The Role of the Expert in Advocacy

This is an Insight article, written by a selected partner as part of GAR's co-published content. Read more on Insight

The distinction between an advocate and expert witness is considered by the UK courts to be crucial to the just disposal of any hearing.[2] Expert evidence presented to the court, it is said, ‘should be, and should be seen to be, the independent product of the expert, uninfluenced as to form or content by the exigencies of litigation’[3] while the expert witness is there to provide ‘independent assistance to the Court by way of objective unbiased opinion in relation to matters within [their] expertise’.[4] An expert witness should not assume the role of an advocate.[5]

Such is the position under English litigation, and experts submitting reports in cases under the English Civil Procedure Rules sign statements confirming this; their understanding of it; and their compliance with it. Many experts will, as a matter of course in arbitration proceedings, include a similar statement to the effect that they are giving objective independent evidence on the matters on which they are instructed, despite these statements not being required by any of the leading institutional rules.[6]

The importance of a competent expert cannot be overstated

The importance of a competent and professional expert cannot be overstated. Experienced experts are expensive, especially in cases requiring complex questions of delay analysis and damage quantification. It is understandable that parties can often be reluctant to make a considerable investment in leading experts; however, it is an investment that is always worth making. I have personal experience of a number of occasions when an otherwise potentially strong case failed because of a flawed expert analysis. Even when a tribunal is otherwise persuaded that a party has a good case, it will not be able to eventually find for that party unless it has detailed and cogent evidence to support the findings when they write up the award. Careful arbitrators tend to be conservative in their decision-making and are looking for a strong and reliable evidentiary basis to reassure them that their decision is the right one.

– Stephen Bond

So, is it right to say experts have no role in the advocacy of a matter in arbitration? After all, reports comprise evidence and argument that support the opinion reached, so experts need to be an advocate for their own opinions and their expert interpretation of documents. To do otherwise, to eschew advocacy at all, would be simply to dump at the tribunal’s collective feet a mass of disorganised material with the instruction to the tribunal to work it out for themselves, which cannot be of much assistance to even the most experienced arbitrators.

For this reason, as in so much in expert evidence, the answer to the question posed must be a qualified ‘yes, but’. In a different sense to advocacy advanced by lawyers on both sides, expert evidence needs to be qualified by a focus on the needs of the tribunal in understanding complex expert evidence and not on advancing the views of the party instructing an expert. Put another way, a ‘good’ expert is an advocate for their own independent view, properly and objectively supported, which may (or may not) align precisely with instructing counsel’s case. A good expert report will equip the tribunal with the language to deal with the expert issues before it and ensure that those issues are properly articulated, and their margins or boundaries defined for the tribunal. A poor expert is one whose evidence simply provides what they think their client wants them to say, leaving the tribunal no way to grapple with the issues it needs to decide.

Experts win cases

Asked, long after the fact, by losing counsel in a case where the weaknesses in his advocacy, if any, lay, I offered two simple words: ‘Your expert.’ That was the truth.

Expert evidence is of the essence in any dispute in arbitration that has the least technical dimension. In more cases than one might imagine, outcomes turn on evidence of a more or less specialised nature. Most leading international arbitrators are generalists and, albeit to a somewhat lesser extent, so too are most leading international arbitration counsel. Expert witnesses plainly fill this gap. Even in disputes having no particularly specialised character, if monetary relief is forthcoming, so too will be expert evidence on damages.

Because arbitral tribunals gauge carefully the objectivity and reliability of expert witnesses, counsel need to admonish experts that poor expert performance can sabotage what might otherwise be a winning case. Who are the experts to avoid? Those who display excessive partisanship, undue defensiveness (including taking umbrage at challenges to their credentials), inconsistencies with prior statements (including prior writings and testimony) and an unwillingness to make strategic concessions.

– George A Bermann, Columbia Law School

Although it is probably better to refer to an expert as a communicator rather than an advocate, expert opinions do form part of the way a case is advanced, and for this reason, the expert has a role throughout the proceedings. This chapter aims to identify the expert’s role at each stage of the arbitral process and to discuss how to get the best out of the expert in a way that contributes effectively to the advancement of the case, bearing in mind the expert’s role as a communicator, not an advocate. Since the previous edition of this Guide, the arbitration world has seen the rise of virtual hearings, and this revised chapter also touches on unique issues arising from giving complex evidence in a virtual environment.

The purpose of expert evidence is twofold: first, to educate the tribunal on technical matters that may not be within their own expertise, but for which understanding is required for the tribunal to do its job; and second, to put forward a view of those technical matters and evidence consistent with the expert’s approach to independence and objectivity and the matters in respect of which they are instructed.

Not all matters require the expert to put forward a view. For instance, accounting teams on both sides may agree that certain costs claimed in proceedings were incurred or expended, with an agreement being reached by each side’s accountants checking and cross-checking a common set of documents. In which case the issue, or the documents, may no longer need to concern the tribunal. Equally, expert evidence may not be required at all if the facts of the case or the contractual matrix are sufficiently clear. However, many matters before the tribunal involve, to a greater or lesser extent, consideration of a counterfactual scenario (and a comparison between that counterfactual and the actual outcome). Hence, consideration at an early stage as to whether experts should assist in devising a counterfactual scenario must be considered, as well as, crucially, the issue of whether opposing experts will agree on the same counterfactual.

In some cases, expert evidence need only be a translation of certain technical aspects of an issue and no expert opinion will really be required. In other words, once the fact evidence has been explained by an expert, the conclusions to be drawn may be obvious. Most practising experts draw an important distinction between resultant evidence (the outcome of factual analysis) and expert opinion, a distinction that may be less obvious to non-specialist counsel and tribunal. For instance, following money through a complex web of offshore and company transactions (to use a common example) involves a degree of forensic skill, but the determination of the final destination of funds is one of resultant evidence. Expert opinion in this scenario may result from the consequence of the loss of those funds, not their determined disposition.

What makes good expert evidence is therefore the expert’s ability to communicate complex issues clearly, precisely and without compromising on detail or meaning. The expert’s understanding and ability to communicate are different skills, and only the latter is useful to the tribunal. In choosing an expert, counsel should consider both expertise and experience: expertise, because someone who is not an expert in their field will generally be a bad choice and will almost invariably come unstuck in cross-examination; experience, because of the unique way in which expertise is challenged in adversarial proceedings. The work of an expert is, in practice, part ‘journey’ and part ‘translation’, as, regardless of what is found, however complex or esoteric, it has to be communicated succinctly and clearly in arbitral proceedings for it to be of use to the tribunal. This ability to communicate will allow an expert to contribute to the overall advancement of a party’s case.

Experts not advocates: style

Flamboyance and expert don’t necessarily go well together. Quiet confidence wins the day with the tribunal.

– London-based partner at a US law firm

If the expert cannot convey the point to the tribunal, it’s irrelevant.

– Arbitrator

Quiet confidence assists the court most – if they are thumping the table advocating a position, they lose trust. I want a tribunal to think: this person knows what they are talking about, and I trust them because they are here to help us.

– Partner at a UK law firm


Instructions are simply the areas for which the expert’s opinion is sought. An expert can assist in framing the precise questions to be addressed. Framing the ‘exam questions’ is a good way of distilling exactly what issues need to be considered. While it may be unavoidable because of the circumstances of the parties’ claims, an issue with instructions arises when they need to address a party’s counterfactual case. Consider the difference between the following.

  • What was the value of Company X on historic date Y? (Or what is the value of Company X today?)
  • What should the value of Company X have been on date Y had Z not happened? (Or what should the value of Company X be today ‘but for’ Z?)

The first question may require no further input. The expert may seek technical clarification, for instance in the context above as to the relevant ‘basis’ of value required,[7] but once the question is clarified, the answer will depend on the quality and availability of documents and the exercise of the expert’s own skill. The expert may need, either because of the lack of available documents or for other reasons, to make assumptions in arriving at an opinion, but those assumptions are the expert’s assumptions, no doubt to be tested in cross-examination.

The second question requires the expert to apply a counterfactual case and the issue may be that there is no agreement between claimant and respondent sides as to precisely what that counterfactual case is; each side may offer different counterfactual scenarios and, ultimately, the ‘correct’ counterfactual case is to be decided by the tribunal based on the evidence put before it.

In these situations, the expert needs to be provided with party-side assumptions and to exercise a great deal of care to remain a neutral communicator and not stray into case advocacy.

For instance, it will generally not be a good idea to restrict an expert to using assumptions only in the client’s favour: an expert who only deals with one-sided assumptions will appear biased, even if following their instructions, and becomes a poor advocate for their own position and the case overall. Equally, an expert asked to value a business based on an assumed set of inputs is simply acting as an expensive calculator, and the resultant ‘valuation’ is merely an exercise in ‘garbage in, garbage out’.

Trust your experts and tribunal!

Recently, I chaired a London-based arbitration where, after first receiving consent from counsel, the tribunal was able to conduct a separate meeting with the quantum experts. The discussion took place prior to the hearing on the merits, but after the filing of two rounds of expert reports. The findings of the experts on rather complex issues differed considerably, and many assumptions and deductions, in particular with regard to the lost profit calculations, could not be reconciled.

Leading up to the meeting with the experts, the tribunal confirmed with counsel the scope of issues to be discussed and distributed a detailed agenda of topics. The tribunal also made clear that any remarks made by its members during the meeting would be without prejudice to its findings on the merits and made solely for the purpose of facilitating the tribunal’s understanding of the expert testimony.

The experts were cooperative and forthcoming during the day’s discussion. There was an open discourse among the experts and members of the tribunal. The experts were given the opportunity to provide their input and feedback on the various subjects set out by the tribunal. The overall atmosphere was more collegial than one normally encounters in a cross-examination setting.

The assembly proved to be very helpful for all parties involved. In the end, the tribunal was able to better understand the points on which the experts were in agreement, as well as the points where there was a clear difference of opinion. An added benefit of the meeting was the identification of documents that both experts were interested in seeing to further update their findings. Although quite onerous to prepare for, the process allowed the tribunal to head into the hearing on the merits with a clearer understanding of the experts’ opinions and consequently the parties’ respective cases. Ultimately, as a result of such an exercise, the presentation of the experts and their cross-examination at the hearing will be more focused and useful.

– Georg von Segesser, von Segesser Law Offices Ltd

Conversely, a good expert can advance a case by considering those client-side assumptions as part of the expert report and forming a view as to reasonableness in the circumstances. This is particularly relevant in matters of damages where an instructing party may have strong views as to financial counterfactual. In these circumstances, the expert serves both the tribunal and instructing party by providing context and independent balance to those client-side assumptions, which in turn enhances his or her credibility with the tribunal and advances the client’s case. Of course, the situation where the expert disagrees explicitly or impliedly (‘I am instructed to assume . . .’, a phrase loaded with meaning when it comes to cross-examination) with his or her instructions or client-side assumptions should be avoided.

Experts not advocates: independence

Lawyers are paid to make an argument they may or may not believe in – by contrast, the tribunal wants to hear an expert’s honestly held, independent opinion, which, as counsel, you hope will advance your client’s case. You wouldn’t present the expert if his or her opinion was not supportive.

– Partner at an arbitration ‘boutique’ firm

Once you instruct an expert, they are a bit like the child you push out the door at 18 to make their own way in life. They are out on their own; you ought not to control them (nor tell them in a meeting what not to concede) – the expert is now on their own. Hopefully, you have sent them in the right direction.

– Partner at a UK law firm

Finally, an expert should (and should be instructed to) deal with a range of relevant permutations. A damages expert will always proceed from the basis that the respondent is liable, or that the claim succeeds; but there may be different combinations of findings on liability that the expert needs to consider. In my experience, this often happens with dates – date of the breach, alternate dates, dates of valuation – and an expert that only considers the client’s proposed valuation date may be exposed where the tribunal prefers the other side’s date. Generally, it is good practice for experts to disclose, or at least summarise, their instructions to avoid confusion on this point.

It is not uncommon to see matters bifurcated between ‘liability’ and ‘quantum’ phases. Although this may be driven by a preference to avoid dealing with complex maths on a Friday afternoon, bifurcating a matter may allow the tribunal to deal with and resolve the different complex counterfactuals so that the expert evidence on quantum deals only with the situation that the tribunal finds. Where bifurcation is being considered, it may be crucial to instruct an expert before any decision, so that the different implications of those scenarios may be considered.

The report

Prior to preparing a report, the expert and his or her team can work with counsel both in framing the issues of the case encapsulated in instructions and with the complex issue of disclosure to ensure that documents relevant to those instructions are available. Experts can be useful in identifying what documents to request (this is particularly true of technical experts), how to go about specifying them with unambiguous particularity, and in assessing the resulting production for gaps: documents that should exist but have not been produced or whose existence is denied. While this may seem reasonable from a legal viewpoint, an expert may have a different view.[8]

The report is the expert’s first contribution to the advancement of the case and a report that does not advance the case may never be seen again. Here it helps to remember that the report should be directed towards the tribunal, and its function is not to advocate client-side assumptions or the preferred counterfactual scenario, but to put forward the expert’s opinions and engage the tribunal so that it sees that they are fair and reasonable, correct in the circumstances and should be preferred to those of the expert on the other side. There are other ‘consumers’ of the report, including the other side and their expert, but they are not the primary focus.

A report is advocacy in written words, numbers and graphics. While it is generally the case that most arbitral tribunals are composed of lawyers, and increasingly those that sit as arbitrators are financially literate,[9] it may always be better to assume a zero-state of technical knowledge in a report. This does not mean a didactic approach (condensing a three-year MBA finance or accounting course is hardly feasible), but it does mean concepts and terminology should be carefully introduced and, crucially, precisely described. The expert should not appear to be hiding behind what the tribunal might see as obscure terminology. To give an example of the confusion that may result from terminology, I was once instructed on a case where a sale and purchase agreement provided for a further payment to be made based on the amount of ‘profit’ made following acquisition but was silent as to the basis or definition of profit. Hence, even simple terms can be complex on expert examination, for instance.

  • Should ‘profit’ be measured before or after tax (PBT or PAT), ‘gross’ or ‘net’ profit, or some other common measure such as operating profit or earnings before interest, taxes, depreciation and amortisation?
  • Should ‘profit’ include or exclude changes of accounting policy or procedures between the date of sale of the business and date of measurement for the purposes of the additional payments?
  • Could ‘profit’ include non-cash amounts earned from a related party (to the seller) transaction where future payment to the company was guaranteed by the assignment of a receivable, being the further payment to be made by the acquiring company?[10]

In the above case, discussion of the meaning of the term ‘profit’ and its various permutations was crucial; but of more general application, the expert does not want to be in a situation where in the context of profit the tribunal is thinking about PBT, when the expert is referring to PAT.

Words are dangerous things: the expert’s choice of words can enhance authority and credibility or land him or her in hot water when it comes to cross-examination. For this reason, qualifiers and superlatives are generally best avoided, and the expert is advised to maintain a dispassionate, balanced and neutral tone. Precision in terminology must be adhered to especially where technical terms and terms that also have an everyday meaning that may, or may not, accord with that technical meaning are used, and where those terms may have multiple meanings. The aforementioned discussion of ‘profit’ is relevant here, but everyday terms such as ‘risk’ and ‘discount’ betray technical meanings distinct from their everyday usage.

If words are dangerous, numbers and graphics can be more so. There can be considerable benefit in providing visual aids in an expert’s report, providing the report takes the time to explain the purpose of the diagram and the conclusions to be drawn from it, neither of which will be obvious to the untrained reader. However, not all complexities can or should be simplified, and there is a danger with oversimplification. Part of the expert’s role must be to make the tribunal aware of the true complexity of an issue and to equip the arbitrators to deal with it properly. Oversimplification, particularly the sort of ‘circles, and arrows and a paragraph on the back of each one explaining what each one was’[11] approach can raise the issue of impartiality, where necessary issues are hidden behind an inauthentic stated goal of making ‘things palatable’ for the tribunal.

The opposing expert

Experts disagree. Understanding why they disagree, and the nature and boundary of that disagreement, provides the tribunal with a reference frame from which to approach expert evidence. Under the English Civil Procedure Rules, experts are generally required to prepare a statement for the court of matters agreed and not agreed and reasons for disagreement (generally known by the shorter name of ‘joint statement’). Although this is not required in an arbitration context, it is not uncommon to see tribunals engage with the concept of the joint statement.

In some cases that I have been involved in, the tribunal has directed experts to meet early in the case and to seek to agree on a common approach or common basis of the approach. While noble in its aim, this may be difficult to achieve in practice. Early on in the case, matters are simply less understood, and it is only by working through these, and the relevant documents, that the experts will start to see the areas where disagreement will arise – which makes putting a joint statement later in the process more appropriate.

Whenever prepared, the joint statement is all about communication. Here the experts are not advocating their respective positions but reconciling the differences between them to establish the key issues that the tribunal will need to decide. For instance, the experts may disagree whether a discount rate is 6 per cent or 9 per cent but agree that the answer lies within that range and the particular factors that, when properly considered, lead to an answer between those values. This avoids the tribunal having to deal with superfluous, immaterial or irrelevant material and focuses attention on the actual issues to decide, often lost in the detail of an individual expert’s report.

Experts not advocates: hearings

The lawyers always want to get the expert into an argument – but get them away from their area of expertise. No lawyer will want to take on accountant on figures – if they do, they lose.

– Arbitrator

Beware of the ‘overstretch’ – you get experts who are inclined to opine on anything; perhaps they like the sound of their own voice. It comes out quickly in cross-examination that they don’t know a thing about the issue, which casts doubt on their credibility as an expert – and contaminates other parts of their work.

– Partner at an arbitration ‘boutique’ firm

The skills needed in a hearing are very, very difficult. Appointing an expert for the first time is a bit like skydiving; you’ll never know how good they are going to be until they have done it once, and the first time can be a disaster.

– London-based partner at a US law firm

The hearing

It is becoming increasingly common for experts to be asked to make opening statements to the tribunal. This is not an opportunity for advocacy; appearing to argue the case in an opening statement may set a tone of partisanship or bias that will not sit well with the tribunal members. Therefore, it is important that the opening statement is not ‘new’ evidence but a helpful distillation of those points that are of most relevance to the issues at the forefront of the tribunal’s collective mind.

It can be instructive for the expert to sit in on, or at least read the transcript of, the opening day of a hearing where the parties’ respective cases are introduced. The questions asked by the tribunal will indicate the issues vexing its members and, to the extent those touch on issues falling with the expert evidence, focusing on those questions – and providing answers from paragraphs in the expert’s report – may be a good use of an opening statement.

Cross-examination follows; its goal may not be to argue with the expert on his or her area of expertise, but it will aim to put an expert under pressure, create fluster and reduce credibility. An expert does not have to be defeated on all points in a report to lose credibility, so cross-examination will be targeted.

From an expert’s perspective, the goal in cross-examination is to show that our approach, review of the evidence and expert opinion remains balanced, fair, impartial and correct. Responses to cross-examination should not be advocacy – we are not there to argue the client’s case – but need to clearly distinguish one scenario (usually the one in our expert reports) from another, typically the one being put in cross-examination. An expert should be measured, serious, convey gravitas, appear calm and not give in to anger or frustration to demonstrate that he or she is there to assist the tribunal. Witnesses are often asked to give simple ‘yes’ or ‘no’ answers, but the ‘yes, but . . .’ answer is a common refrain in expert cross-examination; neither defensive, nor one to be avoided, as answering in this way better equips the tribunal to address the issues. Extending the earlier example, if counsel asks, ‘given X, would not the discount rate be Y?’, an answer that, ‘yes, but X implies Z, which is not observed in this case, so Y will not be correct in this case’ provides better information.

Experts must always assume that cross-examining counsel will be well prepared, immersed in the detail of the case and know the complexity of the expert issue; he or she will be well assisted and prepared by the opposing expert, and the supporting team. In the hearing, he or she will have access to that resource via the tried and tested ‘yellow sticky’ note.

Concurrent evidence is a relatively recent phenomenon, and there are many differing views as to its efficacy, which cannot be dealt with in this chapter. There is agreement that concurrent evidence is not simply experts arguing with each other, as it is difficult to see how a tribunal is assisted by a philosophical debate that it may not be equipped to follow, nor is concurrent evidence about the experts’ skills in advocacy and oratory. Done well, ‘hot tubbing’ allows the tribunal to probe at the heart of the issues of disagreement between experts and, as such, serves much the same role as the joint statement of matters agreed and not agreed.

The virtual hearing

It remains to be seen whether the virtual hearing, a product of exigence at its inception, will be retained as a matter of convenience or cost. From an expert’s view, I estimate (in a strictly non-expert estimate) that a virtual hearing removes something like 95 per cent of the information content usually available to an expert witness in a hearing: distinguished members of the tribunal are reduced to postage stamp-sized moving images while the screen is filled by an unnaturally sized cross-examining counsel.

This latter observation makes it important to remember that the expert is there to assist the tribunal, whereas screen-filling counsel may naturally provoke dialogue, something perhaps best avoided in cross-examination. It becomes harder to direct answers to the tribunal and perhaps harder to gauge whether a point is being understood (or misunderstood), while virtual separation distances the expert from the process, again at odds with the expert’s role to assist the tribunal.

Much also depends on the technology being deployed; how documents are shown to the expert and discussed; and whether the expert has access to documents without screen sharing them with the entire virtual room, so as to locate documents that contextualise those that are shown to the expert. In all virtual settings, what may matter most is whether the expert is looking at the camera, so that the virtual appearance of the expert makes eye contact with those viewing his or her evidence.


In previous editions, this chapter has stressed the tension inherent in the role of an expert witness, where a need to be objective, independent and duty-bound to the tribunal potentially conflicts with the expert’s desire to help the client’s case and the professional obligation to that client. That point must be made again: an expert’s role remains, first and foremost, to reach professional opinions on the matters for which they are qualified, and a good expert will manage that tension – adding value to the advancement of the case by explaining their opinions in a way that is clearly understandable to the tribunal and is consistent with the client’s case. It is that ability to communicate that instructing counsel often seek. After all, a good point badly made or not understood, is, at best, of no use and, at worst, harms the expert’s credibility.

Experts are not hired guns. We do not advocate for our client’s case, and our evidence is undermined if it is shown to be incomplete or one-sided, or avoids tackling interpretations that are unfavourable to our client. Counsel will have selected an expert, not just for their expertise and experience, but also because (it is hoped) that expert’s opinions will support, explain, bolster or at least assist the client’s case. The expert’s role then becomes, principally, one of communicator, and the expert’s contribution to advocacy is to advocate genuinely held and professional opinions in support of the client’s case, but distinct from counsel’s advocacy for the client.


[1] Luke Steadman is a partner at Alvarez & Marsal.

[2] Pickles v. Revenue & Customs (whether crediting a directors’ loan account that was freely available for the directors/members to draw upon) [2020] UKFTT 195 (TC) (22 April 2020), at 6.

[3] Whitehouse v. Jordan [1981] 1 WLR 246, at 256.

[4] Pollivitte Ltd v. Commercial Union Assurance Company PLC [1987] 1 Lloyd’s Rep, 379 at 386.

[5] Civil Procedure Rules, Practice Direction 35, Paragraph 2.2.

[6] The IBA Rules on the Taking of Expert Evidence in International Arbitration require experts to be independent of the parties but say nothing about the content of their evidence. The Chartered Institute of Arbitrators Protocol on the use of Party-Appointed Experts states that party-appointed experts should be ‘impartial, objective, unbiased and uninfluenced’ and imposes a duty on experts to ‘assist the arbitral tribunal to decide the issues in respect of which expert evidence is adduced’. The International Chamber of Commerce, London Court of International Arbitration, United Nations Commission on International Trade Law and International Centre for Settlement of Investment Disputes rules (among others) permit expert evidence but impose no requirements on party-appointed experts.

[7] For instance, ‘market value’, ‘fair value’ or some other basis.

[8] In a High Court matter I was involved in, it came across badly for the other side when they were forced to explain to the judge that the reason that they had not produced ‘accounts’ and ‘budgets’ for the ‘operating companies’ that I had requested as single joint expert in that case, was because ‘financial results’ and ‘forecasts’ were prepared on a ‘group’ basis so that no such documents as requested, technically, existed.

[9] I recall once being told by an arbitrator that he was most looking forward to hearing about my derivation of country risk premium in my capital asset pricing model calculation.

[10] The effect of which was to increase ‘profit’ in circumstances where the earn-out payment was based on a multiple of profit. For every 100 added to profit in this way, 500 was added to the earn-out payment, of which 100 was assigned as guarantee for the amount due to the company.

[11] Arlo Guthrie, ‘Alice’s Restaurant Massacree’ (Warner Bros, 1967).

Unlock unlimited access to all Global Arbitration Review content