribunal (or jointly by the parties).
In cases involving party-appointed experts, particularly in cases in which the experts are exchanging reports simultaneously, there is therefore a risk that the instructions given to the experts may not completely coincide. If this occurs, the experts may be answering different questions, reducing the usefulness of the experts’ opinions to the tribunal (at least until the stage of reply reports, and sometimes not until three reports have been exchanged). Although it is the responsibility of the parties to instruct their experts, the tribunal may decide that its own early involvement in the expert process will help to ensure that experts are addressing the same issues. This can help avoid unnecessary work being carried out (and costs being incurred) by the parties to the arbitration. In the absence of tribunal involvement, the parties should consider whether cooperating on a list of expert issues will be beneficial to the process.
If a tribunal is appointing an expert directly, it should ensure that the expert is instructed to answer all the necessary questions to enable both parties’ arguments to be fully considered. If one party feels that an argument has not been fully considered, it may lead to appeals against any tribunal decision.
Scope of the expert’s report
The expert should be careful not to address topics that are outside of his or her expertise or the issues on which he or she has been instructed to opine. Providing opinion on such matters is not likely to assist the tribunal in its determination of the relevant issues.
As part of fulfilling the ‘clarity’ element of an expert’s duty to the tribunal, the expert should clearly list his or her instructions in any expert report. To the extent that these instructions evolve between reports (for example, an expert may be instructed to respond to another expert’s report or perform additional analysis), this should be highlighted in the subsequent report.
Treatment of disputed matters of fact or expertise
Arbitrations usually do not concern only the quantification of damages, or the determination of whether ‘Expert A’ or ‘Expert B’ is correct. For many experts, and in particular for experts in the quantification of damages, disputed facts or other expert opinions may influence his or her opinions and conclusions.
Consider, for example, an expropriation claim under a bilateral investment treaty. There may well be a dispute as to the date on which the asset in question was expropriated (or if it was expropriated at all). An expert valuer, being asked to assess the fair market value as at the date of expropriation, would not typically have the expertise to offer an opinion as to the date on which expropriation occurred. He or she must therefore be instructed as to the appropriate date at which value should be assessed.
Typically, the valuer in this hypothetical case would be instructed to assume that his or her instructing party’s position is correct. In the interests of clarity (and transparency) the expert should set out the basis of his or her instruction or assumption in the expert report.
In addition to disclosing the instruction, it may also be appropriate for an expert to consider the effect of this instruction, or assumption, on his or her conclusions. In order to demonstrate integrity, and if it would be of assistance to the tribunal, the expert may perform sensitivity analyses that show the answer to the question on alternative bases, potentially including the assumption that the other party’s position is correct.
Of course, the quantification of damages under numerous scenarios is not always practicable. If multiple instructions are required, the number of permutations grows at an exponential rate. In other cases, altering even a single assumption (for example, the valuation date) may result in a disproportionate amount of work. In such cases, an expert may discuss the effect of conflicting assumptions made by the parties qualitatively.
The execution of the expert role in an arbitration
Because of the bespoke nature of arbitration, the input of an expert will be different in each case. In the remainder of this chapter, we discuss the potential involvement of a quantum expert in various aspects of a case.
Although a quantum expert may be retained to advise on the potential quantum of any claim before that claim is issued, the first formal output of a quantum expert will typically be an expert report. This may be followed by further reports as more information comes to light, his or her instructions change, or another expert produces reports affecting his or her opinions or conclusions.
The expert’s report should make clear the expert’s overriding duty to the tribunal, both explicitly and through the demonstration of characteristics such as independence and impartiality. The format and precise requirements of an expert’s reports will be determined by the rules of the arbitration. However, many arbitral rules require an expert to make a statement of independence and even prescribe the content of such a statement.
The quantification of damages almost always requires the creation of a financial model. The complexity of such a model is dependent upon the case in question, and is driven by many factors that we do not consider in this chapter.
There is no single ‘best practice’ for the creation of a financial model, but there are general principles that should be followed:
- the model should be as simple as possible, within the constraints of the calculations that need to be performed. This means that there should be minimal extraneous data or calculations and that the model should be logically structured;
- if appropriate and practicable, the model should be dynamic. This means that it should be possible to run alternative scenarios by changing certain inputs to the model; and
- if necessary, a written explanation of the model should be included. This is required not only in the case of complicated financial models, but in any situation where it will help an opposing expert understand (and test) the model more efficiently.
The application of these principles should ensure that the financial model is of maximum utility to the tribunal. While ‘maximum utility’ may not mean that the tribunal is able to manipulate the model or perform calculations itself, it should enable another expert to assess its accuracy and appropriateness. This may enable opposing party-appointed experts to agree on the mechanics of a model under a number of scenarios, giving the tribunal comfort that the mathematical accuracy of the calculations of damages are not disputed (once the inputs have been determined).
Another way in which an expert can assist the tribunal is to perform sensitivity analyses. We discussed this above, in the context of performing sensitivities around any instructed assumptions, but this also applies to the expert’s own conclusions. He or she may show the tribunal the effect of variations in his or her conclusions through the use of a dynamic model. For example, if an expert determines that an appropriate discount rateis 12.5 per cent, he or she may wish to show the effect on the quantification of damages of applying a 10 per cent or 15 per cent discount rate.
Sensitivity analyses such as these can be useful in allowing the tribunal (and the parties) to establish which of the expert’s assumptions or areas of disagreement between the parties have a material effect on the quantification of damages.
Pre-hearing joint expert meetings and statements
In some cases, it may be of assistance to the tribunal, and the parties, if opposing party-appointed experts are able to identify areas of agreement and disagreement. A meeting between experts or the preparation of a joint statement may result in:
- the narrowing of issues in dispute that need to be considered by the tribunal;
- a useful reference document for the tribunal during its deliberations;
- identification of conflicting assumptions or instructions adopted by the experts;
- identification of any information asymmetry that may exist between the experts;
- an agreed financial model; or
- agreement as to the effect on quantum of certain assumptions.
A meeting of experts before the hearing typically takes place on a without-prejudice basis, without counsel for the parties being present. In some cases, the entire production of a joint statement is carried out between the experts, without involvement from the parties or their legal teams. Such an approach can reduce any pressure being placed on the experts’ independence by their instructing parties.
The scope of any joint expert meeting or statement is usually defined in advance of the meeting; the experts do not have the ability to reach agreements on behalf of the parties, or to agree to offset differences of opinion between themselves. For example, if there were two offsetting areas of disagreement between the experts, they do not have the authority to agree that one difference should be resolved in favour of each party, resulting in no difference between their ultimate quantification of damages.
The usefulness of such a process to the tribunal, and therefore the extent to which such a process is consistent with the overriding duty of the experts, depends upon the time available and the level of cooperation between the parties and experts.
It is broadly accepted that a statement of matters agreed is of assistance to the tribunal (and the parties). If such a statement can be prepared, it narrows the issues to be considered by the tribunal and can result in shorter hearings and quicker awards. It may also allow the experts to agree a financial model and the effect of instructions or assumptions on the quantification of damages (to the extent that agreement hasn’t been reached in the preparation of sequential expert reports).
A statement of areas of disagreement may also be helpful, but this is not always the case. In principle, a summary of areas of disagreement between experts, setting out the respective positions of each expert, is a useful reference document when the tribunal is deliberating. However, in practice, the creation of such a document can be inefficient, with multiple iterations of the draft document passing between experts and a proliferation of restated (or even new) arguments. In order to be of maximum use to the tribunal, the experts may need to be instructed as to the scope of such a statement; for example, it may be sufficient to identify the areas of disagreement and each expert’s position, with cross-references to their reports, but without each expert setting out the reasons for their opinions at length (which should already be included in the experts’ reports).
Expert participation at the hearing
It is generally up to the parties to agree which experts, if any, need to participate in the hearing. In the case of quantum experts, an appearance is usually considered to be helpful to the tribunal, unless a significant level of agreement has already been reached.
It is increasingly common for experts to give a summary presentation to the tribunal as part of their direct examination. The expert should use this opportunity to summarise his or her conclusions and discuss any key differences in opinion with the other expert. This can be particularly helpful to the tribunal if there have been joint expert meetings and statements since the date of the last reports.
In addition to a direct presentation and cross-examination, an expert may be instructed to perform concurrently with his or her equivalent expert appointed by the other party. Expert conferencing (or ‘hot-tubbing’) can assist the tribunal as it gives the opportunity for the experts to answer the same questions, posed directly by the tribunal. This may enable a tribunal to uncover areas of agreement and reasons for areas of disagreement, to the extent these have not already been made apparent in other work performed by the experts. In the authors’ experience, this can be a useful addition to oral evidence tested by cross-examination, but is not a proper substitute for it.
During his or her oral evidence, particularly during cross-examination and when answering questions from the tribunal, the key characteristics of a good expert (independence, integrity and clarity) should again be apparent. In particular, the expert should not modify answers in order to enhance his or her appointing party’s position, conceal the effect of alternative assumptions (or errors) or deliberately complicate matters being considered by the tribunal. An expert whose evidence is helpful to the tribunal is much more likely to have that evidence relied upon by the tribunal in its award.
It is common nowadays for counsel to highlight which experts have been helpful or unhelpful to the tribunal in their closing submissions. Sometimes this is no more than a variant of ‘my expert is prettier than yours’, but sometimes differences in the experience or impartiality of the experts merit serious scrutiny. It is to be hoped that tribunals show a greater willingness to consider and comment on whether the experts before them have properly fulfilled their role of assisting the tribunal.
 Richard Boulton QC is a barrister at One Essex Court and chairman of Berkeley Research Group (UK) Limited. Joe Skilton is a director at Berkeley Research Group.
 P. 3, 2012 International Arbitration Survey, Queen Mary University of London.
 Although not directly relevant to the topics covered in this book, the common themes identified in this chapter are also present in the English Civil Procedure Rules (Part 35) dealing with experts and assessors.
 Clause 4, Article 4, CIArb Protocol for the Use of Party-Appointed Expert Witnesses in International Arbitration (2007).
 Clauses 2(c) and 2(g), Article 5, IBA Rules on the Taking of Evidence in International Arbitration (29 May 2010).
 Clause 1(b), Article 8, CIArb Protocol for the Use of Party-Appointed Expert Witnesses in International Arbitration (2007).
 Article 21.2, LCIA Arbitration Rules (2014).
 Herbert Smith Freehills translation of Article 257 of the UAE Penal Code, ‘Arbitration in Dubai: Two Steps Forward, One Step Back’, Stuart Paterson (9 December 2016).
 For example, Article 8 of CIArb Protocol for the Use of Party-Appointed Expert Witnesses in International Arbitration (2007).
 We discuss joint expert meetings and statements in more detail below.
 Discount rates are discussed in Chapter 17 on determining the weighted average cost of capital.
 Such an agreement is not always possible, for example, if each expert has produced his or own financial model, it is unlikely that a joint model will be created (absent an instruction to do so).