Economic Expert Evidence Under The Microscope: Lessons From The U.K. Courts

This article first appeared in an issue of International Comparative Legal Guide - Class and Group Actions 2026.

Economic analysis can be instrumental in competition litigation cases by providing structured and rigorous frameworks and methodologies through which the effects of alleged breaches of competition law may be assessed. In recent years, economics has come to occupy a central position in competition litigation, shaping both substantive outcomes and evidentiary practice. Nonetheless, there remains considerable debate about the most effective ways of incorporating economic evidence into competition litigation proceedings. This chapter explores that debate from an economist's perspective, highlighting key principles – drawn from recent cases – that can guide the credible and effective presentation of economic evidence to assist the courts.¹

U.K. competition litigation: A shifting landscape

In recent years, U.K. competition litigation has undergone a notable transformation. Historically, the focus was on follow-on cartel damages claims, where companies sought compensation after a competition authority had already established an infringement. In such cases, claimants were not required to prove liability; instead, economists' primary role was to assist the courts in determining the quantum of damages suffered by claimants, if any, due to the alleged infringement.

However, the litigation landscape is evolving. There is now a marked increase in standalone actions, including class actions and private abuse of dominance claims, where liability must
be established from the outset. These cases often involve more complex theories of harm than simply assessing whether prices were elevated by a cartel.

The evolving nature and scope of economic expert evidence

This shift in the litigation landscape has elevated the importance of economic expert evidence. Economists have become increasingly central to competition disputes, not only for the quantification of damages but also for the assessment of liability. Their input is often required from the earliest stages of proceedings – particularly in class actions, where courts must assess whether a case is suitable to be certified prior to seeing any factual or witness evidence. In such cases, the proposed economic methodologies have assumed a pivotal role.

As competition claims have become more diverse and complex, so too has the nature and breadth of the evidence being presented. Experts are increasingly deploying sophisticated methodologies to test novel theories of harm, construct complex counterfactual scenarios (i.e., what outcomes would have occurred absent the conduct in question), and quantify damages. Their analysis has been drawing not only on competition economics tools, but also insights from behavioural economics (for example, to understand how consumer behaviour may have been influenced by specific conduct), and technical/industry expert evidence (for example, to understand the operation of algorithms/evolution of markets).

Judicial challenges and emerging principles

For the courts, this evolution presents a growing challenge as they are grappling with how best to evaluate and interpret highly technical expert submissions. Judges are increasingly confronted with voluminous and technically complex expert reports. They must navigate competing methodologies – often grounded in intricate economic modelling and theoretical assumptions – while ensuring that the evidence remains reliable and fact-based.

Against this backdrop, we have reviewed several recent cases to identify key themes in how courts in the U.K. have assessed expert evidence. Our analysis highlights important principles that underpin the effective use of economics in competition litigation.

Key principles underpinning effective use of economic evidence in competition litigation

Based on our review of recent competition litigation cases, the following core principles emerge that can be considered central to the effective use of economic evidence in such cases.

• Applying robust economic frameworks grounded in facts: Economic frameworks and methodologies must be robust and firmly grounded in the commercial realities and market dynamics at hand.
• Assisting the court effectively: Experts must remain aware that their role is to assist the court. This requires clarity, transparency, and brevity in presenting evidence – ensuring that complex analysis is accessible and helpful to judicial decision-making.
• Demonstrating independence: The credibility of expert evidence depends on the expert's independence. Opinions should be objective, impartial, and free from advocacy.
• Engaging with evidence from different disciplines: Interdisciplinary collaboration can be important. Economic evidence may need to be integrated with insights from behavioural science, data science, or technical domains (e.g., algorithms or digital platforms) to fully capture the nature and impact of the conduct in question.

Applying robust economic frameworks grounded in facts

Economic frameworks can help to understand and assess the issue at hand in competition litigation cases, and in doing so they can provide deep insights. However, there may be several potentially relevant economic frameworks, depending on the nature of the markets in question. Also, how economic frameworks are applied in a case will vary depending on the interpretation of the facts.

The first principle in providing economic evidence is to set out the appropriate economic framework with some careful thought, ensuring it reflects the economic expert's understanding of the market and facts of the case. While economic models cannot capture every aspect of a market, they should reflect its most salient features.

The starting point is to establish a testable hypothesis around the theory of harm – acknowledging any uncertainties regarding the scope, nature, or extent of the conduct or agreements in question. Once the theory of harm is defined, the next step is to consider the available economic frameworks and methodologies for testing it, alongside the likely availability of data. This ensures that the analysis is both feasible and grounded in empirical evidence.

The recent Amazon carriage disputes offer good examples of how proposed economic frameworks can materially influence judicial outcomes in competition litigation. These disputes involved two parallel collective proceedings – one at the consumer level and one at the retail level – both alleging abusive conduct, including algorithm bias, by Amazon, in its marketplace operations. In evaluating the competing applications, the Competition Appeal Tribunal ("CAT" or "the Tribunal") placed significant weight on the robustness, relevance, and methodological clarity of the economic evidence presented, which ultimately shaped its decisions.

In Hunter v Amazon & Hammond v Amazon,² the CAT ruled in favour of Mr Hammond, finding that his expert's methodology – based on re-running Amazon's algorithm without the alleged bias – was more closely aligned with the nature of the abuse and therefore the counterfactual.³ Ms Hunter's expert, by contrast, proposed using a choice-based conjoint analysis⁴ based on a survey of Amazon's consumers' preferences to identify those instances where purchasers would have chosen an alternative offer to the one featured by Amazon's algorithm. The Tribunal viewed this approach as less relevant to the core allegations, stating that: "It will readily be appreciated that the approach does not necessarily align itself with the true counterfactual (the operation of Amazon's algorithm without the abuse), and runs the risk of importing a different way of ascertaining consumer preference which is at variance with the non-infringing aspects of
the operation of the Amazon algorithm."⁵

The fundamentally different conceptual approaches adopted by the economic experts in identifying the counterfactual were key to the Tribunal's decision-making as regards carriage. A similar approach was adopted by the CAT in BIRA v Amazon & Stephan v Amazon,⁶ where carriage was awarded to Professor Stephan as it found that the approach proposed by his expert,
also based on re-running Amazon's algorithms absent the alleged abuse(s), "has the benefit of more directly tracking the effects of Amazon's alleged abuses".⁷ By contrast, the Tribunal expressed concerns about BIRA's expert's approach, which it described as appearing "to be based on very simplistic assumptions of what is likely to have happened in the counterfactual", and depending "on a number of different specifications at each stage, requiring adequate and reliable data".⁸

Economic frameworks and methodologies are unlikely to persuade courts if they are detached from the factual and commercial realities of a case. The Cabo v MGA⁹ case highlights the importance of ensuring that economic models are not only theoretically sound but also grounded in the commercial and factual context of the dispute.

Cabo Concepts, a UK toy start-up, alleged that MGA Entertainment – the maker of LOL Surprise! – engaged in exclusionary conduct that prevented the successful launch of Cabo's rival product, Worldeez. The case required the High Court to assess not only whether MGA's conduct breached competition law, but also whether it caused Cabo to suffer quantifiable loss.

In this case, the Court scrutinised the economic logic and evidentiary basis of the experts' economic models but ultimately relied more on factual evidence in reaching its conclusion, finding that neither side's expert analysis was sufficiently tied to the realities of the market. For example, the Court appeared to agree with the criticism of the claimant's expert's simulation model for assuming a form of competition that did not reflect how the market operated. The claimant's expert used a differentiated product Bertrand simulation model to predict Worldeez's counterfactual sales, volumes and revenues.¹⁰ This was a model that assumed static price competition, in a market with a relatively small number of firms competing on price.¹¹ The reality of the toy market that this model was attempting to simulate was found to be different – it was a volatile market with a large number of competitors, with significant competition on non-price factors.¹²

While the Court acknowledged that simulation models of the type used by the claimant's expert could be useful in such cases, their reliability depends, however, on the right assumptions being made, particularly as to the likely conditions of competition and consumer demand in the non-infringement scenario.¹³ In this case, the claimant's expert's model was found not to be aligned with the realities of the market – for instance, it assumed a form of competition that was not tied to facts and it assumed too high a quality of Worldeez (based on inputting into the model what the appropriate comparators to Worldeez were). Consequently, the model predicted counterfactual prices for Worldeez that were "wholly unrealistic".¹⁴ While the Court was more accepting of the defendant's expert's simpler analysis of Wordleez's counterfactual profits based on comparator products, his model was also criticised for not using realistic cost assumptions that were aligned with his lower revenue projection of Worldeez.¹⁵

It is worth acknowledging that it can be particularly challenging to establish the counterfactual and estimate damages in exclusionary abuse of dominance cases such as this, where the excluded business was in its infancy and has little track record that can reliably assist the experts with their analysis of what might have happened in the counterfactual. In such circumstances, it is even more important for experts to appropriately analyse and interrogate the facts available (for example, whether the claimant's business projections are realistic) as well as engage thoroughly with other complementary forms of evidence, such as industry expert evidence, which may be critical in identifying realistic comparators in the counterfactual. The Court has highlighted that this is preferable to engaging in "spurious precision"¹⁶ through sophisticated modelling techniques.

This theme is consistent with earlier cases where courts have rejected economic models that were misaligned with the factual evidence. A notable example is BritNed v ABB,¹⁷ where the High Court rejected the claimant's econometric model for estimating overcharges.

BritNed, a joint venture between National Grid and Dutch operator TenneT, sued ABB following the European Commission's finding of a cartel in the high-voltage submarine and underground cable market. BritNed claimed the cartel led to inflated prices for its U.K.-Netherlands electricity interconnector.

The claimant's model was found to be largely divorced from the factual context of the case.¹⁸ For instance, it included data from all of ABB's projects, despite BritNed being a bespoke project, and incorporated both underground and submarine power cables, even though the claim concerned only submarine cables. Additionally, the model used control variables such as order backlog and time trend, which were not supported by the factual record.¹⁹

While the Court was more receptive to the defendant's margin analysis, it still noted that also that analysis "cannot be followed blindly", due to some limitations.²⁰

Ultimately, the Court's conclusion on damages was highly influenced by the factual evidence – such as the fact that ABB employees involved in negotiating the BritNed tender were unaware of the cartel – rather than on economic modelling. It is worth noting that the Court thoroughly engaged with the economic evidence presented in that case (including the econometric modelling) and the rejection of economic evidence was due to those not being tied to the facts of the case. This case reinforces the principle that economic evidence must be closely tied to the facts to be relevant and persuasive.

The Granville v LG Display ²¹ case, a follow-on claim based on the European Commission's LCD cartel decision, also highlights the importance of using an appropriate damages estimation methodology that is aligned with market realities. The High Court rejected the claimant's simplistic price extrapolation model and gave greater weight to the defendant's more robust econometric analysis, which accounted for the impact of multiple market factors unrelated to the cartel that may have affected prices of LCD panels²² – a critical element of estimating how much overcharge is caused by a cartel.²³

While the Court applied the "broad axe" principle to estimate damages amid imperfect evidence, it emphasised that methodological rigour and factual grounding are essential.²⁴ The case reinforces that economic expert evidence must be both credible and context-specific to carry weight.

It follows from the above that economic models must be firmly grounded in the factual context of the case – this is both sensible and uncontroversial to state. However, challenges can arise when key facts are unclear or not yet established, raising questions about the appropriate approach experts should adopt in such circumstances. This issue was central in Stellantis v Autoliv,²⁵ a standalone cartel damages claim.

Faced with limited documentary and witness evidence, the claimant's expert employed an econometric model to test for the presence of cartel effects from various points in time. The CAT criticised this approach, stating that the existence and duration of a cartel are factual matters and should not be inferred solely through econometric modelling. While not the sole reason for the Tribunal's finding on the overcharge, the case highlights the practical difficulties experts face when required to construct robust economic models without a settled factual foundation. This issue is particularly relevant as the landscape evolves with more standalone cases where facts are often decided at trial.

Notably, the defendant's expert did not present their own overcharge estimation model²⁶ and the only useable data that was available was the claimant's data.²⁷ This, however, was not criticised by the Tribunal. This raises an important question regarding the role of defendant experts in competition litigation cases: will it be sufficient for these experts to critique the claimant's model without presenting an alternative, particularly where, as noted in the judgment, "it is more likely than not that Autoliv has engaged in some cartel activity by object in relation to the Stellantis groups"?²⁸

This naturally leads to the next theme we explore: the duty of economic experts to the court.

Experts must consistently uphold their duty to assist the court effectively

In addition to producing economic analysis that is consistent with the facts, economic experts must remain mindful that their duty is to the court, which includes presenting evidence with transparency, clarity, and brevity. Courts should not be expected to navigate lengthy or constantly evolving submissions. What they need are the precise reasons behind differing expert conclusions – not just the assumptions made, but also their rationale and materiality.

This challenge was evident in Cabo v MGA,²⁹ where the Court described the process as "unsatisfactory" due to frequent model updates – even post-trial – and unclear submissions.³⁰ The Court was critical of the sheer volume and timing of expert reports – 33 economic and valuation reports and letters were filed, many revised during the trial itself.³¹ This made it difficult for the Court to follow the evolution of the evidence and to distinguish material issues from peripheral details. The judgment made clear that expert evidence should not resemble a negotiation, where each side begins from an extreme position and gradually converges under pressure.³² Instead, experts are expected to start from a defensible, objective standpoint and engage constructively with opposing views during joint meetings. The failure to do so, and the maintenance of entrenched positions up to and beyond trial, was seen as undermining the credibility of the evidence.

Similarly, in Royal Mail & BT v DAF Trucks,³³ the CAT criticised the expert evidence in the case for being excessive and disproportionate. Although expert input was central to the trial, the sheer volume – 48 reports spanning thousands of pages – was seen as unjustified, especially for secondary issues, which had limited impact on the overall quantum. The Tribunal acknowledged the need for thoroughness in a precedent-setting case but emphasised that a more measured and proportionate approach would have been preferable.³⁴

In this regard, joint expert statements can play an important role – they can help experts to assist courts more effectively by clarifying agreements and pinpointing genuine points of disagreement. By requiring experts to meet and discuss their views, these statements can facilitate narrowing of the issues in dispute between the experts, allowing courts to focus on the most material differences. They can also enhance judicial understanding by explaining not just what the experts conclude, but why they diverge – making technical economic evidence more accessible and meaningful.

In addition to improving clarity, joint statements can promote procedural efficiency and proportionality. They can reduce duplication, streamline submissions, and encourage constructive engagement between experts.

Courts have criticised poorly prepared joint statements that fail to narrow issues or clearly articulate the basis of disagreement, highlighting the need for careful and purposeful drafting. In the ongoing Consumers' Association v Qualcomm ³⁵ case, the joint expert statements were deemed too technical and unnecessarily detailed, prompting the Tribunal to request a revision.³⁶ This highlights the Tribunal's insistence that expert evidence remains focused, accessible, and manageable.

In addition to the use of joint expert statements, courts are increasingly developing various procedural innovations to better manage the growing complexity and volume of economic evidence in competition litigation. One such innovation includes the imposition of page limits on expert reports. In Consumers' Association v Qualcomm, the Tribunal introduced page limits not only on some expert reports but also on annexes and joint statements ³⁷ – seeking to balance analytical depth with accessibility and a focus on issues that really move the needle. Likewise, in BIRA v Stephan, the introduction of expert summaries sought to distil key points while preserving substance. These measures reflect an increasing judicial emphasis on ensuring that expert evidence is both comprehensible and useful to the decision-making process.

Courts have also introduced structural innovations such as expert-led processes in disclosure and methodology development. These approaches, trialled in collective proceedings like The Trucks Second Wave Proceedings ³⁸ and Merchant Interchange Fee Umbrella Proceedings,³⁹ were aimed at encouraging early collaboration between experts to align methodologies and data requirements, thereby attempting to improve the relevance and coherence of evidence presented at trial.

Courts may adopt an expert-led process where the structure of the proceedings (for example, the Merchant Interchange Fee Umbrella Proceedings) may dictate that a traditional disclosure process involving all parties to the claim would be impractical and disproportionate. Trial 2A of those proceedings involved the assessment of whether and the extent to which merchants paying higher merchant service charges as a result of potentially anti-competitive Multilateral Interchange Fees ("MIFs") would have passed these on to their retail customers. The case involved many hundreds if not even thousands of retailers spanning a wide range of sectors in the economy.

In such cases, courts may wish to understand what methodologies the economic experts would propose to adopt, and what data and/or disclosure would be necessary in order to implement those methods. In some cases, the experts' preliminary engagement (including reports setting out the reasoning for their proposed approaches) may assist in narrowing the areas of dispute, or may provide a basis to the court to reach decisions that would lessen the risk of expert methodologies leading to "ships passing in the night". For instance, the CAT, in the Umbrella Proceedings, agreed with the majority of experts that a simulation exercise would not likely be informative so as to be useful for the purpose of addressing pass-on.

However, the expert-led process was also controversial in these proceedings, with some parties concerned that the process (and any preliminary decision arising from the experts' discussion of methodologies) may restrict their ability to put forward their preferred defence; or, alternatively, that the limitations on required disclosure may raise questions of fairness to all parties in meeting the case put to them. Judgment for the pass-on trial is awaited, and it will be interesting to see how the Tribunal engaged on these types of concerns.

Another less recent yet noteworthy development is the increasing use of concurrent expert evidence – or "hot tubs". This format allows experts to present and respond to questions simultaneously, facilitating direct comparison and enhancing transparency. While implementation varies, the CAT continues to refine this approach to ensure fairness and consistency.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.