Evans v Barclays Bank Plc & Ors

Factual and Procedural Background

The case concerns the collective proceedings regime introduced into the Competition Act 1998 by amendments in 2015. Under section 47B, collective actions may be brought in a specialist competition tribunal (“the Tribunal”) to obtain aggregate damages on behalf of a class of persons affected by alleged infringements of competition law, without proving each individual’s loss.

The statutory scheme requires the Tribunal to act as gatekeeper. Collective proceedings may go forward only if the Tribunal grants a collective proceedings order, which must state whether the proceedings are to be “opt-in” or “opt-out”. In opt-in proceedings, only those class members who actively join the claim are represented; in opt-out proceedings, all within the class are included unless they take steps to opt out (with a separate opt-in requirement for persons domiciled outside the United Kingdom).

The Respondent (the proposed class representative) applied to the Tribunal to bring follow-on damages claims, under section 47A of the 1998 Act, arising out of two settlement decisions of a European competition authority (the “Commission”) issued on 16 May 2019 concerning foreign exchange (“FX”) spot trading in certain major currencies. Those decisions (often referred to by the names of chatrooms used by participating traders) found that several banks had engaged in a single, continuous infringement by object of the EU and EEA prohibitions on anticompetitive agreements. The infringement involved traders at competing banks exchanging commercially sensitive, current or forward-looking information in private chatrooms, thereby enabling occasional coordination of spreads and trading behaviour.

The Respondent’s proposed collective proceedings are follow-on claims, relying on those Commission decisions to establish infringement. The proposed claim alleges that, by virtue of the infringements, the defendant banks were able to widen bid-ask spreads on FX spot (and related) transactions in the relevant currencies, causing loss to two proposed classes: (i) Class A, comprising persons who traded directly with the defendant banks in the EEA during the periods when those banks participated in the infringements; and (ii) Class B, comprising persons who traded with other financial institutions in the EEA, or with a defendant at times when that defendant was not itself participating in the chatrooms, but who allegedly suffered from “umbrella” effects whereby the unlawful conduct reduced competitive pressure across the market.

The Respondent estimated that Class A would comprise between approximately 14,000 and 42,000 members, and Class B between approximately 28,000 and 42,000 members (with overlap). Indicative aggregate damages were estimated at roughly £2.155 billion before interest, based on the value of commerce and a generic percentage overcharge drawn from economic literature on cartels.

Another proposed class representative (the “Other Applicant”) lodged a competing application for similar collective proceedings. After a substantial certification hearing at which extensive factual and expert economic evidence was adduced, the Tribunal resolved the “carriage” issue in favour of the Respondent, and that ruling was later affirmed on appeal. The Other Applicant has since discontinued and did not participate in the present appeal.

In its main certification judgment, the Tribunal concluded that both applications could in principle be certified as collective proceedings. However, the Tribunal raised on its own initiative the possibility of striking out the claims, expressing serious concern about how the Respondent proposed to establish causation between the limited infringements identified in the Commission decisions (information exchanges among a handful of traders, of short-lived commercial value) and the very broad alleged market-wide loss affecting trillions of pounds of FX trading by numerous institutions and customers.

Having considered the pleadings and extensive expert material, including post-hearing submissions requested by the Tribunal to articulate the “theory of harm”, the Tribunal concluded that the Respondent’s pleaded case on causation did not disclose reasonable grounds for the claim and was liable to be struck out. Nonetheless, in light of the novelty and developing nature of the law, and because the applicants had not had the benefit of seeing the Tribunal’s fully developed reasoning before, the Tribunal decided not to strike out immediately, and instead allowed an opportunity for reformulation.

On the separate question whether to certify the claim on an opt-in or opt-out basis, a majority of the Tribunal refused to grant an opt-out order. The majority treated the perceived weakness of the pleaded case on causation as a powerful factor against opt-out proceedings and also concluded that it was, overall, practicable for the claims to be brought on an opt-in basis, particularly for large and sophisticated financial institutions with substantial alleged losses. The Tribunal accepted that, as a matter of fact, the proceedings would not be brought as opt-in (because key potential claimants and funders lacked appetite for such proceedings), but considered that this reflected commercial decisions rather than lack of practical feasibility. The Tribunal therefore stayed the applications and granted permission to submit revised applications for opt-in certification within a specified period.

A dissenting Tribunal member would have certified the claim on an opt-out basis, placing heavy weight on access to justice and the evidence that opt-in proceedings would in practice not proceed or would attract very few class members.

The Respondent and the Other Applicant appealed. Before the appeal hearing, the Commission issued two further decisions concerning similar FX conduct in a different chatroom. One was a settlement decision against certain banks; the other (the “ordinary decision”) was an infringement decision against another bank after a contested procedure, setting out more detailed evidential analysis. The Respondent amended the claim form to add a follow-on claim based on these later decisions and joined additional defendant banks.

The Court of Appeal allowed the Respondent’s appeal, holding that the Tribunal had erred in how it treated both the strength of the claim and the practicability of opt-in proceedings, and in its handling of access to justice and policy considerations. It remitted the matter to the Tribunal for reconsideration, while expressing a more favourable view of the merits by reference inter alia to the later Commission ordinary decision. The present appeal is brought by the defendant banks (the “Appellants”), who contend that the Court of Appeal wrongly interfered with the Tribunal’s evaluative and discretionary judgments and misdirected itself on several legal issues.

Legal Issues Presented

1. Whether the Court of Appeal was wrong to find that the Tribunal erred in relying on its assessment that the claims were weak as a factor weighing against certification of opt-out collective proceedings (Issue 1).
2. Whether the Court of Appeal was wrong to find that the Tribunal erred in its assessment of the practicability of bringing the claims as opt-in collective proceedings (Issue 2).
3. Whether the Court of Appeal was wrong to hold that the policy principles of facilitating the vindication of rights and deterring future wrongdoers are factors which weigh in favour of opt-out proceedings generally or in the circumstances of this case (Issue 3).
4. Whether the Court of Appeal was wrong to treat the later Commission ordinary decision (concerning a different infringement and a third party bank) as admissible and probative on a potential strike-out application and to rely on it when assessing the strength of the Respondent’s claim (Issue 4).

Arguments of the Parties

Appellants’ Arguments

• The Tribunal, as the specialist body entrusted with managing collective proceedings, has a broad discretion in assessing both the strength of claims and the practicability of opt-in proceedings under rule 79(3) of the Tribunal’s procedural rules. The Court of Appeal had no proper basis to interfere with those evaluative judgments.
• The Court of Appeal’s approach would, if generally followed, allow almost all collective proceedings that are not liable to be struck out or summarily dismissed to proceed on an opt-out basis, contrary to the carefully balanced statutory scheme and legislative intent to protect defendants from unmeritorious but oppressive litigation.
• The Tribunal was entitled to regard the Respondent’s causation case as very weak even if it decided, as a matter of procedural fairness, to postpone a final strike-out decision. It was legitimate to treat that weakness as a powerful factor against conferring the additional leverage of opt-out proceedings.
• The Tribunal’s conclusion that, viewed objectively, it was practicable for larger, sophisticated class members with substantial claims to proceed on an opt-in basis was a permissible assessment open to the Tribunal. The fact that many such potential claimants had chosen not to participate in opt-in proceedings reflected commercial choices rather than practical impossibility.
• The Court of Appeal erred in elevating general policy aims of facilitating redress and deterrence into factors that positively favour opt-out proceedings, and in treating access to justice as effectively requiring opt-out certification whenever opt-in proceedings were unlikely to be brought.
• The Court of Appeal wrongly held that the later Commission ordinary decision against a third-party bank (concerning a different chatroom and infringement) was admissible and probative. Under principles derived from the common law rule excluding prior findings of other decision-makers, such findings are not admissible as evidence of facts against non-parties, and in any event the later decision did not assist in overcoming the fundamental causation difficulties in the Respondent’s pleaded case.

Respondent’s Arguments

• The Tribunal, having expressly decided not to reach a final strike-out decision and to allow the Respondent an opportunity to reformulate the case, acted inconsistently and unfairly in treating its necessarily provisional assessment of the merits as if it were final for the purpose of refusing opt-out certification, particularly knowing that this would effectively terminate the claims.
• The strength of the claim should, in most cases, be a neutral factor when choosing between opt-in and opt-out proceedings, particularly at the certification stage when evidence and disclosure are incomplete, and the Tribunal should not operate a “sliding scale” whereby weaker claims are channelled to opt-in and stronger claims to opt-out.
• The Tribunal erred in its practicability assessment by focusing on overall average claim size and by giving insufficient weight to evidence showing that most potential class members had relatively modest individual claims and that, in practice, opt-in proceedings would not be viable. In particular, the Tribunal gave inadequate weight to the fact that, absent opt-out certification, there would effectively be no proceedings.
• Policy goals of enabling the vindication of rights and deterring future infringements, as recognised in earlier case law and legislative materials, should be treated as strong factors favouring opt-out certification in circumstances where opt-in is economically or practically unattractive.
• The later Commission ordinary decision, although not addressed to the Appellants, was admissible at least for interlocutory purposes (such as resisting strike-out) and had substantial probative value, because it involved similar FX trading conduct in a parallel chatroom and contained detailed economic and evidential analysis which shed light on the likely effects of the infringements in issue.
• More generally, the Tribunal should not be constrained by the common law rule excluding prior findings of other decision-makers, given its broad statutory powers over admissibility of evidence and its sophistication in assessing what weight, if any, to attach to such material.

Table of Precedents Cited

Precedent	Rule or Principle Cited For	Application by the Court
Case A [2020] UKSC 51; [2021] Bus LR 25; [2021] 3 All ER 285	Set out principles for granting collective proceedings orders in the Tribunal; merits not relevant to deciding whether to grant a collective proceedings order, but legislative history and discussion highlight both the access-to-justice benefits and the potential misuse and “leveraging” risks of opt-out, aggregate-damages class actions.	Used to distinguish the certification decision (where merits are generally irrelevant) from the opt-in/opt-out choice (where rule 79(3) expressly makes strength relevant), and to support the view that the Tribunal may take into account the leveraging effect and risks of abuse when deciding whether to permit opt-out proceedings.
Case B (Case C-441/11P) EU:C:2012:778; [2013] 4 CMLR 8	Explains the concept of a single, continuous infringement in EU competition law and its consequence that an undertaking can be treated as liable for the entire infringement, including acts of others, during its period of participation.	Cited to clarify the legal significance of the Commission’s finding that the FX conduct in the relevant chatrooms constituted a single, continuous infringement and that each addressee could be held liable for the infringement as a whole.
Case C (Case T-84/22) EU:T:2025:752; [2025] 5 CMLR 12	Describes the Commission’s method for calculating fines, using a proxy for “value of sales” which is not restricted to sales actually affected by the infringement but is intended to reflect the economic importance of the infringement.	Referred to illustrate how the Commission approached fines in the FX decisions, using revenues and adjustment factors based on bid-ask spreads, thereby explaining the background to the administrative penalties on the banks.
Case D [2021] UKSC 50; [2022] AC 1217	(i) Records that an earlier consumer collective action regime resulted in only one case with very limited take-up, demonstrating its ineffectiveness; (ii) confirms in a representative action context that, for certain collective claims, there is no need to quantify and prove each individual’s loss.	Used to show why a more effective collective proceedings regime, including opt-out, was introduced, and to emphasise that the Respondent’s burden is eased by the ability to claim aggregate damages without proving individual loss, making the absence of a coherent aggregate causation theory more striking.
Case E (Case C-557/12) EU:C:2014:1317; [2014] 5 CMLR 5	Discusses the “umbrella effect”, whereby anticompetitive conduct by some firms can affect prices charged by non-participants.	Cited to explain the economic concept underlying the Respondent’s Class B “umbrella” claims, namely that the cartel allegedly enabled non-participating banks also to widen spreads.
Case F [2022] EWCA Civ 593; [2022] Bus LR 660; [2023] 1 All ER (Comm) 667	Holds that there is no statutory presumption in favour of opt-in or opt-out collective proceedings; strengthens the principle that the Tribunal has a broad evaluative discretion and that appellate courts should be slow to interfere. Also observes that in many cases at certification stage the Tribunal may be hard-pressed to form a clear view on merits, so merits may sometimes be a neutral factor.	Relied on to reaffirm the breadth of the Tribunal’s discretion and the need for appellate restraint, while clarifying that where the Tribunal can form a reasoned view of strengths or weaknesses, merits should not be treated as neutral by default.
Case G [2004] EWCA Civ 266; [2004] PNLR 35	Indicates that strike-out may be inappropriate where the law is uncertain and developing, particularly in novel areas.	Cited to support the Tribunal’s decision not to strike out the claims immediately, despite finding the pleading deficient, because collective competition damages claims are in a developing area and the applicants had not yet addressed the Tribunal’s fully articulated concerns.
Case H [2022] EWCA Civ 464; [2022] 4 WLR 48	Confirms that appellate courts should assume that first-instance courts have considered all the evidence unless there are clear indications otherwise, and should be slow to infer oversight simply from lack of express reference.	Applied to reject an argument that the Tribunal failed to consider certain sub-groups within the proposed class when assessing practicability, noting there was no clear basis to conclude the Tribunal had ignored relevant evidence.
Case I [2025] UKPC 37	Reiterates limits on appellate interference with fact-finding and evaluative judgments, stressing that absence of express mention in a judgment does not mean a point was ignored.	Cited alongside Case H to support the conclusion that the Tribunal was entitled to be assumed to have considered the detailed data on claim sizes and class composition when assessing practicability.
Case J [1943] KB 587	Establishes the common law rule that prior findings of fact by another decision-maker (such as a criminal conviction) are not admissible as evidence of those facts in subsequent civil proceedings, based on fairness and the need for the tribunal itself to evaluate the underlying evidence.	Used as the foundational authority for the principle that findings in a Commission decision addressed to a third party cannot be admitted as factual evidence against non-addressee defendants in follow-on damages claims.
Case K [2014] EWCA Civ 257; [2015] QB 265	Explains that the rule in Case J rests on fairness: the fact-finder must form its own view on the evidence rather than adopt another body’s conclusions. Clarifies that prior reports or decisions may be used to identify or record evidence, but not as evidential opinions on disputed factual issues.	Cited both to reinforce the fairness rationale for excluding third-party findings and to distinguish between admissible use of prior material as a source of primary evidence and inadmissible reliance on another decision-maker’s evaluative conclusions.
Case L [2003] 2 AC 1	Applies the principle underlying Case J to inquiries, holding that an inquirer must make findings based on its own assessment of evidence.	Relied on to show that the fairness principle extends beyond courts to other decision-makers, and by analogy to support its application in the Tribunal.
Case M [2020] EWCA Civ 1475; [2021] Bus LR 1610	Addresses which aspects of a Commission decision are binding on national courts in follow-on claims and which factual findings remain open to challenge.	Referenced to highlight that, although infringement findings against addressees are binding in follow-on claims, not all factual recitals in such decisions are conclusive, and that the present issue concerns a different question: the admissibility of findings in a decision addressed to a third party.
Case N [2023] CAT 9	Holds that while the Tribunal is not formally bound by the rule in Case J, it should adopt the same principle and, at trial, should not attach weight to findings of fact by other courts or regulators, because it must make its own primary findings.	Quoted to demonstrate the Tribunal’s own adopted practice on prior findings and to question the Court of Appeal’s contrary view that the Tribunal could freely rely on detailed findings in a later Commission decision involving a third party.
Case O [2021] EWHC 3440 (Ch)	Warns against “satellite litigation” over the circumstances and evidential basis of earlier decisions if such decisions are treated as probative in later proceedings.	Invoked via Case N to illustrate practical difficulties in determining the weight of prior decisions, which further support reluctance to admit or rely on such findings in subsequent litigation.
Case P [2011] UKSC 34; [2012] 1 AC 531	Affirms that procedural discretion and case management powers are constrained by fundamental principles of natural justice and fairness, especially where a party has had no opportunity to contest evidence used against it.	Used to reinforce that it would be fundamentally unfair to use findings in a Commission decision against defendants who were not parties to that decision and had no opportunity to contest them or appeal.
Case Q [1964] AC 40	Rejects the “heresy” that principles of fairness in administrative law apply only to decisions characterised as judicial or quasi-judicial.	Cited to rebut an attempt to limit the rule in Case J to judicial or quasi-judicial decisions; the fairness-based exclusion of prior findings applies irrespective of the decision-maker’s formal status.
Case R [1970] 2 QB 417	Applies the broad fairness approach endorsed in Case Q, confirming that procedural fairness requirements are not confined to traditional judicial bodies.	Reinforces the view that fairness-based evidential rules, including the rule against admitting third-party findings, apply to Commission decisions as much as to court judgments.
Case S [2006] UKHL 38; [2007] 1 AC 333	Holds that domestic courts are not required, as a matter of EU law, to defer to factual findings in Commission decisions involving different parties; Commission decisions can be considered as part of the evidence but are not binding absent estoppel.	The Court distinguished an obiter statement suggesting Commission decisions are admissible and potentially persuasive as evidence, concluding that it could not override the fairness-based exclusion of third-party findings derived from Case J in circumstances like the present.
Case T [2009] UKPC 34; [2010] 2 LRC 280	States that a second court cannot sensibly determine what weight to attach to an earlier decision’s findings without itself going into the underlying facts; hence, prior findings themselves cannot properly be relied upon as evidence.	Applied to support the conclusion that the Tribunal would have to examine the underlying evidence in the later Commission decision and could not treat its findings as probative shortcuts.
Case U [2013] EWCA Civ 784; [2013] 2 Lloyd’s Rep 242	Permits reliance, at an interlocutory stage, on findings of another court to show that there is a serious issue to be tried.	Cited (together with later authorities) in discussing whether prior findings may be considered, not as trial evidence, but as indicating the type of evidence that might be available at trial in resisting strike-out or summary judgment.
Case V [2014] EWHC 3233 (Comm)	Similarly accepts that earlier findings can, at a preliminary stage, assist in demonstrating that there is an arguable case and in identifying potential evidence.	Used in conjunction with Cases U and W to ground the proposition that there is a limited interlocutory use for prior decisions, consistent with but not displacing the rule in Case J.
Case W [2023] EWHC 2437 (Ch)	Summarises that there is a limited “preparatory stage” use of prior findings: although inadmissible as trial evidence, they may help to identify what evidence can reasonably be expected to be available at trial.	Endorsed to clarify that, even on this limited interlocutory approach, the later Commission ordinary decision did not materially assist the Respondent’s causation case in this appeal.

Court’s Reasoning and Analysis

Issue 1: Relevance of the Strength of the Claims to Opt-In/Opt-Out

The court held that the Tribunal’s assessment of the merits and the weight it attached to that assessment when refusing opt-out certification were within its lawful discretion and that the Court of Appeal had no proper basis to interfere.

First, the court rejected the Court of Appeal’s view that there was an inconsistency or unfairness in the Tribunal’s approach. Rule 41 and rule 43 expressly empower the Tribunal, on its own initiative, to consider strike-out or summary judgment at any stage, even absent a defendant’s application, reflecting the need to conserve judicial and party resources in large collective actions. The Tribunal had interrogated the causation theory in depth, drawing on extensive pleadings, expert reports and post-hearing submissions, and including the specialist expertise of one of its members in financial economics. It concluded that, even assuming further evidence could be obtained, the pleaded theory did not articulate a plausible route from the limited and episodic information exchanges identified in the Commission decisions to the sweeping allegation of widened spreads on essentially all relevant transactions during long periods.

The Tribunal gave detailed reasons, including that: (i) Class A was not confined to trades actually subject to the information exchanges, but effectively treated all trades with certain banks during wide time windows as affected without a coherent explanation of mechanism; and (ii) Class B relied on an asserted umbrella effect in a market where products were fungible and intra-market competition and substitutability were likely to be high, again without a properly pleaded causal account of how the specific infringements translated into the alleged widespread overcharge. The Tribunal concluded that the pleadings and the underlying expert material amounted only to a broad theoretical position and did not support a legally adequate plea of causation and loss.

Nonetheless, the Tribunal declined to strike out immediately, principally because the area is novel and evolving and because the applicants had not had an opportunity to address the Tribunal’s fully developed concerns. It described the applicants as being in the “last chance saloon” but allowed an attempt at re-pleading. The Supreme Court held that this procedural leniency did not preclude the Tribunal from treating the claim, as presently formulated, as very weak and from giving that assessment substantial weight when deciding whether to superadd the significant pressures inherent in opt-out proceedings.

Secondly, on the broader legal point, the court rejected the Court of Appeal’s suggestion that the strength of the claim is generally a neutral factor in the opt-in/opt-out choice. Rule 79(3) specifically picks out “the strength of the claims” as a relevant consideration “additional” to the general suitability factors in rule 79(2). The court reasoned that the open-textured discretion must be exercised consistently with the Tribunal’s overriding duty under rule 4 to deal with cases justly and at proportionate cost, including maintaining equality of footing between parties, saving expense, ensuring proportionality to the amounts and issues involved, and properly allocating Tribunal resources.

Opt-out proceedings confer powerful advantages on claimants and impose heavy burdens and settlement pressures on defendants, particularly because damages are awarded for the entire class even though not all members will claim. If the Tribunal assesses a claim as very weak, it is materially harder to justify exposing defendants to those additional burdens and the associated risk of settlements that do not reflect the underlying merits. A weak claim tends to undermine the justification for conferring the extra leverage of opt-out proceedings, both in fairness terms and in terms of proportionate cost and resource allocation. Conversely, a strong claim can support an opt-out order by making it more likely that the substantial costs and pressures of opt-out are warranted in the pursuit of meritorious redress.

The court therefore endorsed the Tribunal’s “sliding scale” approach: as a general rule, the weaker the case (even if not so weak as to be struck out), the less justification there is for opt-out certification; the stronger the case, the more readily opt-out may be justified. The court emphasised that this is only one factor among many, but rejected the Court of Appeal’s view that clearing the summary judgment or strike-out threshold logically precludes a finding that a claim is nonetheless “very weak” for the purpose of choosing between opt-in and opt-out.

Applying those principles, the court held that the Tribunal’s conclusion—namely, that the Respondent’s case on causation was fundamentally weak and that this provided a “powerful reason” against opt-out—was carefully reasoned and fully within the scope of its evaluative discretion. The Court of Appeal erred by re-assessing the merits, including by relying heavily on material (the later Commission ordinary decision) that was not before the Tribunal and by speculating that disclosure and broad-brush economic analysis might yield supporting evidence, without addressing the Tribunal’s core point about the mismatch between the limited infringements and the breadth of the alleged loss.

Issue 2: Practicability of Opt-In Proceedings

Rule 79(3)(b) directs the Tribunal, when deciding between opt-in and opt-out, to consider “whether it is practicable for the proceedings to be brought as opt-in collective proceedings, having regard to all the circumstances, including the estimated amount of damages that individual class members may recover.” The Supreme Court held that, where opt-in proceedings are practicable, there is a general implication that opt-in will usually be more appropriate: conferring the extra procedural advantages of opt-out (and corresponding burdens on defendants) is then harder to justify.

The court stressed that this assessment is highly fact and context sensitive, involves balancing competing risks of injustice to both claimants and defendants, and falls within the Tribunal’s specialist case-management discretion. The Tribunal, with expertise in economics and practical experience of collective actions, is best placed to gauge, on an objective basis, the likelihood that reasonable potential claimants would opt in, given the size of individual claims, sophistication of parties, and commercial realities. The court endorsed the Tribunal’s view that practicability must be assessed objectively (by reference to what reasonable claimants in relevant categories would do) rather than by attempting to gather and interpret subjective states of mind and privileged legal advice of numerous potential class members.

The Tribunal had distinguished between (a) a group of financial institutions and larger commercial entities, generally sophisticated and with substantial potential claims, and (b) a large number of individuals and smaller entities with modest claims. It accepted that, for the latter group, opt-in proceedings were not practicable: many such class members would not be aware of their claims or would not regard the small amounts at stake as worth the effort. However, the Tribunal found, based on evidence (including the Respondent’s own expert tables), that many larger institutions had significant potential claims, often in the millions, and were capable of litigating on an opt-in basis, whether individually or in coordinated claims. Their lack of enthusiasm for joining opt-in proceedings was reasonably interpreted as a commercial choice rather than a demonstration of non-practicability.

Standing back, the Tribunal concluded that it would be inappropriate to allow a relatively small “tail” of low-value claims to determine the practicability assessment for the entire class, particularly where the bulk of aggregate claim value lay with large, sophisticated entities for whom opt-in litigation was realistically available. The Supreme Court characterised this as a rational evaluative choice: the Tribunal was entitled to decide that the interests of claimants whose individual claims were commercially negligible (and unattractive to funders) could not, in this configuration, justify granting an opt-out order that would significantly enhance the leverage of large institutional claimants.

The Court of Appeal, by contrast, undertook its own detailed analysis of claim-size distributions, focusing on typical rather than average claim size and emphasising that, for many claimants, even “large” claims might be relatively modest compared with litigation costs. It preferred the reasoning of the dissenting Tribunal member. The Supreme Court held that this was an improper substitution of the appellate court’s factual evaluation for that of the Tribunal, contrary to guidance given in earlier authority. The Tribunal had considered the relevant data, expressly acknowledged the limitations of averages, and accepted that opt-in was not practicable for most small claimants. Its ultimate conclusion—that overall practicability was not negated—fell squarely within its broad margin of judgment. No legal error in the Tribunal’s approach was identified; thus the Court of Appeal ought not to have interfered.

Issue 3: Vindication of Rights, Deterrence and Access to Justice

The court accepted that facilitating the vindication of rights and deterring future wrongdoers are important policy objectives underpinning the collective proceedings regime, as recognised in earlier case law and legislative materials. However, it held that these aims cannot be treated as factors that, in themselves, point towards opt-out over opt-in proceedings.

The statutory framework, including the legislative history and the design of rules 79(2) and 79(3), reflects a balance between two sets of policy goals: (i) enabling effective redress for victims of competition infringements, particularly consumers and small enterprises who might otherwise be unable to litigate; and (ii) protecting businesses from exposure to burdensome and unmeritorious collective litigation. When determining opt-in versus opt-out, the Tribunal must maintain an even-handed stance, starting from neutrality rather than from any presumption in favour of one mechanism.

Accordingly, while the fact that no proceedings would occur absent opt-out is a relevant and potentially weighty consideration, it is not a “trump card”. Rule 79(3) directs the Tribunal to assess practicability of opt-in, not to ask whether opt-out is the only way to bring proceedings. The Tribunal in fact recognised that the likely collapse of the claim if limited to opt-in was a factor strongly in favour of opt-out, but correctly regarded it as one factor among many, to be balanced against the weakness of the claim and the nature of the class.

The court also rejected a one-sided conception of “access to justice”. Both claimants and defendants have rights to fair procedures and to the proper adjudication of disputes on their merits. The defendant banks had already either paid significant fines or obtained immunity in the public enforcement process, reflecting co-operation incentives. Allowing clearly unmeritorious or highly speculative damages claims to proceed on an opt-out basis, thereby leveraging settlement pressure well beyond their substantive merits, would not further the genuine enforcement of competition rules but rather risk over-enforcement, which is contrary to the public interest. The court endorsed the Tribunal’s statement that opt-out certification is not a “last resort” entitlement whenever opt-in is unattractive; it must be justified as appropriate in light of all relevant factors.

Issue 4: Use of a Later Commission Decision Against a Third Party

The Court of Appeal had treated the later Commission ordinary decision concerning a different chatroom and a third-party bank as admissible and as having “probative value” when assessing the Respondent’s prospects of reformulating the claim to survive strike-out. It described the various Commission decisions as following a common template and appeared to infer, from more detailed economic analysis in the later decision, that similar harmful effects were likely to have occurred in relation to the earlier infringements, thereby bolstering the Respondent’s case.

The Supreme Court held that this approach was misconceived. It first reaffirmed the common law principle, derived from Case J and explained in Case K and others, that findings by another decision-maker are not admissible as evidence of the facts found in subsequent proceedings, save in limited categories such as issue estoppel or abuse of process. The principle is rooted in fairness: the tribunal charged with resolving a dispute must base its findings on its own assessment of the evidence, not on a previous adjudicator’s evaluation, particularly where the party against whom the findings are invoked was not a party to the earlier decision and had no opportunity to contest or appeal.

The court rejected the argument that the rule should be confined to judicial or “quasi-judicial” decisions, noting that fairness-based principles apply equally to administrative or regulatory decisions. It also rejected reliance on a brief dictum in Case S suggesting that Commission decisions might be treated as persuasive evidence, holding that this could not displace or qualify the fairness-based exclusion of prior factual findings in proceedings between different parties.

The Tribunal had, in Case N, recognised that it is not formally bound by Case J but had nonetheless adopted the same principle as a matter of its own practice. The Supreme Court endorsed that approach, emphasising that the Tribunal’s broad powers over admissibility of evidence (under its procedural rules) must be exercised consistently with fundamental fairness and cannot authorise admission of third-party findings to prove contested facts against non-parties.

The court acknowledged, drawing on Cases U, V and W, that at an interlocutory stage prior findings can sometimes be used in a limited way: not as evidence of their truth, but as indicators of the type of evidence that might reasonably be expected to be available at trial. However, this does not conflict with Case J, because it does not involve treating prior evaluative findings as factual proof.

Applying this framework, the court concluded that the later Commission ordinary decision did not record evidence directly relevant to the claims against the Appellants based on the earlier FX chatrooms, nor did it materially assist in identifying additional evidence likely to be available at trial that would cure the fundamental causation gap in the Respondent’s case. The decision involved a distinct infringement in a different chatroom, with partially different participants and time periods, and while it elaborated on why the exchanges of information were likely to be capable of harming competition, the earlier settlement decisions already contained comparable findings. The central difficulty for the Respondent—linking sporadic information exchanges among a few traders to an alleged systemic widening of spreads across the entire G10 FX market for prolonged periods—remained unaddressed.

The Supreme Court therefore held that the findings in the later Commission ordinary decision were both inadmissible as evidence against the Appellants and, even on a limited interlocutory view, of no real assistance in strengthening the Respondent’s claim. The Court of Appeal erred in treating them as admissible and probative and in basing its more optimistic view of the merits on that material.

Overall Assessment and Appellate Restraint

Bringing the analysis together, the court reiterated that determining whether to certify collective proceedings as opt-in or opt-out is an exercise in pragmatic, specialist judgment by the Tribunal, involving the weighing of multiple factors, including strength of the claims, practicability of opt-in, class composition, and broader justice and proportionality considerations. The Tribunal’s margin of appreciation in such matters is broad. Appellate courts should not interfere merely because they might have reached a different conclusion; intervention is warranted only if there is a legal misdirection or an irrational or impermissible evaluation.

In this case, the Tribunal carefully considered all relevant statutory and rule-based factors, including those in rule 79(2) and 79(3), the overriding objective, and the evidence on class size, claim values, funding, and access to justice. It reached a reasoned conclusion that the Respondent’s causation case, as pleaded and supported by the existing expert material, was weak and that this strongly weighed against opt-out proceedings. It further concluded, based on an objective assessment of class composition and likely behaviour of reasonable claimants, that opt-in proceedings were, overall, practicable for major institutional claimants, even though they were unlikely to be pursued in practice. The Tribunal also expressly weighed the fact that the claims would probably not go forward on an opt-in basis, but legitimately declined to treat this as determinative.

The Supreme Court held that no legal error in this reasoning had been shown. The Court of Appeal, in contrast, had wrongly recast the merits factor as generally neutral, substituted its own evaluation of practicability for that of the Tribunal, gave undue weight to general policy aims of redress and deterrence without adequately balancing defendant interests, and improperly treated a later Commission decision against a third party as admissible and probative. Accordingly, the Court of Appeal’s interference could not be sustained.

Holding and Implications

APPEAL ALLOWED

The Supreme Court allowed the Appellants’ appeal and reinstated the Tribunal’s decision refusing to grant an opt-out collective proceedings order to the Respondent. The Tribunal’s original orders, including its conclusion that the claims could only proceed (if at all) on an opt-in basis and its grant of an opportunity to reformulate the claims for that purpose, stand restored.

Directly, the consequence is that the Respondent cannot pursue these FX claims on an opt-out basis. Any further attempt to advance them must satisfy the Tribunal, in fresh applications, both as to pleading adequacy (particularly on causation) and as to the appropriateness of opt-in collective proceedings. The judgment does not itself finally determine whether the claims will be struck out, but it confirms that, as presently pleaded and analysed by the Tribunal, they are regarded as weak and unsuitable for opt-out treatment.

More broadly, the decision clarifies several important aspects of the collective proceedings regime:

• Strength of the claims under rule 79(3)(a) is not generally a neutral factor in the opt-in/opt-out choice. Where the Tribunal can form a reasoned, even if provisional, view of the merits, weaker claims normally weigh against opt-out certification, and stronger claims may support it.
• Practicability of opt-in under rule 79(3)(b) is to be assessed objectively, taking account of class composition and realistic litigation behaviour of reasonable claimants, and recognising that the regime offers an opportunity, not a guarantee, of redress. The Tribunal has considerable discretion in this evaluative assessment, and appellate courts should respect that margin.
• Policy aims of vindicating rights and deterring wrongdoing do not create any presumption in favour of opt-out proceedings; access to justice must be balanced against the need to protect defendants from oppressive, unmeritorious or over-leveraged litigation.
• Findings in competition authority decisions against third parties are not admissible as factual evidence against non-addressee defendants in private follow-on actions and do not, in themselves, carry probative weight at trial. At most, such material may be used at an interlocutory stage to indicate potential avenues of evidence, but in this case the later Commission decision did not meaningfully assist the Respondent’s causation case.

The judgment therefore reinforces the Tribunal’s central gatekeeping role in collective competition litigation and delineates the proper boundaries of appellate review and evidential use of regulatory decisions in such proceedings.