Bruno Augustin and Rob van der Laan discuss some key questions relating to the quantification of damages in private enforcement actions involving competition law
Does competition law require a different approach in damages quantification compared to other areas of the law?
Bruno: Courts and tribunals have been dealing with the quantification of damages in private enforcement actions in all areas of the law for many years, and there is much jurisprudence and professional experience that competition law practitioners can call upon from other legal disciplines for this task.
Basically, I see no reason why competition law should be treated any differently – the need to estimate the “but-for” or “counterfactual” scenario against which to compare what really happened is an obvious task in any damages action no matter what area of the law the claim arises from. It is something I encounter every time I attempt the task, which I have done so on numerous occasions mostly without the need for complex models.
Yes, competition law has some unique features, especially when the whole market for a particular product needs to be evaluated, but other areas of law (such as Intellectual Property, for example) also have their unique features, which courts have found ways to handle without over-complicating matters. Whatever area of the law you are dealing with, the fundamental principles underlying how you quantify loss remain the same.
Rob: My expertise is very much restricted to economic issues arising from competition law and I do not feel qualified to make statements in relation to damages quantification in relation to other areas of law. I would not be surprised if most of the claims based on other areas of law deal with similar situations compared to damages actions based on an infringement of competition law.
However, my gut feeling is that there are some issues that are very much specific to competition law. It is generally required to have a good understanding of markets in order to assess damages from an infringement of competition law. Both claims based on an abuse of a dominant position and claims based on a cartel infringement suggest that only a limited part of the market may be unaffected and suitable as a base for the counterfactual.
The counterfactual in competition may therefore require a more fundamental understanding of the working of markets compared to what is required for the quantification of damages in other legal areas. For example, how a market would operate without the infringement can depend on detailed analysis of, for example, the impact of capacity constraints and customer loyalty.
What quantification techniques do you prefer to use?
Bruno: There are a number of techniques one could use, and much literature has been published about them that can be referred to. However, it is most important to remember the underlying principle of what you want to achieve, and that is to determine the loss suffered by the claimant by means of comparing what would have happened in the absence of a breach to what happened in reality.
Therefore, the method adopted must be the one most suitable to the facts and commercial realities of the case. For example, if the claimant is a new entrant to the market, it would be difficult to apply a “before and after” method to quantifying its loss, as by definition there was no “before” performance with which to compare its performance during the period of the infringement. The facts must never be a slave to a favoured method, but instead be the driving force behind the choice of method and how it should be applied.
Facts may also reveal commercial realities of the case that are not consistent with conventional wisdom. For example, the claimant may have a long term agreement to sell on a cartelised product at a fixed price, so in those circumstances could not be said to have passed on the overcharge it suffered from the defendant, even though economic opinion may suggest this to be likely in general terms.
Rob: I agree with the notion that the facts should drive the method. However, any kind of model or description will by definition only reflect a part of reality, ignoring the rest for convenience’ sake. For example, a price elasticity of demand may very well capture the effects on sales of a structural price increase over a short period of time; however, changed perceptions by the customers – the image as a cheap or a dear supplier – may not be properly incorporated in the information conveyed by this single number.
Once the fundamental elements of a market are identified, I would advocate the use of various techniques rather than a single method in order to derive several estimates that are compatible with the fundamentals of the market. In case the outcomes are very similar, this may provide additional confidence in the results. In case the outcomes are very different, analysis of the differences may provide insight into a better estimation. In general, taking various approaches to test the robustness of the model or method will improve the result. Of course, there are limits to this approach on the basis of the available time, data and budget.
How important is it to have full disclosure of relevant data?
Bruno: Ideally, it would be nice to have access to as much data as possible when quantifying damages, but the key word here is “relevant”.
For example, in cartel cases, one can make the case for any loosely significant factor having an influence on the counterfactual price, and be tempted to include it in a regression model, just because data exists that allows one to do this. It may be better in such circumstances to go for a simpler technique that focusses on the more directly relevant factors, even if the available data is limited in this area. At the end of the day, it is the quality, rather than the quantity, of the data that counts.
Expert accountants and economists are used to having to make assumptions when key data is missing, and if the other side is at first reluctant to provide full disclosure, the need to challenge these assumptions often persuades it to reveal more evidence as the case progresses.
Rob: I agree that data needs to be relevant to be included; however, we should not forget that often there is not a strict boundary between relevant and irrelevant, between significant and insignificant. This applies in particular to statistical data, where one generally opts for a pre-defined level of confidence before drawing conclusions. Therefore, one needs to be cautious before labelling information as irrelevant from the outset of an assessment.
Furthermore, quantity has a quality of itself. There is no point spending resources on analysing huge amounts of data just for the sake of them being available; however, in my experience, it is the analysis of a large dataset that often provides some unexpected qualitative insights. There is a fine balance to be struck between controlling costs and allowing sufficient freedom of action to obtain additional insights.
How precise does the calculation need to be?
Bruno: As precise as the evidence allows, is the short answer. We must not forget that in calculating the “counterfactual” scenario, we are dealing with a situation that never happened. Therefore, it is impossible to determine scientifically what the alternative scenario would definitely have looked like if there was no breach. There is no “truth” that can be discovered – it is by necessity always going to be an estimate.
Accepting this reality, it is not cost-effective to spend considerable time and money searching for an elusive “truth” and in the process produce a calculation so complex that most people cannot understand. The balance has to be right between aiming for as much precision as the evidence allows, and producing a calculation that is readily understandable, supported by the facts, and makes sense intuitively.
In all my years of dealing with damages quantification in different areas of the law, very rarely have I found judges or arbitrators to be won over by complex models – they are more likely to accept the calculations that are aimed at a level that they can follow with minimal difficulty, and which make more sense and are better supported by the evidence than the calculation produced by the other side.
Rob: There is no simple answer to this, different cases require different solutions. For example, assume that the key question is whether different market conditions would have resulted in market entry in the counterfactual. The entry/no entry question may only need superficial attention if either entry is only plausible in very narrowly defined circumstances (long term contract) or entry would be plausible even if market conditions would change only marginally (perfectly contestable market). A precise estimation would not be of much use.
On the other hand, there are situations where the most precise level of estimation is desirable in order to narrow down the range of plausible levels of entry as far as possible. Precision is generally required where the claimant and the defendant agree on the relevant elements that need to be taken into account. In many cases, however, the importance of the various variables proposed is disputed and the question is more about plausibility and fairness of incorporating the respective variables than about the accuracy of the prediction produced by one of these specific variables.
Can expert evidence also help establish liability and causation?
Bruno: This is an area which has been overlooked in much of the discussion around the use of expert evidence in private enforcement of competition law. Especially with the greater likelihood of stand-alone actions, it is vital that expert evidence is proffered to establish various factors such as market share, or the existence of predatory and excessive pricing.
Causation is also a key factor in any damages claim. Just because a regulator may have found the defendant to have breached the law, does not mean that that the breach necessarily caused the loss allegedly suffered by the claimant. Particularly in abuse of dominance cases, where the claimant has gone out of business allegedly due to the defendant’s behaviour, expert opinion is often helpful to the court in exploring other possible causes of the business’s demise, such as poor management decisions or the inability to handle other commercial pressures.
Rob: Economic and accounting evidence obviously has a place in various stages preceding the quantification of damages – including evidencing or disproving the infringement, or establishing liability.
At times, clients assume a causal relationship between the infringement and damages that sounds unconvincing for an expert. On the other hand, a perfectly plausible explanation within the parameters of economic theory may not sound plausible to, for example, CEOs.
In practice, it is very difficult to establish a causal relationship in any complex economic environment with 100% certainty. An economist can provide a credible economic theory that may allow the interpretation of, for example, a strong correlation. If an explanation of the correlation is accepted by the courts, causality is established. It is important to be receptive to potential explanations from experts with different backgrounds in order to establish a proper chain of causality.
Is there a role for expert accountants or economists during a negotiated settlement?
Bruno: There is always a role for experts to play in assisting a settlement negotiation. Through examining the available evidence supporting or disproving the quantification of a particular claim, an expert can come up with a range of figures which in his/her view represent the potential value of the claim that would be acceptable to a court.
A party can then choose various points within that range as its minimum and maximum positions when negotiating with the other side, safe in the knowledge that they are within the boundaries of what is supportable by the evidence. All the more reason why experts should be engaged at the earliest point in a case when it is practical to do so.
Rob: In my opinion, involving an expert accountant or economist early during the process may help in reaching a settlement. In the first place, the expert may be able to progressively narrow the range of potential damages thereby facilitating a compromise.
In the second place, the expert may identify the remaining areas of uncertainty and help manage expectations regarding finding a single value that is established beyond any doubt.
In what circumstances should an expert economist be appointed?
Bruno: There are many circumstances when an expert economist’s evidence will be useful. If the quantum of the claim hinges on the relative market strengths of the claimant and defendant, or an identification of the relevant market, an economist’s input is essential.
Even though I have cautioned against the use of complex models, in some circumstances, particularly in cartel cases where the whole of a national or regional market is involved, a sophisticated regression model may be a useful tool to estimate a likely overcharge. This is as long as sufficient relevant data is available, the model is kept as easy to understand as possible, and the assumptions are clearly explained.
In what circumstances should an expert accountant be appointed?
Rob: Different disciplines look at the world in different ways. When data are used in an economic model it is essential that, for example, the origin and definitions of the data are identified (or confirmed) objectively. As an economist, one has to work on the basis of information provided by companies almost without exception. It goes without saying that an expert accountant’s involvement is almost a conditio sine qua non for damage quantification and often proves invaluable for the economist.
Are there benefits for experts from different disciplines working together on damages quantification?
Bruno: This has been my experience of damages quantification in many areas of the law. I have often worked together with other professionals on the same case, such as quantity surveyors, valuers, economists and other technical experts, pooling our joint expertise to come up with a robust case to put before a court or tribunal.
Which expert or combination of experts to go for will always depend on the facts of the case and the legal strategy formulated to run it. In competition cases, the financial and economic factors involved would support a combination of expert economic and accounting evidence as a good way to get a sensible result. We should work together more often!
Rob: In line with what I said above, modelling is an abstraction of reality to make analysis feasible. How we view the world is determined by our background and experiences. In general, it is recommended to incorporate experts from different backgrounds in order to allow questioning of the assumptions and conclusions. Creating a team of experts that recognise their respective limitations and respect each other’s input may sound over the top, but may in fact be the most effective solution.