Access to collective redress: divergent gatekeeping standards
U.S. courts have reinforced that a lack of commonality and predominance of an issue between class members can be a major barrier to class certification. In both In re Keurig Green Mountain Single Serve Coffee Antitrust Litig. and Miami Products & Chemical Co. v. Olin Corp., federal courts in New York held that a record showing that plaintiffs and defendants engaged in individual price negotiations meant that individual issues predominated over common ones for the proposed class.
Similarly, in Mr. Dee’s Inc. v. Inmar, Inc., the Fourth Circuit affirmed a lower court’s refusal to certify a class of manufacturers who allegedly overpaid for coupon processing services due to defendants’ alleged price-fixing conspiracy. The Fourth Circuit panel found that plaintiffs' defective proposed classes were impermissible in that they both excluded alleged victims of the same alleged Sherman Act violation and included proposed members who had suffered no injury.
The Mr. Dee’s case also deepened a circuit split as to whether a class can be certified that contains uninjured class members. The Fourth Circuit hinted it may apply the more permissive approach (taken by the Seventh, Ninth, and Eleventh Circuits) and permit class certification even when some portion of the proposed class may not have suffered injury, provided that the presence of such members does not overwhelm the common issues or render the class unmanageable.
Nevertheless, the Fourth Circuit denied class certification as, regardless of whether a class can be certified containing uninjured class members, “the presence of 32% of uninjured members in a proposed class strikes us as much too high.” These class certification denials emphasize the difficulty of meeting the predominance requirement in markets with individualized and complex pricing.
Statutes of limitation also remain a key battleground in the U.S. In Scharpf v. General Dynamics Corp., the Fourth Circuit held that allegations of a “non‑ink‑to‑paper” no‑poach agreement could constitute active concealment sufficient to suspend the four‑year Sherman Act statute of limitations. By contrast in 2025, former NCAA athletes’ claims were held time‑barred in Chalmers v. NCAA and Pryor v. NCAA, where courts found that claims accrued long before the applicable limitations period when the college athletes signed contracts giving the NCAA the right to use their names, images, and likenesses, and later promotional uses were merely manifestations of those contracts and did not give rise to new claims.
Further, in Litovich v. Bank of America, the U.S. District Court for the Southern District of New York likewise dismissed an attempt to revive an antitrust class action alleging major U.S. financial institutions organized a group boycott of rival bond-trading companies as time barred, finding that the plaintiffs did not allege any overt acts that occurred within the limitations period, even though plaintiffs claimed that they engaged in bond trading during that period and that there was no fraudulent concealment as the plaintiffs’ claims were predicated on publicly-sourced articles dating back to 2000.
In the UK, the Competition Appeal Tribunal (CAT) is the “gatekeeper” of the class actions regime. It applies a two-part test for certification; broadly, whether the claim is suitable for class action proceedings and whether the class representative would act fairly and adequately in the class’s best interests. Although competition class actions were introduced in 2015, they made little progress before a claimant-friendly decision of the UK Supreme Court in 2020 which confirmed the low threshold test for certification and paved the way for many high-value claims. Nonetheless, the CAT will refuse certification where warranted (and refused three claims certification in 2025 for a range of reasons).
Importantly, at certification the CAT will determine whether certification is to be awarded on an opt-out basis—meaning that UK persons meeting the class definition are automatically part of the claim—or only an opt-in basis, where a person is part of the claim only on registering by a deadline. Opt-out claims produce bigger and higher-value claims, which (amongst other features) make them more appealing to litigation funders and claimant lawyers.
However, the UK Supreme Court recently upheld a decision of the CAT to refuse to certify the FX class action on an opt-out basis. The court found that the CAT was entitled to conclude that opt-out certification was not justified, including because the claim was inherently weak and the evidence suggested that it would be practicable for potential class members (sophisticated businesses with sizeable claims) to opt in.
The court emphasized that a balance is to be struck between assisting claimants to obtain redress via the opt-out mechanism, whilst avoiding defendants facing massive but unmeritorious opt-out claims (which they are driven to settle because of their scale notwithstanding their lack of merits). The decision is likely to lead defendants to increasingly challenge opt-out certification, particularly where the proposed class comprises sophisticated people with large claims (for whom opting in might be practicable) or where there are compelling arguments at the outset that the claim is inherently unlikely to succeed.
In the EU, there is no single certification test for collective redress. Member states apply their own procedures to representative actions (or even claims bundling), constrained only by the EU principles of effectiveness and equivalence. Opt‑in models remain the norm (e.g., France, Germany, Italy), while opt‑out systems (e.g., Portugal) or hybrid regimes (e.g., the Netherlands and Belgium, although the latter uses opt‑in as the default rule) make those for a attractive for claimant groups.
In Australia, class actions proceed on an opt-out basis by default. Anyone who meets the group definition is included unless they opt out by the court-set date. This promotes access to justice but can make it difficult for defendants to estimate their potential exposure because final group size is uncertain. To manage settlement risk around group size and quantum, courts often make “soft class closure” orders, which require group members to register by a set date if they wish to share in any settlement, thereby improving certainty without formally changing the proceeding from opt-out to opt-in. In 2025, the High Court, in Lendlease v. Pallas, confirmed that courts have power to make such orders.
Across the four jurisdictions, access to collective redress is expanding, but the gatekeeping rules differ sharply. These differences make forum choice strategic, and they require claims to be tailored to each system’s specific certification logic.
Managing procedural complexity: experts, evidence and modelling
Across all four jurisdictions, managing procedural complexity has become central to antitrust damages actions, covering expert evidence, economic modelling, data issues, and judicial case management. Although all systems rely heavily on expert input, they differ significantly in how courts engage with experts, the evidentiary burdens they place on claimants and defendants, and the extent to which economic models influence both certification and ultimate liability.
In the U.S., expert testimony continues to be central to antitrust class actions, with parties almost always retaining their own expert. However, 2025 saw a rare invocation of judicial power to appoint a court expert. In the Epic v. Google litigation, the Northern District of California appointed a Massachusetts Institute of Technology economist under the Federal Rules of Evidence to independently evaluate the competitive and economic implications of proposed modifications by the parties to an injunction related to the Google Play Store. The expert concluded that the parties’ proposed adjustments would fall short of the competitive remedies in the original injunction and could produce lower developer fees than those contemplated by the parties’ proffered agreement.
In the UK, experts are fundamental to the determination of antitrust damages claims in the CAT at the certification and trial stage. While parties typically nominate their experts and the issues to be addressed, the CAT may seek to understand at an early stage the respective methodologies on which those experts intend to rely (indeed, whether a class representative has a sufficiently credible or plausible methodology is part of the test for certification) and, where there is overlap, may require parties to jointly instruct an expert (in particular across parties whose positions are broadly similar).
In recent years experts have taken an increasingly pivotal role in case management within the CAT and have been granted a wide berth to identify the material required for their analyses and influence decisions around the scope of evidence required. However, that tide may be turning with recently renewed guidance from the CAT seeking to rein in the scale and proliferation of expert evidence in antitrust damages actions and to reinforce the positive obligations placed on experts with respect to their independence and duty to engage constructively with each other rather than act as advocates for the person who instructed them.
The EU shows the greatest procedural variation in how courts handle experts, reflecting distinct judicial cultures, resources, and approaches to economic evidence. In some jurisdictions, such as France and the Netherlands, courts regularly engage with party-appointed experts, and expert evidence forms an integral part of the adversarial process. In others, including Belgium, Denmark, Germany, Hungary and Italy, courts often rely on court-appointed experts to produce an independent analysis, which can carry significant persuasive weight, sometimes serving as the primary basis for judicial findings.
By contrast, Spanish courts cannot appoint an expert on their own motion; they may do so only at the request of a party and, as a general rule, only where that party has not already submitted its own expert report on the same subject. In antitrust damages actions, however, parties almost invariably file expert reports, meaning that court-appointed expertise is rarely available. Spanish courts sometimes find the parties’ expert reports unreliable or overly complex and, therefore, turn to judicial estimation. This approach carries the risk of producing outcomes that may appear arbitrary or disproportionate.
This fragmentation matters in practice: in cross‑border antitrust damages actions, expert strategies must, for example, adapt to different evidentiary expectations, varying judicial comfort with economic analysis, and inconsistent approaches to assessing pass‑on.
Despite these procedural differences, parties across the EU routinely submit their own expert reports. These typically serve as the foundation of the economic debate, reflecting the growing importance of robust economic evidence as claims become increasingly complex.
In Australia, while courts rely heavily on expert evidence in antitrust damages claims, judicial caution has emerged regarding disproportionate expert costs. Judges have warned against expensive “materiality” evidence on matters of common sense, which can escalate into an unnecessary and costly expert battle with little value to group members (Parkin v Boral Limited). Similarly, courts may prefer a court-appointed referee over a firm-retained assessor to test legal costs independently (Yasmin v Commonwealth of Australia (No 2)). This reflects a broader concern that expert proliferation should not outweigh the practical benefits to claimants.
In summary, approaches differ—from the U.S.’s rigorous scrutiny to the UK’s expert‑heavy model and the EU’s fragmented practices—but everywhere the direction of travel is the same: robust economic evidence now drives outcomes in antitrust collective actions.
Distribution, settlement design and funding
As antitrust class actions grow in size and complexity, the practical tasks of verifying claims, allocating compensation, approving settlements, and structuring funding have become central to how effective and credible collective redress systems are. Courts in all major jurisdictions now examine not just the headline value of settlements, but also the economics of participation, administration, and funding—though each system applies this scrutiny through its own institutional and procedural lens.
In the U.S., scrutiny of class action settlements continues to intensify. In the long‑running Payment Card Interchange Fee and Merchant Discount litigation, several large retailers—including Walmart—objected to the 2025 proposed settlement with Defendants, arguing that the deal failed to treat large merchants equitably and offered relief that would not benefit a significant portion of the class. Objectors urged the court to reject the settlement and decertify the class, carve out large national merchants from the settlement, or permit opt‑outs.
Also in 2025, whilst a federal court in Caccuri v. Sony Interactive Entertainment LLC, denied preliminary approval of a proposed class settlement for technical deficiencies, it made it clear its skepticism about the proposed settlement’s structure, in which the class would receive credits rather than cash. The court directed the parties to explain the value and distribution of such credits in any revised motion. While the court did not hold that non‑cash relief was impermissible, its concerns reflect a judicial trend toward insisting on transparent valuation of class recovery.
In the UK, the 2023 Supreme Court decision in PACCAR, which rendered common forms of damages-based litigation funding unenforceable in many circumstances, led to significant uncertainty for claimants and litigation funders (and the adoption of alternative funding models where existing models were held to be unenforceable). However, in 2025 the UK government confirmed that it will introduce legislation to reverse PACCAR. It has also accepted a recommendation from a civil justice advisory body to implement light-touch regulation of litigation funders. The content of the proposed legislation has not been confirmed, but these announcements indicate broad support for the continued growth of commercial litigation funding in the UK (and more generally support for private enforcement of antitrust law).
A number of collective settlements have now been approved, with the decisions demonstrating that the overarching consideration for the CAT is whether the settlement is just and reasonable for the class (not for the lawyers or funders involved). However, the CAT recognizes the role played by funders in the success of the regime, and that funders should achieve commercial returns provided these fairly reflect the success (or failure) of the litigation.
The 2025 approval of a USD263.7 million settlement in the Merricks Interchange Litigation provided the first detailed picture of how high-value, opt-out collective settlements will be evaluated in the UK. The CAT approved the settlement despite it representing a massive discount against the original claim value (at one point estimated at USD21.1 billion). The CAT recognized—on the basis of the evidence from both sides—that the figure reasonably reflected the prospects of success at that stage of the litigation. Notably, the settlement withstood strong opposition from the funder, who contested the overall settlement amount and its share. The CAT emphasized that a funder’s return should reflect the outcome of the case.
In 2025 concerns were raised about the effectiveness of the distribution process for class actions in the UK after it emerged that in the Gutmann litigation (described below) less than 1% of a settlement fund (resulting from a settlement between some but not all of the parties) had been claimed by class members. This outcome lent support to commentators who argue that class actions principally serve the interests of lawyers and funders. We anticipate that the CAT will now place greater emphasis on the need for class representatives to show at the outset of the litigation that they have credible, evidence-based plans for distribution (although the UK Supreme Court has previously ruled that consideration of distribution proposals at the certification stage will generally—but not always—be premature).
It is also likely that this “distribution problem” will be an area of focus for the government as part of its ongoing review of the opt-out class action regime (also announced in 2025). The review focuses on whether the right balance is being struck between ensuring consumer protection through class actions and limiting the burden of claims on business. The results of the consultation are expected during 2026. In our view, the mix of claimant- and defendant-friendly decisions in 2025 (set out below) will be seen as a sign that the regime is broadly working and should be allowed to continue to evolve through judicial supervision, albeit perhaps with some tweaks (for example, in relation to distribution).
The EU shows the widest variation in settlement design because nearly all aspects of distributing damages—from claim verification to funding—remain matters of national procedural autonomy. This leads to markedly different approaches, including to identifying eligible claimants: Germany often requires individual proof of purchase, unless simplified evidentiary standards are agreed in a settlement structure, while Portugal’s opt‑out system eases the burden of proof on individuals but lacks established claim verification tools, forcing courts to craft bespoke solutions.
Distribution models diverge just as sharply. France favors a court‑controlled scheme with detailed judicial oversight over the parameters for compensation, whereas the Netherlands allows claimant—or funder‑managed structures under the WCAM/WAMCA framework, subject to reasonableness and transparency.
Funding rules are similarly inconsistent: Dutch courts accept success‑fee arrangements, if not excessive; Swedish courts take a more cautious, interventionist stance, scrutinizing fee structures and funder influence (conflicts of interest); and, in Portugal, continued uncertainty over the legality of third‑party funding complicates enforceability of funding arrangements. Residual balances add another layer of divergence, with France directing unclaimed amounts to the State Treasury, and Spain giving courts broad discretion as regards distribution.
Settlement approval is equally uneven. The Representative Actions Directive introduces only basic safeguards for cross‑border settlements and leaves substantive fairness or adequacy criteria to national law. The Netherlands stands out with the EU’s most developed statutory settlement approval regime and, to date, generally permissive approach focused on reasonableness and transparent distribution of the settlement amount. Elsewhere, practices vary widely; for example, Italian courts may even propose settlements and related terms to ease caseloads. These differences materially shape the net compensation ultimately reaching claimants, lead to uncertainty in settlement negotiations, and drive forum choice.
Third-party funders remain central to the viability of many class actions in Australia. Funding has grown since the High Court confirmed in Campbells Cash and Carry Pty Limited v Fostif Pty Ltd that litigation funding is not contrary to public policy or an abuse of process. However, litigation funders may become more selective following the decision in Davis v Wilson, which confirmed that litigation funders may be subject to adverse costs orders. There have been calls to cap the proportion of returns payable to funders and plaintiff firms, including a proposal to guarantee at least 70% of gross recoveries to class members, although a bill to create a presumption against distributions above 30% to non-group members lapsed in 2022 and has not been revived.
Despite different national models, courts everywhere now require clear, credible and transparent settlement and funding structures. Robust, data‑driven design is essential to obtain approval and ensure compensation reaches claimants.
Substantive theories of harm
Private enforcement increasingly focuses on harms linked to digital market power—such as self‑preferencing, tying, restrictive access terms, and excessive commissions—and, more recently in the U.S., on claims that shared algorithms or data systems may support coordinated outcomes. The legal labels differ (abuse of dominance in the UK/EU, monopolization and restraints in the U.S., misuse of market power in Australia), but the economic analysis largely converges on defining the market, assessing foreclosure, and establishing a credible counterfactual.
Abuse of dominance
Big Tech remains a major focus of U.S. private antitrust litigation. Similar to courts in other jurisdictions, U.S. federal courts addressed a case brought by Epic Games against Google in 2025. In Epic Games v. Google, the Ninth Circuit affirmed a 2023 jury verdict that Google monopolized Android app distribution and billing through the Play Store. The court upheld an injunction requiring Google to allow rivals access to the Play Store and prohibiting exclusive distribution agreements with app developers.
Google’s argument that the adverse verdict was precluded by Epic’s unsuccessful antitrust claims against Apple was rejected. The Ninth Circuit held that technological differences and distinct theories of harm rendered the cases materially different.
In the EU, private damages claims arising from alleged abuses of dominance are gaining momentum, fuelled by intensified public enforcement activity—particularly in digital markets—where the European Commission (EC) and national antitrust authorities have pursued high profile investigations into major platforms and Big Tech companies, such as Google and Microsoft.
These decisions have already prompted follow-on actions, notably in Germany after the EC’s Google Shopping decision, where comparison-shopping services allege harm derived from foreclosure caused by Google’s self-preferencing and discriminatory display practices. Stand-alone abuse actions are also increasing, as illustrated by representative actions in the Netherlands against Apple and Google, alleging excessive (30%) commissions on apps and in app purchases.
Collectively, these developments signal a clear and accelerating upward trend in EU private enforcement for abuse of dominance, transforming an area that historically lagged far behind cartel damages litigation. However, these cases can be expected to be procedurally and evidentially demanding, relying on complex theories of harm—typically supported by economic evidence—and requiring courts to assess and weigh competing expert analyses, which they may well not be adequately equipped to do.
Likewise in the UK, claimants’ focus has been on abuse of dominance, including claims against the likes of Google, Microsoft and Apple that are analogous to claims brought in other jurisdictions such as the EU. Of the claims filed or certified in the last year, only one does not involve an abuse of dominance allegation.
Similarly, most claims in the UK are brought on a standalone basis. In some instances, claimants have relied on some form of investigation or review by an antitrust authority or a regulator. However, these are not binding on the CAT (unlike decisions in strict follow-on claims) and the CAT’s judgments in Le Patourel and Kent (described below) illustrate that the CAT will form an independent view and (as in Le Patourel) may ascribe little weight to prior non-binding assessments by public bodies.
As noted above, these claims are evidentially dense and complex. Across the first three class action judgments delivered by the CAT—Le Patourel v BT (telecoms pricing), Gutmann v First MTR South Western (train fares) and Kent v Apple (AppStore terms and commission rates)—there have been both successful and unsuccessful outcomes for claimants. These judgments have demonstrated that abuse of dominance’s subjective elements can make the end result of these types of claims harder to predict.
All three cases alleged abuse of dominance based on excessive and unfair pricing and/or exclusionary abuses and were “standalone” claims. Le Patourel and Gutmann failed because the class representative was unable to establish any anticompetitive abuse. Both cases turned on the CAT’s evaluation of whether the facts were sufficient to cross the line of an abuse.
The Le Patourel case was appealed unsuccessfully, with the Court of Appeal emphasizing strongly that it will be reluctant to interfere with the CAT’s evaluation of the facts (reinforcing the CAT’s “gatekeeper” role). In contrast, Kent was successful, with the CAT finding both an exclusionary abuse, and excess and unfair pricing. Again, Kent turned principally on the CAT’s assessment of the facts, both in how it defined the market (leading to a finding that Apple had a 100% market share) and its views about whether the practices were abusive. Subject to any appeal, Apple may be required to pay damages exceeding USD1.3bn.
These mixed results mean that abuse of dominance claims are likely to remain attractive to claimant lawyers and litigation funders, but equally, they demonstrate to defendants that it may be worth fighting such allegations to trial rather than agreeing substantial settlements to resolve them early.
In Australia, since a 2017 reform which changed the misuse of market power test from a solely “purpose-based” test to a broader “effects-based” test, now prohibiting corporations with substantial market power from engaging in conduct that has the purpose, effect, or likely effect of substantially lessening competition, there has been an increase in private misuse of market power claims. By late 2025, the overwhelming majority of such cases had been brought by private parties, while regulatory enforcement has been comparatively modest.
Private antitrust class actions continue to be filed in Australia. Sony is facing a class action alleging misuse of market power and exclusive dealing in relation to the PlayStation ecosystem. Apple and Google also face class actions in which liability findings have been made and relief is to be determined. In 2025, the Federal Court delivered major judgments in proceedings brought by Epic Games against Apple and Google, together with follow-on class actions by app developers. These were among the first Australian cases to apply the post-2017 misuse of market power test in digital platform markets.
Epic alleged Apple engaged in misuse of market power by forcing app developers to use only Apple’s App Store to distribute their apps to iOS device users and tying app distribution to the exclusive use of Apple’s payment platform for in-app purchases. The court found that Apple’s conduct substantially lessened competition in the iOS app distribution and in-app payment markets, and was a misuse of market power. The court accepted Epic’s case on both purpose and effect in app distribution, and on effect (but not purpose) for in-app payments. Apple’s security justification was not accepted. In determining Epic Game’s allegations against Google, the court recognized Android’s greater openness but found contraventions relating to app distribution and payments. Claims of exclusive dealing, anticompetitive arrangements and unconscionable conduct were rejected in both cases. In the related class actions, the court found for the applicants. The cases are now before the court for determination of relief.
Algorithmic pricing
In the U.S., AI‑driven pricing is becoming both a legislative and litigation priority. In 2025, U.S. states enacted a variety of laws targeting the use of AI and algorithms to set prices, including in California where amendments to the Cartwright Act aimed directly at algorithms used to set prices were enacted, effective January 1, 2026. Other states enacted their own frameworks—sometimes inconsistent with each other—creating a complex compliance landscape. No analogous federal law has yet been enacted, although proposals have been introduced.
Courts continue to hear cases involving alleged algorithm‑facilitated collusion. In In re MultiPlan Health Insurance Provider Litig., plaintiffs sufficiently alleged healthcare insurers’ use of algorithms to provide recommendations for certain provider payments constituted an “agreement” for purposes of Section 1 of the Sherman Act. The court found persuasive plaintiffs’ allegations that defendant insurers delegated rate decisions to the same third-party rate negotiation company that allegedly acted as a go-between to facilitate competitor communications of sensitive data. Conversely, in Gibson v. Cendyn Group, LLC, the Ninth Circuit upheld dismissal where plaintiffs failed to allege an overarching agreement among Las Vegas hotels to collude via shared confidential data.
Private enforcement is shifting sharply toward digital‑market conduct, where platform control, design choices, and pricing structures drive alleged harm. Across jurisdictions, success increasingly depends on clear market definition and robust economic evidence of foreclosure and pricing effects, while weak facts or uneven expert analysis remain the main causes of failure.
