Summary

Damien Geradin, founding partner, Geradin Partners

A decade after the Consumer Rights Act 2015 grafted an opt-out class action onto UK competition law, the Competition Appeal Tribunal has become the busiest forum for competition class actions in Europe. Claims worth tens of billions of pounds are now pending before a tribunal that, ten years ago, had never certified a single collective claim. This article examines the regime from the vantage point of a competition lawyer trained in the civil-law tradition. Its thesis is that the very features that make the CAT so formidable an instrument of redress are also the source of its principal pathologies, and that the regime now stands at an inflection point at which a measure of continental discipline might usefully be borrowed. The criticisms are directed at the design of the regime, not at the institution or its judges.

Origins, architecture and growth

The regime is a creature of the Consumer Rights Act 2015, which inserted sections 47B to 47E into the Competition Act 1998 and allowed damages to be awarded on an aggregate basis. Its opt-in predecessor under the Enterprise Act 2002 produced a single claim and failed. Merricks then lowered the certification bar and released a wave of litigation.

The Tribunal itself is unusual. A collective action is heard by a chairman, usually a High Court judge or a competition silk, sitting with two ordinary members who are often economists. Economic expertise sits on the bench rather than in the witness box. Continental systems achieve something similar only through specialised chambers or court-appointed experts.

Growth has been explosive, but the record of delivery is thin. Of the sixty-nine claims brought by May 2026, three had produced a final judgment after trial, two of which failed. Only Kent v Apple has delivered a first-instance judgment for a class, of around GBP 1.5 billion. The first contested collective settlement was approved in February 2025, nine years after the claim was issued. However, a second settlement was approved in September 2026.

The Tribunal through continental eyes

Four differences stand out between the UK and continental regimes. The first is opt-out. Almost every CAT claim is opt-out; on the Continent the opt-in model prevails, with the Netherlands and Portugal the exceptions. The consequence is not merely doctrinal. Where loss is dispersed in small amounts across many victims, nothing binds a class unless its members come forward, so the rational course for each is to do nothing and the harm goes uncompensated. Continental practitioners bundle claims instead through powers of attorney, assignment vehicles or the French fiducie, models the Court of Justice endorsed in ASG2 and the German Federal Court of Justice confirmed in May 2026.

The second is orality. English procedure treats the trial as a single concentrated event, with live cross-examination and hot-tubbed experts. The result is a long succession of case management conferences and trials of six to twelve weeks. A continental collective hearing may take a day. The oral model tests evidence far more searchingly, but it is expensive, slow, and hospitable to tactical exploitation.

The third is disclosure, which corrects the information asymmetry between wrongdoer and victim and is what makes standalone abuse claims viable at all. The Antitrust Damages Directive gives continental claimants a pale shadow of it. But disclosure is also the single largest driver of cost and delay in CAT proceedings.

The fourth is duration. Both systems are slow, but their delay is differently distributed: the continental claim is slowed at the front end, the English claim in the oral theatre of trial and in the appellate corridors beyond it. English delay is markedly the more expensive, because every month of it is billed to the class.

Criticisms

Three criticisms are developed. The first is cost. In Waterside v Mowi the Tribunal refused certification where a budget of some GBP 21 million served a claim worth between GBP 1.60 and GBP 8.80 per class member, and warned that class actions can offer irresistible benefit to lawyers and funders while delivering little to the class. The picture must be qualified: Waterside is not representative, classes made of businesses can ensure significant take up, and lawyers and funders carry real risk.

The second is the predicament of the class representative, who is asked to govern complex litigation, supervise lawyers and funders, and face cross-examination, while expected to be remunerated by reference to public-sector pay scales. Underpaying the representative does not protect the class; it removes the representative’s independence. The representative is simultaneously told to build an advisory committee and criticised for its cost. This does not make sense and could disincentivize competent people to become class representatives.

The third is procedural attrition. Jurisdiction challenges, strike-out applications, contested certification, appeals, funding challenges after PACCAR, and now decertification and variation applications after Evans v Barclays Bank allow a well-resourced defendant to shift the centre of gravity away from the merits. Ennis v Apple has absorbed nine successive applications, almost all unsuccessful, in a claim the Tribunal called unusually strong. The inequality is informational as well as financial: the class representative’s budget is disclosed and scrutinised, the defendant’s is not. Further concerns include low distribution to consumers, the reliability of expert methodology, copycat claims, the shoehorning of non-competition grievances into competition clothing, and the Tribunal’s own resources.

Proposals

The paper makes seven proposals. Reverse PACCAR by legislation, with retrospective effect, and regulate funding agreements separately. Certify intensively but once, and set a stringent threshold for reopening orders after Evans. Make cost budgeting and cost capping routine, on both sides of the record, and treat reasonable governance expenditure as legitimate rather than as a presumptive vice. Make a robust distribution plan a condition of certification and decide openly whether the regime serves compensation, deterrence or both. Mark procedural attrition in costs. Borrow selectively from the Continent by disciplining the length of trials and deciding more business on the papers. Finally, resource the Tribunal, for private collective actions are the mechanism by which harm that public enforcement cannot reach is confronted, and they deserve to be funded as an instrument of public policy.

(*) The full version of the paper is available here: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=7258898