Maria José Azar-Baud
International Collective Redress Expert, Paris

Representative actions, an important device for the effectiveness of competition law

Collective redress mechanisms, including class actions and representative proceedings, pursue three interrelated objectives: access to justice, in terms of redress and preserving the effectiveness of such rights; judicial efficiency; and deterrence.

In the field of competition law, these objectives acquire particular salience. First, they enable effective compensation for those who have suffered harm as a result of anti-competitive conduct. Second, they allow courts to manage, in a rational and structured manner, the volume of claims generated by systemic infringements. Third, they reinforce both general and specific deterrence by ensuring that wrongdoers are held accountable at scale and are dissuaded from new wrongdoings, complementing public enforcement by skimming off unlawful gains, through compensation, which is the property of victims. This paper aims to demonstrate that when opt-out collective actions and funding are made available, they remain the superior means of achieving effective compensation for consumers, as opposed to the bundling and assignment of claims. Indeed, despite the risk of parallel actions, they contribute to improving the efficiency of the judiciary.

A structural question arises at the outset. In competition law, public enforcement already produces a sanction: infringements are declared and substantial fines imposed. Yet those fines are paid into the public treasury, and victims receive no direct compensation. Private enforcement, therefore, operates not as a substitute, but as an indispensable complement to public enforcement. Where a regulatory fine has been imposed, collective redress remains necessary to secure compensation for victims. What remains largely unresolved, particularly in the European Union, is the fate of unlawful profits retained beyond the fine, in particular, given the absence in most Member States of a developed doctrine of punitive damages.

From the perspective of access to justice, the central obstacle is rational apathy. Where individual loss is modest, litigation is economically irrational. The harm remains uncompensated not because the relevant rights lack substance, but because procedural design renders their exercise impracticable. Empirical research confirms this structural deficit. A survey of consumer antitrust enforcement across Europe concluded that EU consumers have not, in any meaningful number, been compensated for anti-competitive infringements.^[1] Only jurisdictions combining opt-out procedures with viable litigation funding, most notably the UK, Portugal and the Netherlands, appear capable of altering that landscape. But that has not yet occurred, as litigation is ongoing and no meaningful settlements are in the public domain.

Indeed, the limitations of opt-in mechanisms are well documented. Prior to 2015, England and Wales provided an opt-in collective procedure before the Competition Appeal Tribunal. Only one action was ever brought. Recovery amounted to approximately £21,000 in total. Fewer than 0.1 per cent of eligible claimants participated. The authorised consumer body concluded that opt-in proceedings were not a proportionate use of resources.^[2] The UK responded through legislative reform in 2015, introducing collective proceedings that can proceed on either an opt-in or opt-out basis before the Competition Appeal Tribunal. That has changed the landscape.

Across the EU, there was no example of an opt-in mechanism in the competition field that compensated a significant number of consumers. Where participation requires affirmative action for a small individual recovery, participation does not occur.

Comparative practice confirms the significance of the procedural design of collective actions, and the Apple App Store litigation provides a paradigmatic illustration. In the Netherlands, two opt-out collective actions are pending. In Portugal, Apple’s App Store conduct is the subject of three separate opt-out consumer actions before the Competition, Regulation and Supervision Court. Ius Omnibus filed a popular action in March 2022, representing over 2.9 million consumers, for purchases made since July 2008. Professor Fabrizio Esposito filed a citizen popular action in 2022 against Apple’s Portuguese subsidiary, with the compensation claimed exceeding €100m, limited to Article 101 of the Treaty on the Functioning of the European Union (TFEU). DECO filed a further action in 2025 as a follow-on to the European Commission’s Apple Music App Streaming decision, seeking on average €2.60 per user per monthly payment, on behalf of approximately 36,000 consumers. In October 2025, the Competition, Regulation and Supervision Court deemed a large part of the DECO claim inadmissible for lis pendens, allowing the remainder to proceed on the basis that the defendant, facts or damage differed sufficiently. The first two actions are suspended pending the Commission’s decision becoming final.

Furthermore, in Spain and in France, collective actions were also initiated and a coordinated initiative from Euroconsumers spans multiple Member States. As for the UK, in Kent v Apple, the Competition Appeal Tribunal’s first successful opt-out competition trial awarded £1.18bn in damages (pre-interest) for excessive App Store commissions^[3], a clear contrast with its predecessor’s opt-in procedure.^[4]

These developments also underscore the relationship between public and private enforcement. In March 2024, the European Commission imposed a €1.8bn fine on Apple for the same conduct. Public enforcement declared the infringement had occurred and imposed a sanction; private enforcement followed to secure compensation for victims.^[5]

These developments also highlight the importance of the procedural design in terms of scope. Indeed, a collective action was only brought in Member States in which collective actions are allowed in the competition sector.

In this respect, it is worth highlighting that while the Representative Actions Directive^[6] (RAD) does not include competition law within its scope (it does include Digital Markets Act (DMA) infringements), some Member States had already included competition law within the scope of collective actions (the Netherlands, Portugal and Belgium). Other Member States have also extended collective actions to competition law when transposing the legislation (for example, France, Greece and Slovenia). In Germany, lawmakers have included competition violations within the Consumer Rights Enforcement Act, which has been in force since October 2023, making it the first Member State to integrate DMA enforcement into its domestic private enforcement framework.^[7] An assessment of the German implementation of the mechanism is that the ‘redress action falls short’.^[8] It does not apply to companies, its reach in terms of small- and medium-sized enterprises (SMEs) is limited and whether it will be used by consumers in practice is still an open question. There, the assignment model remains the principal instrument and adopts a business-to-business (B2B) model (serving those whose losses are large enough to attract a litigation funding vehicle). In January 2025, in ASG 2 v Land Nordrhein-Westfalen,^[9] the Court of Justice of the European Union’s Grand Chamber confirmed that a national ban on assigning cartel damages claims to a legal service provider is incompatible with the principle of effectiveness where alternative collective redress mechanisms are not genuinely effective. For many cartel victims, the assignment model remains the only one available, and so the challenge persists.^[10]

Parallel cross-border and domestic competition-related representative actions

The expansion of collective redress mechanisms has generated another structural challenge: the proliferation of parallel proceedings both domestically, that is within a single Member State, and cross-border.

The first arises when a representative action and an individual action concerning the same subject matter proceed simultaneously before courts in the same Member State. The Court of Justice of the European Union had confronted this in Sales Sinúes.^[11] In that case, individual consumers had brought actions before a Spanish court and a consumer association, ADICAE, had already initiated a representative action against the same banking institutions. The Court held that individual and representative actions serve different aims: representative actions are preventive and serve as a deterrent, they may be brought without specific contracts and they do not engage individual circumstances. A rule automatically binding individual actions to the outcome of representative proceedings deprives the consumer of effective judicial protection and is incompatible with the principle of effectiveness.

Due to their procedural autonomy, Member States can handle parallel actions differently. Portugal for instance has taken the position that parallel proceedings entail lis pendens unless there are different parties, facts or damages.

Cross-border parallelism presents greater difficulties. Since June 2023, more than 134 representative actions have been recorded across multiple Member States. Technology companies such as Meta, Google and Apple face simultaneous proceedings in several jurisdictions. The RAD expressly states that the Directive exists without prejudice of private international law; jurisdiction, therefore, falls to the Brussels I bis Regulation. Article 29 provides a lis pendens rule: where the same cause of action between the same parties is pending before courts of different Member States, all courts other than the court first seised must stay and ultimately decline jurisdiction. The difficulty is that under the Directive, consumers are not considered to be ‘formal’ parties to the proceedings (but material ones). Only the Qualified Entity acts as a formal claimant, as confirmed explicitly in Recital 36 and Article 7(6). For example, Article 12 of the RAD clearly expresses the ‘substitutive’ nature of representative actions, whereby consumers shall not pay the costs of the proceedings. This provision serves as a direct reflection of their non-party status.

Aside from the inefficiency faced by judiciaries, the risk of duplicative recovery in opt-out systems, even if low in practice, is not merely academic. The more principled reading of Article 29, consistent with Drouot,^[12] and our previous works, is that represented consumers must be treated as material parties, since the outcome directly affects their legal position. Where Article 29 cannot be engaged, Article 30 of the Brussels I bis Regulation offers a secondary instrument, permitting a stay on the basis of connexity alone. Legislative clarification is needed on whether the identity of parties refers to the formal plaintiff or the material consumer group.^[13] In a related matter, in December 2025, the Grand Chamber ruled in Case C‑34/24 that any competent national court has territorial jurisdiction to hear claims in respect of all users of a nationally targeted digital platform.^[14]

Lastly, case management challenges follow from these structural gaps. Where collective mechanisms are unavailable or ineffective, alternative aggregation devices emerge, such as assignment models and bundling.

The Spanish trucks cartel illustrates the systemic consequences of this situation. Tens of thousands of appeals accumulated before the Spanish Supreme Court. Substantive questions had already been resolved; what remained was repetition at scale. In January 2026, the Court referred the pending appeals to mediation, invoking the sustainability of its constitutional function.^[15] Mediation, thus, filled a procedural lacuna that a structured collective mechanism might have addressed more coherently. Spain remains one of the few Member States yet to transpose the RAD.

Conclusion

The broader conclusion is clear. Where effective opt-out mechanisms exist, compensation becomes feasible at scale. Where they do not, rational apathy prevails, and alternative aggregation devices put a strain judicial resources. Collective redress in competition law is, therefore, not an optional adjunct but a structural complement to public enforcement. It delivers access to redress where individual litigation is economically irrational, it enhances efficiency where fragmentation would otherwise prevail and it reinforces deterrence by ensuring that unlawful conduct does not remain economically advantageous.

Firstly, opt-out collective proceedings ensure comprehensive inclusion of affected individuals, thereby overcoming the structural under-enforcement that characterises mass harm scenarios. By dispensing with the need for individual claimants to take active steps to participate, they capture the full scope of the harm, including that suffered by passive, uninformed or economically disincentivised victims. In contrast, bundling or assignment models depend upon the voluntary mobilisation of claimants, which inevitably results in partial aggregation and leaves a significant proportion of harm unremedied.

Secondly, opt-out mechanisms promote procedural economy and judicial efficiency. A single set of proceedings enables the court to determine common issues in a unified manner, avoiding the duplication of evidence, inconsistent findings and fragmented litigation strategies that often arise in claim-by-claim aggregation or through multiple assignments. This centralisation enhances the manageability of complex litigation and reduces the burden on judicial systems.

Thirdly, opt-out collective actions reinforce equality of arms and deterrence. By consolidating claims into a single, representative procedure, they enable claimants to match the economic and organisational capacity of well-resourced defendants. This aggregation of claims increases the overall quantum at stake, thereby strengthening the deterrent effect of litigation and incentivising compliance with substantive legal norms. By contrast, assignment structures, often driven by commercial intermediaries, may prioritise portfolio selection and profitability over the comprehensive vindication of rights.

Fourthly, opt-out regimes offer a higher degree of procedural legitimacy and safeguards. They are typically embedded within a structured judicial framework, involving certification or admissibility controls, court supervision of settlements and scrutiny of representation and funding arrangements. This contrasts with assignment-based models, which may operate through private law mechanisms with more limited judicial oversight, raising concerns about transparency, conflicts of interest and the equitable distribution of redress.

Finally, opt-out collective actions are particularly suited to cross-border and systemic infringements, where the ubiquity of harm calls for a mechanism capable of addressing it in a coherent and unified manner. They, thus, align more closely with the objectives of modern collective redress systems, notably those underpinning EU developments, by ensuring both effective enforcement and the orderly administration of justice.

For these reasons, opt-out collective proceedings are widely regarded as a superior procedural vehicle for addressing mass harm when compared to the bundling of claims or assignment-based aggregation.