Regulatory Norms in EU Competition Law Assessment

I. Introduction

What roles do regulatory norms play in EU competition law assessment? The question matters both for the consistency of the supervisory framework for economic activity in the EU, and because it helps us to better understand the legitimate concern of the competition rules, a key theme in contemporary competition policy debates. Competition law (also known as antitrust) aims to promote and protect effective competitive behaviour by economic actors—the jurisprudence uses the term ‘undistorted’^[1] competition to describe the ultimate policy objective. Regulation, a broader and more ambiguous concept,^[2] often applies to and constrains the same market conduct, though typically in pursuit of alternative or additional policy objectives. Regulation can enhance opportunities for effective competition or limit the scope for abusive market behaviour; but it can also ‘distort’^[3] competition in ways that raise concerns from a competition policy perspective, whether by diminishing overall competitive dynamics or by facilitating or incentivising anticompetitive conduct.^[4] EU law has long endorsed the concurrent application of the competition rules to regulated firm behaviour, an approach that maximises the effectiveness of competition policy. Yet the jurisprudence struggles with a problem of internal coherence, seeking to protect the distinctiveness and primacy of the competition rules while also finding space for the increasing centrality of contextual analysis in contemporary competition enforcement. Moreover, applying competition law to behaviour that has already been affected by other regulatory norms creates its own challenges, including issues of legitimacy, consistency and proportionality. These questions are the focus of this piece.

In principle, the competition rules apply only to the autonomous behaviour of firms, and such conduct—whether taking the form of agreements, dominant firm practices or mergers—is prohibited only where it is demonstrated that anticompetitive effects—whether actual, potential, or presumed—are likely to result. Yet where competition law is applied to regulated behaviour, there is a higher likelihood that suspect conduct has been influenced or even directed by the regulatory obligations and that any resulting market distortions can be attributed, wholly or partly, to the impact of the regulation. In such contexts, competition law assessment must consider not only whether and how the defendant’s behaviour has caused anticompetitive harm, but also the extent to which that behaviour and its effects were determined by the underlying regulatory framework. The greater the overlap between the scope of the regulatory obligations and the competition law theory of harm, the more complex this task becomes. Since competition law and regulation must co-exist as legitimate and socially-valuable mechanisms of market control, there are arguments—both practical and legal^[5]—for ensuring that these separate regimes are interpreted and applied in a consistent, ideally mutually reinforcing, manner. But because this question typically arises where competition law is applied in markets already subject to ex ante regulation, there can be tension between preserving the effectiveness of competition law and its underlying goal of protecting undistorted competition, while simultaneously respecting other public interest values pursued by ‘distortive’ regulation.

A great deal has already been written about the relationship between competition law and other regulation, including much of my own work. This article seeks to make an original contribution by focusing on how substantive regulatory norms are accommodated within competition law analysis and what this tells us about competition law’s place within the wider market regulatory sphere. We leave aside the distinct questions of when competition law itself exhibits certain ‘regulatory’ characteristics,^[6] and of the appropriate competition policy response to anti-competitive State regulation.^[7] In doing so, the article adopts an essentially nonnormative, conceptual approach, which aims to work effectively within the confirms of the existing jurisprudence rather than making more aspirational claims for wholesale change.^[8] Our analysis is not directly concerned with the goals of competition law, although there are plenty of articles of this sort about.^[9] But when considering how competition law interacts with other regulatory fields that pursue a more diverse range of policy objectives, it is impossible to avoid the question of what competition law itself seeks to achieve. Above, we described the goal of competition law broadly as the protection of ‘undistorted’ competition, an approach that, although not without its critics, finds support in recent jurisprudence and legislation.^[10] The article thus seeks to provide a more granular exploration of what we mean by undistorted competition, in a context where markets can never really be free of distortions, whether stemming from private or public sources. By considering the extent to which other regulatory norms can provide a conduit by which to reflect a wider range of public interest values within competition enforcement, while also recognising the legal and practical limits on this approach, we seek to obtain a better understanding of what Odudu termed ‘the legitimate scope’^[11] of competition law.

The article is structured as follows. Section II introduces and distinguishes competition law from other regulatory instruments of market supervision. Section III sets out and analyses the ‘orthodox’ understanding of the inter-relationship, which prioritises the independence and effectiveness of competition law as a mechanism of market control. Section IV discusses the more ‘contextual’ contemporary approach, considering the ways in which regulation may provide a relevant element of context within competition law analysis. Section V steps back, to consider the normatively-oriented question of how the ‘outsourcing’ of liability standards in competition law to other regulatory fields can be explained. It examines a variety of perspectives: the effectiveness of competition law as a regulatory enforcement mechanism; regulatory norms as a constraint on the acceptable parameters of competition on the merits; and regulatory (non-)compliance as a positive anticompetitive action. Section VI brings these disparate strands of analysis together and concludes.

II. Distinguishing competition law from ‘regulation’ as instruments of market supervision

A necessary starting-point is to distinguish competition law from other forms of market regulation. Competition law comprises a set of legal rules that aim to ensure open, undistorted and fair competition in the internal market, by controlling the acquisition and exercise of market power (as distinct, inter alia, from public power^[12]). Within the framework of the EU treaties, the competition rules operate as important flanking provisions to secure the objective of ‘a highly competitive social market economy’.^[13] Yet most competition cases do not belabour this point. Instead, enforcement tends to be a relatively technical exercise aimed at identifying and proscribing various agreements (under Article 101 of the Treaty on the Functioning of the European Union, or TFEU), practices of dominant undertakings (under Article 102 TFEU) and mergers (under the EU Merger Control Regulation, or EUMR)^[14] that are deemed restrictive of competition in some legally relevant sense. A key theme within recent case-law is the notion of ‘competition on the merits’^[15]: although competition law is aimed at anticompetitive firm behaviour, the rules do not seek to prevent firms, even where they are dominant or acting in concert, from competing ‘on the merits’ (what is also termed ‘normal competition’^[16]). What amounts to normal or meritorious competition in an individual case is a question of competition law. It is thus determined by applying the legal tests and principles developed in the competition jurisprudence,^[17] instead of deferring to the behavioural or regulatory norms within a sector.^[18]

Contemporary EU competition law has several salient aspects for our purposes. First, competition assessment is, in almost every instance,^[19] a highly context-specific exercise: conduct is held to restrict competition within its market circumstances,^[20] while also taking account of any countervailing pro-competitive efficiencies that might be generated.^[21] Second, and somewhat by contrast, competition law is largely sector-indifferent in its application. With a few exceptions,^[22] the competition rules apply across all sectors, supervising vastly different industry practices under the same legal provisions. Finally, the competition rules governing anticompetitive agreements and dominant firm practices are largely ex post in application, serving to punish infringing behaviour after-the-fact.^[23] The EUMR, conversely, is prospective in analysis,^[24] but the object of the exercise is not (generally^[25]) to regulate the behaviour of newly-merged firms going forward, but rather to catch and proscribe potentially harmful concentrations before these occur. In this article, we focus on the impact of pre-existing regulation on the application of Articles 101 and 102 TFEU to market behaviour, while considering insights from merger control where appropriate.

The term regulation is less precise in its meaning. It can be defined as broadly as ‘all forms of economic or social influence’ that may stem from public or private sources, or as a narrowly as ‘a specific set of commands’ imposed by the state and pertaining to closely identified activities.^[26] For the purposes of this article, we adopt a relatively loose definition, referring to state-imposed (or at least state-endorsed^[27]) norms that, directly or indirectly, control or constrain market activity by economic actors, broadly understood. This encompasses both sector-specific regulation, which applies to market participants that are active only within particular sectors (such as telecommunications, energy, transport or digital platform services), and to regulatory regimes of more general application (such as data protection, health and safety or environmental law).

Applying this definition, it can readily be seen that most markets today are subject to multiple layers of regulation that, to borrow the language of the competition jurisprudence, may ‘distort’ competitive outcomes to a greater or lesser extent. In some instances, it is obvious how overlap with competition law arises: for example, regulation may grant an economic actor special or exclusive rights that give it a position of market dominance, or regulation may mandate pro-competitive behaviour (such as imposing access requirements) or proscribe direct abuses of market power (through, for example, pricing controls). In other instances, the potential for interaction with competition law arises more obliquely, for example where regulation constrains a firm’s ability to maximise its profits or provides opportunities for exploitative or exclusionary behaviour.

Given the breadth of our understanding of regulation, and by implication the variety of regulatory models that it encompasses, it is impossible to neatly distinguish competition law from other forms of market regulation that may apply to same conduct.^[28] Two distinctions of relevance might, however, be advanced. First, although the precise ‘goals’ of EU competition law are debated, broadly speaking the rules aim at a straightforward objective, namely ‘the protection of undistorted competition’.^[29] While a regulatory regime may seek to enhance effective competition, this need not be its only policy objective, and most of the examples of regulation that overlap with competition law within the case-law have pursued other regulatory goals additionally or alternatively,^[30] some of which may sit uncomfortably with the efficiency-focused perspective of the competition rules. Second, one of the necessary trade-offs that comes from the wide applicability of competition law is that its prohibitions are sketched in broad and abstract terms. This creates flexibility and scope for progressive development of the law to address emerging market problems, but such advantages arise at the expense of ex ante precision and certainty. Some regulatory regimes operate in a similar manner, yet regulation can be more prescriptive, identifying with greater exactness the specific regulated actors or the content of mandatory norms. Specification can assist enforcement, enabling the supervision of complex market problems. Yet where regulated entities are eager to resist regulation, a precisely defined and limited set of rules may facilitate strategic behaviour to escape or exploit regulatory controls, a problem considered in Section V.

The broad scope of the competition rules coupled with the pervasiveness of regulation in modern markets leads to overlapping jurisdiction and potential conflict. The jurisprudence demarcating the inter-relationship of competition law and other regulation within the EU legal system is complex. One strand of case law emphasises the independence and effectiveness of competition law as a mechanism of market control, reflecting an ‘orthodox’ understanding of the inter-relationship that prioritises but also effectively siloes competition law. Another strand embraces the more contextual approach that has come to the fore in the general competition case-law, recognising the extent to which regulation may feed into and even determine the substance of competition law assessment. While these jurisprudential currents are not inherently at odds, they reflect notably different visions of this inter-relationship. There are also points of inconsistency and even conflict within the case law, which remain underexplored in the existing jurisprudence. In what follows, we set out first the apparently resolute ‘orthodox’ position (section III) and then the more nuanced contextual approach (section IV). Doing so identifies several cases in which it becomes more difficult to reconcile the orthodox and contextual accounts, which are explored in section V.

III. The Interaction of Competition Law and Regulation: the Orthodox View

Within EU law, the orthodox position is that the competition rules are essentially distinct from and unaffected by other regulatory norms that may apply to the same market behaviour. This approach is exemplified by Deutsche Telekom, in which the defendant was held to have breached Article 102 in circumstances where it had complied with its sector-specific regulatory obligations, where the sector-regulator had either set or endorsed the pricing practices that comprised the abuse, and where the national regulatory framework pursued a legitimate public policy objective (social solidarity) by endorsing a pricing structure that the Commission subsequently held to be inefficient.^[31] An analogy of ‘two barriers’ to lawful market participation was suggested by the Advocate General in Deutsche Telekom: economic actors must comply with the overarching obligations imposed by the competition rules, while simultaneously discharging any parallel regulatory obligations.^[32] Typically, moreover, these legal requirements are enforced by different regulators, who each decide independently whether the norms falling within their jurisdictions have been breached. This orthodoxy has several important implications.

(i) No ouster of competition law in the presence of regulation

First, the existence of regulation within a market does not generally oust the applicability of competition law to the same behaviour, even if the regulatory regime has a significant impact on the conduct or underlying level of competitiveness within the market.^[33] The principal exception, the ‘State action’ doctrine, is narrowly construed,^[34] applying only where anticompetitive conduct is either required by regulation or where the regulatory framework removes all scope for competitive (and, by implication, anticompetitive) behaviour.^[35] The logic behind this defence is that competition law regulates anticompetitive behaviour by firms and not merely uncompetitive market structures; but its narrowness in practice suggests a deep reluctance to decouple the abstract notion of competition from the concrete actions of competitors.^[36]

This can be contrasted with the approach in US antitrust law, which more readily embraces a rule of antitrust preclusion where ‘a regulatory structure designed to deter and remedy anticompetitive harm’ can be identified.^[37] There are at least three reasons which explain the distinctive position to be found in EU law. First, the competition rules are primary EU law, meaning that there is a formal obstacle within the EU’s hierarchical legal structure to allowing ‘special’ regulatory rules found in secondary or national legislation to take precedence over the ‘general’ competition rules.^[38] Second, even if regulation takes account, inter alia, of the goal of effective competition, the assumption is that it almost always pursues ‘objectives which differ from those of [EU] competition policy’.^[39] The policy objectives of one set of rules cannot be subsumed into the other; with the result that, to the extent we wish to pursue both sets of goals simultaneously, both sets of rules must remain in operation.^[40] Finally, EU competition law incorporates a remarkably strong principle of effectiveness, which has aided the progressive development and expansion of these rules in many areas.^[41] Limiting any rule of antitrust preclusion to only the most exceptional circumstances where regulation removes all scope for (anti)competitive conduct serves to protect the effectiveness of competition law by ensuring that these rules continue to apply in the largest number of circumstances.

This means, for instance, that a dominant firm can be held to have engaged in an abusive margin squeeze (which entails manipulating the relationship between wholesale and retail prices for a product) where both price levels have been set or approved by a sector-specific regulator.^[42] Similarly, industry efforts to reorganise a sector suffering from an overcapacity crisis can be treated effectively as a cartel, even if undertaken at the behest of and in accordance with a plan devised by national government.^[43] And under the EUMR, strong governmental support has not conventionally provided a compelling counterweight where the Commission concludes that a proposed concentration is likely to significantly impede effective competition,^[44] although this position may prove more difficult to sustain in future.^[45]

The fact that, in EU competition law, defendants can breach the rules even where their conduct or the anticompetitive effects that follow are heavily influenced by state-imposed regulation is made possible by the lack of a fault or ‘bad’ intention requirement. Thus, under Article 102, ‘proof of the deliberate nature of the conduct and of the bad faith of the undertaking in a dominant position is not required for the purposes of identifying an abuse’.^[46] Under Article 101, it is ‘irrelevant’ to the assessment of whether a restriction by object exists that the coordinating parties acted in pursuit of ‘legitimate’ (i.e. not deliberately anticompetitive) objectives.^[47] The ‘by effects’ limb of that provision, additionally, considers only the impact in fact of an agreement on the market, rather than the reasons motivating its adoption.^[48] Intention also plays a very limited role in determining the essentially factual question of how a merger is likely to affect competition if it proceeds.^[49]

The absence of a fault requirement can be explained by the fact that the competition rules are aimed at behaviour that harms competition; yet the neoclassical economics that underpin the conventional competition law framework assumes that all firms ultimately intend to ‘harm’ competition in a literal sense by increasing their market share and profits at the expensive of competitors.^[50] Indeed, such self-interested behaviour has long been considered an essentially virtuous aim to the extent that it drives the competitive dynamics necessary to maximise efficiency.^[51] The upshot is that competition law takes an explicitly ‘objective’ approach, at least in principle, to the assessment of whether behaviour amounts to normal, meritorious—and so permissible—competition. To the extent that an absence of fault or bad intention is relevant within EU competition enforcement, it goes primarily to the penalty imposed: breaches must be committed intentionally or negligently to attract fines under Regulation 1/2003,^[52] while the influence of regulation on (otherwise autonomous) firm behaviour can be reflected in a discount on the overall fine.^[53]

(ii) Competition law and regulatory compliance are unrelated

Second, the question of whether behaviour violates competition law is formally ‘unrelated to its compliance or non-compliance with other legal rules,’^[54] regardless of how strong the link between the regulatory framework and the allegedly abusive behaviour. This means, on the one hand, that behaviour that conforms to a relevant regulatory standard might still breach competition law, if it amounts to an anticompetitive agreement or abuse of dominance.^[55] This position has been defended within EU law on the bases, inter alia, that decisions of (national) regulators should not bind the Commission,^[56] even if the regulator purports to apply and find compliance with the competition rules in parallel;^[57] and that the absence of a fault requirement, discussed above, means that good faith regulatory compliance is insufficient to avoid competition liability.^[58] The general principle that regulatory compliance is ‘unrelated to’ competition law liability would seem to infer that regulatory noncompliance is also largely irrelevant from an antitrust perspective, though Meta Platforms (discussed in Section IV(i)) nuances this assumption.

Accordingly, the case-law recognises that ‘in the majority of cases, abuses of dominant positions consist of behaviour which is otherwise lawful under branches of law other than competition law.’^[59] The jurisprudence is more complicated under Article 101, insofar as the sanction of nullity in Article 101(2) means that agreements that violate the provision are void as a matter of contract law too. But even in this context, the Court considers that compliance with other regulatory norms is, generally, ‘not, as such, a matter for competition law,’^[60] and so does not determine the question of whether an arrangement has the object or effect of restricting competition. Under the EUMR, when assessing whether a merged entity is likely to engage in harmful behaviour, the Commission takes account of any pre-existing regulation that might effectively inhibit the merged entity from doing so.^[61] It has also sought, in certain instances, to repackage regulatory compliance as a dimension of quality, to the extent that users place economic value on the underlying public interest protected (such as the level of data privacy offered by a product).^[62] Yet the Commission is clear that, if the objection to a concentration is a fear the merged entity will fail to comply with other regulatory rules in future, ‘[s]uch concerns are not within the remit of merger control and there are regulatory tools better placed to address them.’^[63]

(iii) Constrained powers of Commission when acting as competition regulator

This links to a third implication of the orthodox separation of competition law and other regulation within EU law, namely the limits of the Commission’s powers when ‘acting as competition regulator’.^[64] In the recent case of Orlen, the Court of Justice emphasised that, when exercising its competition powers, the Commission is restricted to imposing remedies or penalties that ‘may respond to the competition concerns [it has identified] on the market concerned.’^[65] Thus the Commission cannot legitimately use competition enforcement to advance other regulatory goals or interests,^[66] unless these can be absorbed into the ‘competition issues’ at stake.^[67] The Court acknowledged that the Commission’s competition enforcement activity ‘could not lead to a result which is contrary to’ other regulatory norms within EU law,^[68] and in particular, it cannot make binding remedies that would conflict with other (EU-level, at least) regulation. But that it quite different from saying that the Commission might positively impose ‘obligations…motivated by policies other than competition law policies’^[69] through the avenue of competition enforcement; this is something, the Court was clear, the Commission cannot do.

The attitude of the Court in Orlen potentially poses a not-insignificant significant hurdle to realising more ambitious applications of the competition rules,^[70] by establishing that competition law enforcement can pursue only ‘competition issues’. Of course, there is a risk of circularity, insofar as what we consider to legitimately constitute a ‘competition issue’ may vary depending upon how broadly we treat the legitimate objectives of competition law. But the judgment clarifies a point of importance to this article: competition enforcement is not an instrumental tool by which to advance a grab bag of ‘other’ regulatory objectives but must confine itself to responding to competition-related concerns.

(iv) Ne bis in idem and parallel enforcement of competition law and regulation

Finally, this orthodox understanding of the relationship between competition law and other regulatory norms—as discrete sources of legal obligations for market actors, applied by distinct regulators pursuing different public interest goals—creates potential for overlapping jurisdiction and enforcement. Compliance with regulatory obligations does not, as explained, certify that the regulated behaviour also complies with competition law. Yet overlapping jurisdiction does not always generate conflict, and it is entirely possible that the same behaviour might be found to violate both the competition law and the regulatory standards.

For many years, EU competition law recognised a distinctive principle of double jeopardy or ne bis in idem, which precluded the parallel application of multiple sanctions only if it was demonstrated that the regulatory regime pursued the same legal interest as the competition rules, in addition to unity of facts and offender.^[71] This unique approach reinforced the orthodox view of competition law and regulation as distinct and unrelated tools for market supervision. More recently, the Grand Chamber in bPost has realigned competition law with other areas of EU law by shifting to an idem factum standard requiring only the same facts and offender.^[72] The Court nonetheless recognised that duplicate proceedings and penalties might be permitted if a necessary and proportionate response to achieve public interest objectives or to protect the rights and freedoms of others.^[73] Of particular relevance is the Court’s strong assertion, when considering the proportionality of parallel proceedings, that ‘public authorities can legitimately choose complementary legal responses to certain conduct that is harmful to society through different procedures forming a coherent whole so as to address different aspects of the social problem involved, provided that the accumulated legal responses do not represent an excessive burden for the individual concerned’.^[74] Yet parallel jurisdiction cannot be exercised wholly unilaterally if it is to comprise a proportionate response. Instead, the Court in bPost stressed the need for sufficient ‘coordination’ between the competition regulator and other prosecuting authority to ensure a truly ‘complementary’ response.^[75]

IV. The Interaction of Competition Law and Regulation: the Contextual View

One shortcoming of the ‘two barriers’ understanding of the interaction of competition law and other regulation is that it sets up a dichotomy between these instruments of market supervision that does not reflect the reality of how most markets work today. In its competition case-law, the Court of Justice increasingly emphasises that the determination of whether conduct amounts to a breach of the rules requires a detailed assessment of context, meaning that decision-makers must take account of ‘all of the relevant factual circumstances’^[76] and the ‘economic and legal context’^[77] in which allegedly anticompetitive conduct occurs. The underlying regulatory landscape, insofar as it may shape competition dynamics and/or influence the defendant’s own behaviour, is an obvious contextual element that should be considered. Although EU law rejects the proposition that the presence of specific regulation should oust the more general competition rules as a matter of law, the former may play a significant role in determining whether competition law is breached on the facts.

(i) Relevance of regulatory norms in principle: from Deutsche Telekom to Meta Platforms

Indeed, even as the Court in Deutsche Telekom sidelined the State action defence and discounted the significance of good faith regulatory compliance in competition assessment,^[78] it recognised the relevance of regulatory context insofar as it ‘contributes to the determination of the competitive conditions under which an undertaking…carries on its business in the relevant markets’.^[79] The Court elaborated three areas where regulatory context might be relevant: ‘defining the relevant markets, assessing the abusive nature of such conduct or setting the amount of the fines’.^[80] In reality, the invocation of regulatory context in Deutsche Telekom was somewhat disingenuous, enabling the Court to endorse a cost-price calculation mechanism that was ‘at odds with the realities of the market’^[81] for the defendant and its rivals.^[82] Despite this, in both Slovak Telekom and Baltic Rail, the Court drew upon this statement from Deutsche Telekom to conclude that the regulatory context justified application of a less exacting legal standard under Article 102, reiterating that ‘a regulatory obligation can be relevant for the assessment of abusive conduct’.^[83] This language is echoed in later cases.^[84] It is the 2023 judgment of the Grand Chamber in Meta Platforms, however, that provides the most explicit support for a potentially pivotal role for regulatory norms within competition law assessment.

The Meta Platforms case began with an infringement decision of the German competition authority against the owner of the Facebook social network, which aimed at the latter’s user data-gathering policies. In essence, the decision construed Meta’s policies for use of Facebook services as an abuse of market power,^[85] using as a proxy for unfairness in competition law terms the fact that these policies were non-compliant with relevant data protection standards as reflected in the EU’s General Data Protection Regulation (GDPR).^[86] The Court of Justice, called upon to advise the fairly unconvinced national court, began by reiterating the orthodoxy of the distinctiveness of competition law from other regulation, including the different legal interests protected and enforcement by different regulators.^[87] Yet it quickly turned to acknowledge the applicability of regulatory standards and the role of compliance within the nominally distinct task of competition assessment.

The Court went much further than in previous cases, where the regulatory context had been described as potentially ‘relevant’ to this determination; in Meta Platforms, the Court held that ‘compliance or non-compliance…may, depending on the circumstances, be a vital clue’ as to the permissibility of behaviour under competition law.^[88] Notably, although the questions referred pertained only to data protection, the Court’s response was pitched in broader terms.^[89] Though the Court did not elaborate at length on how regulatory compliance might offer a ‘vital clue’ to competition liability, it noted the ‘great importance’ of having access to user data for many digital economy business models,^[90] so that it might undermine the effectiveness of competition enforcement to ‘disregard the reality of this economic development’.^[91] Echoing bPost,^[92] the Court then considered how national authorities should collaborate to minimise ‘the risk of divergences’,^[93] an approach that it grounded in the general EU law duty of sincere cooperation in Article 4(3) TEU. The judgment thus offers a more nuanced account of the inter-relationship between competition law and other regulation, by acknowledging that the latter may influence or even determine the application of the former in practice, depending upon its effects within a relevant market.

(ii) Relevance of regulatory norms in practice

The impact of the regulatory context on competition law assessment can be grouped into three broad categories: where regulation effectively excludes competition law liability; where it changes the structural competition dynamics within a market; and where it influences the behaviour of economic actors in a manner that prevents—or conversely provokes—competition law infringements.

a. Regulation as exemption of competition law liability

Most radically, though also most exceptionally, regulation might remove the scope for applying the competition rules to market behaviour or at least preclude a finding of liability. The competition rules apply to ‘undertakings,’ a concept that has been interpreted to exclude certain (typically highly regulated) market actors that pursue objectives related to social solidarity^[94] or that exercise quasi-public powers.^[95] The State action defence, noted above, provides an effective means to avoid Articles 101 and 102, but only where the applicable regulatory regime removes all capacity for ‘autonomous conduct,’ competitive or otherwise, by regulated undertakings.^[96] Similarly, under Article 106(2) undertakings that are ‘entrusted’ by Member States with the operation of services of general economic interest can be exempted from application of competition law, but only to the extent that being subject to such obligations would ‘obstruct the performance, in law or in fact’ of this task. These three disparate exceptions share a common theme: although the would-be defendant is engaged in what is, strictly speaking, ‘economic’ activity on the market, the nature or impact of the underlying regulatory framework removes its actions from the purview of the competition rules as a matter of law, regardless of their effects on competition in fact.

Alternatively, the operation of the regulatory framework may provide a good defence that enables a defendant to rebut the claim that its conduct is harmful from a competition perspective. The Article 101 jurisprudence recognises the ‘Wouters exception,’ which enables certain prima facie restrictive arrangements to escape prohibition ‘by effect’ on the basis that the coordination pursues ‘legitimate objectives in the public interest which are not per se anticompetitive,’ and subject to indispensability and proportionality requirements.^[97] Formally, the exception does not hinge upon the impact of regulation on the activities under scrutiny; that is, the concept of ‘legitimate objectives in the public interest’ does not require that defendants are empowered to achieve those objectives under public regulation. But it is notable that, on the facts of Wouters, the organisation concerned (the Dutch bar association) was expressly tasked under domestic legislation to adopt the rules under scrutiny, ‘in the interests of the proper practice of the [legal] profession’.^[98] In this sense, a parallel can be drawn to the exceptions to the ‘undertaking’ concept that have similarly developed in the jurisprudence.^[99] In each instance, although the activity concerned is ‘economic’ in nature the predominant interests at stake are not, and typically (indeed, perhaps necessarily in practice) this will be reflected in the underlying regulatory framework.^[100]

Under Article 102, it is difficult to successfully argue that regulatory obligations deprive a dominant undertaking’s conduct of its ‘autonomous’ nature so as to preclude antitrust scrutiny.^[101] But there may be greater scope, where behaviour is driven by such considerations, for advancing a claim of ‘objective necessity’ which can serve as an objective justification, or defence, to a prima facie finding of abuse.^[102] This potential is seen most clearly in the case-law on refusal to deal, where the jurisprudence explicitly recognises that the ‘applicable regulatory framework’ may impose constraints on dominant firm behaviour in a manner that essentially legitimates what would otherwise be an abusive refusal to grant access to the dominant firm’s property.^[103] But objective necessity claims are narrowly construed in practice, and defendants have the burden of establishing that their behaviour is both necessary to achieve the public interest aim and that the actual or potential exclusionary effects resulting from the conduct are proportionate to the allegedly necessary aim.^[104] The objective necessity defence moreover does not apply where a dominant defendant claims, ‘rightly or wrongly,’ that its behaviour was necessary to prevent other market actors from breaching their regulatory obligations,^[105] the idea being that private actors should leave the task of enforcement to the professionals, i.e. the applicable public regulator.

b. Regulation as a structural feature of market competition

Most of the time the underlying regulatory framework does not provide anything so dramatic as a wholesale dispensation from competition scrutiny. It is more usual for regulation to feed into antitrust assessment either by heightening or dampening competition dynamics (i.e. the possibilities for effective competition within a market) or by altering (whether by constraining or directing) the behaviour of economic actors in ways that make anticompetitive conduct more or less likely.

Regulation, first, can function structurally as a barrier to entry or expansion by competitors, typically enhancing the power of incumbent market actors and making anticompetitive effects more likely or plausible. The case law recognises a wide variety of such circumstances, including statutory monopolies^[106] and intellectual property rights.^[107] Consideration of barriers to entry is a key component of the initial assessment of dominance under Article 102, and can also be highly informative of whether a merger is likely to create or strengthen a dominant position, so as to violate the ‘significant impediment to effective competition’ standard under the EUMR. Such structural effects may endure even after regulation: so, for instance, entities that previously benefitted from special or exclusive rights under regulation may continue to experience de facto benefits after such legislation is removed,^[108] firms may maintain anticompetitive arrangements after the regulatory justification for doing so has disappeared,^[109] or even seek subsequently to reimplement regulatory barriers through private arrangements after these have been dismantled by market liberalisation.^[110]

Regulatory barriers to entry may also be relevant to the assessment of likely anticompetitive effects that follow from behaviour.^[111] It was noted above that contemporary competition assessment is highly attuned to the specific context in which conduct occurs; increasingly, the case-law also emphasises that competition law aims at behaviour that has at least potential anticompetitive effects.^[112] The regulatory framework may therefore feed into the evaluation of whether coordination violates Article 101(1) or that single firm conduct constitutes an abuse of existing market power, insofar as the regulatory context may influence ‘the consequences of a certain practice in the market or for consumers’.^[113]

A more complex question is whether the anticompetitive effects must be caused by the impugned behaviour. In Google Shopping, the Court of Justice described the establishment of a causal link between a firm’s behaviour and the claimed actual or potential anticompetitive effects as one of the ‘essential constituent elements of an infringement of competition law’.^[114] This would appear to preclude a finding of breach of Articles 101 or 102 where the harmful effects on competition are attributable more or less entirely to the impact of the relevant regulation, even if defendant undertaking(s) have engaged in what looks like wholly unmeritorious behaviour from a competition perspective. Accordingly, while there is a high threshold to prove that regulation robs firm behaviour of its autonomous character, there may be greater scope to argue that the pervasive distortive effects of regulation mean the necessary causal link cannot be found. Yet Google Shopping does not require an exacting causation standard, with the Court accepting that it is sufficient merely to demonstrate a ‘correlation’ between suspect practices and the evolution of the market concerned, provided that a causal relationship can be confirmed by ‘additional information’ such as the position of market participants.^[115] The question in regulated markets is therefore whether it is possible to identify a sufficient diminution in effective competition arising from the suspect behaviour, over and above any limitations or distortions arising from the regulatory restrictions.

Conversely, regulation may facilitate new entry which enhances competition overall. This is, for instance, a principal objective of much of the market liberalisation legislation introduced in the EU since the 1990s.^[116] There are a number of common regulatory techniques which are deployed frequently to increase opportunities for competition: such as requiring the removal of existing monopoly rights that protect (typically dominant) incumbent operators from new entry; imposing structural separation on vertically-integrated incumbents to reduce incentives to engage in exclusionary behaviour; and requiring incumbents to share existing infrastructure or even data with potential new rivals, to enable or assist entry. Regulation aimed at market-opening may be supplemented by regulation that controls the behaviour of former monopolists in liberalised sectors, a point developed further below. In such circumstances, antitrust takes account of the altered market dynamics effectuated by the presence of the regulation, so that a defendant cannot, for instance, argue that its behaviour should be assessed as if unencumbered by its regulatory burdens.^[117] On the other hand, the regulatory framework may make it more plausible for a defendant to claim that its superficially restrictive behaviour generates countervailing efficiencies.^[118]

c. Regulation and its impact on market behaviour

Regulation may also influence a firm’s market behaviour in ways that have implications for competition assessment: either because it directly controls market behaviour or because it incentivises conduct that generates (anti)competitive effects. Here, several scenarios can be envisaged.

Most constructively, regulation may guide a firm to competition law-compliant behaviour. Where appropriately calibrated and effectively implemented, a regulatory framework may prevent regulated firms from engaging in the sorts of behaviour that would otherwise cause competition problems: by limiting a firm’s capacity to charge prices that are excessively high or have exclusionary effect, for instance, or by preventing use of ‘unfair trading conditions’ in contractual relations. This outcome is most likely to manifest itself obliquely, in an absence of antitrust problems in a sector.^[119] To the extent that context-specific competition assessment takes place prospectively, the general starting point is that the market behaviour of economic actors is shaped and constrained by other regulatory norms that apply within that sector and to that conduct.^[120] This is most salient in merger control, where it is assumed that firms comply with their regulatory obligations, and that their scope for ‘free’ market conduct is curtailed accordingly.

More contentiously, regulation may guide a firm to behaviour that is essentially at odds with competition law, a scenario most likely to arise where the regulatory regime endorses ‘inefficient’ market outcomes in pursuit of other policy goals. In this case we may encounter the possibility flagged in section III, namely prosecution under the competition rules of behaviour that is not only fully compliant with, but has also been largely influenced by, a regulatory framework that deliberately prioritises other socially valuable goals.^[121] This is an essentially unsatisfactory outcome which is considered further below.

The first two scenarios each envisage effective compliance with regulatory obligations, whether this leads to competitive market outcomes or not. But firms do not always respect their regulatory obligations or make good faith efforts to do so. Within the competition jurisprudence, it is possible to identify several scenarios whereby an antitrust defendant’s liability is essentially contingent upon its interaction with another regulatory framework. It is these cases, in which regulatory norms function as the ‘hook’ for additional antitrust liability, which raise the most interesting but also the most difficult questions regarding the impact of regulatory norms on competition law.

Most straightforwardly, a regulated firm may simply disregard the constraints that are placed on its behaviour. In Meta Platforms, a digital firm’s data-gathering policy was deemed to amount to an abuse of its market power because the policy failed to respect the relevant data protection standard. A similar approach has been adopted in the Commission’s enforcement activity in liberalised utilities sectors, where failure to abide by regulatory duties to share infrastructure is also construed as an abuse of dominance by the incumbent operator.^[122] Here, the regulatory violation—which may or may not have been prosecuted by the sector regulator in parallel—is usually obvious; the more interesting question is how and why the behaviour might additionally become the focus of an antitrust investigation.

Beyond the problem of mere non-compliance, the competition case-law discloses a variety of instances where liability hinges on a firm’s disruptive interaction with a regulatory framework. In essence, these cases involve so-called ‘regulatory gaming,’ namely ‘[b]ehaviour that abuses a neutral or procompetitive regulatory structure and wields it as a tool to accomplish [anticompetitive] results’.^[123] What distinguishes the non-compliance cases from the regulatory gaming ones is that, generally speaking, in the latter the firm’s behaviour is unproblematic (or at least unprosecutable) under the relevant regulatory regime. Instead, antitrust liability derives from the defendant’s efforts to avoid the regulatory constraints while remaining within the strict letter of the law (what we term ‘bad faith compliance’), or to exploit opportunities created by the existing regulation to anticompetitive ends (‘abusive compliance’).

Many examples of bad faith compliance relate to the highly regulated pharmaceutical sector. In AstraZeneca,^[124] the first EU competition case that directly embraced regulatory gaming as a theory of harm, the defendant took advantage of loopholes in the rules on pharmaceutical marketing authorisation and weaknesses in the administration of national patent offices to exclude would-be generic competitors. In Teva,^[125] the defendant strategically lodged and withdrew divisional patent applications in a manner permitted by the regulatory framework but which was clearly intended to frustrate new entry. Most remarkably, in Aspen Pharma^[126] the defendant successfully implemented an excessive pricing campaign in a context where every Member State had domestic price regulation, by exploiting design flaws which allowed it to manipulate the regulatory price-setting processes. For our purposes, what is most notable about these cases is how they make the leap from attributing antitrust liability for non-compliance with the express rules of a regulatory framework to non-compliance with the broader spirit or underlying policy objectives of those rules.

In abusive compliance cases, the regulatory context is primarily of relevance insofar as it provides an explanation for strategic behaviour that otherwise lacks an anticompetitive rationale. This is exemplified by AB InBev,^[127] where the defendant changed its packaging for beer sold in France and the Netherlands to prevent cross-border resales in Belgium, where it charged higher wholesale prices. This change—to no longer include information in both Dutch and French on beer sold in countries other than Belgium—only made sense as an exclusionary strategy in view of Belgian legislation on food labelling, which meant that larger retailers only purchased products labelled in both languages. In Roche & Novartis,^[128] the defendants engaged in collusive lobbying efforts to persuade the Commission to change its advice on the off-label prescribing of a certain drug, which enhanced the profits of both cartelists. Similarly, in Car Emissions,^[129] car manufacturers agreed to design new vehicles to the minimum regulatory standard for nitrogen oxide cleaning, an arrangement that the Commission prosecuted as a secret cartel. What is notable is that the cartelists explicitly agreed to adhere to their existing regulatory obligations, albeit knowing that these were likely to become more demanding in future; in doing so, however, they effectively agreed not to compete on cleaning technology as a dimension of competition, which the Commission construed as a limitation both of consumer choice and future innovation. Pharmaceutical pay-to-delay agreements, whereby drug companies pay off potential new entrants in the guise of patent litigation settlements, can also be seen as abusive compliance.^[130]

The real world is complicated, of course, and some regulations may be equivocal in terms of their overall effects on competition, so that a neat dichotomy between regulation that limits competition versus that which enhances competition often cannot be drawn. The GDPR provides an example of such mixed effects. On the one hand, by imposing limits on what market actors can do with personal data, it constrains the market power of Big Tech companies which have disproportionately plentiful access to such commercially valuable information.^[131] On the other, GDPR compliance is an expensive business, and so the very existence of these rules, which apply to all data controllers regardless of size, may constitute a barrier to entry.^[132] The GDPR says little about its intended effects on competition beyond a vague reference to levelling the playing field,^[133] yet it has significant impact on many digital business models, as recognised in Meta Platforms.^[134]

(iii) Understanding the role of regulatory norms

Regulation accordingly feeds into competition assessment in numerous (occasionally contradictory) dimensions, from reducing the scope to identify autonomous market behaviour to enhancing the risk that firm conduct generates harmful effects. Where competition law is applied in regulated markets, the competition rules do not generally endorse or apply the regulatory norms as such. But by accepting a market context where these norms reflect the accepted ‘rules of the game’ for market participants, competition law essentially reinforces the more diverse public interest values that are reflected within other regulatory regimes. Occasionally, this means deferring entirely to alternative values by allowing the regulatory regime to displace the application of competition law or to provide at least a good defence. More usually, these norms are absorbed within the contextual analysis, and the case-law increasingly recognises the legitimacy of aligning the thrust of competition analysis with pre-existing regulatory obligations (provided, of course, that these are not fundamentally at odds). There is obvious potential for tension, where for instance competition law turns its focus on market power that has been created by regulation as, say, a reward for innovation or to ensure the universal delivery of public services. But even here, competition analysis typically finds a way to accommodate potentially conflicting values, by distinguishing between the lawful scope of intellectual property rights and efforts to extend those rights beyond what is envisaged by IP law,^[135] for instance, or by recognising the fundamentally different position of liberalised and non-liberalised markets.^[136] Even if the goal of competition law is merely to achieve and protect undistorted competition, the space given to other regulatory norms within the contextual approach allows for the accommodation of complementary (and sometimes conflicting) values. These cases thus illustrate the tendency towards a ‘sponge’^[137]-like or ‘polycentric’^[138] quality within competition law, to the extent that a diversity of values can be accommodated—if not necessarily pursued directly—within contemporary enforcement.

The above discussion nonetheless identified two broad areas where this narrative of an essentially passive-but-tolerant inter-relationship between competition law and other regulatory norms does not provide an effective explanation. The first scenario arises where a (typically national) regulatory framework favours inefficient market outcomes in pursuit of non-economic objectives, albeit leaving sufficient scope for independent conduct to attract antitrust attention. The unsatisfactory position in the current law has been noted: in line with the orthodox approach, unless the narrow State action defence applies distortions generated by the regulatory framework can be reflected only by discounting the penalty that might be imposed.^[139] This position can be critiqued from two dimensions. It creates unfairness for, or at least imposes a heavy burden on defendants, who seem to have a positive responsibility to seek align their regulatory obligations with the perceived demands of competition law.^[140] But even leaving the question of fairness aside, when viewed through the lens of contextual analysis this approach gives little weight to the obviously significant impact of the regulatory framework on competition dynamics in the market concerned. Instead, there is a strong argument that the (typically, non-economic) considerations factored into the regulatory framework should inform, though not necessarily determine, the competition enforcer’s understanding of what ‘competition on the merits’ looks like in a sector. While this may be insufficient to provide a good defence where there is further evidence that the defendant’s conduct is objectively non-meritorious in competition law terms, it would at least provide some recognition of the public interests that motivate the regulatory framework. To the extent that the latter are pretextual or at least insufficient to justify the restrictions on competition entailed, the more appropriate competition policy response is to attack the regulation directly.^[141]

A second category of cases that require more in-depth consideration are those involving non-compliance and regulatory gaming, which pose a conceptual dilemma from a competition policy perspective. Through such behaviour, a firm may engage in conduct that is harmful to effective market competition, which seems to fall within the legitimate scope of the competition rules. Yet insofar as the firm’s ‘unmeritorious’ behaviour revolves principally around its interactions with another regulatory framework, one might ask what makes this an appropriate subject for antitrust intervention, as opposed to the more obviously and directly applicable regulatory regime. In cases of non-compliance, where the behaviour violates the regulatory norm, the question is how to justify competition enforcement as opposed to (or in addition to) direct regulatory enforcement. In cases of bad faith compliance, the question is why competition law might take it upon itself to redesign or fill in perceived gaps within a regulatory framework rather than deferring to the regulation itself. Cases of abusive compliance are less intimately entwined with the obligations imposed by a regulatory framework, making these easier to explain as examples of ‘regulation as context,’ discussed above. Yet such cases almost unavoidably involve value judgments regarding the nature of the defendant’s interaction with the regulatory regime, and so raise a similar question to other categories: how can we explain the role played by regulatory norms within the competition law assessment in such cases? This is the focus of the next section of this article.

V. Understanding the role of regulatory norms in competition assessment: explaining the non-compliance and regulatory gaming case-law

The crux of concern is that these cases involve, in essence, the ‘outsourcing’ of liability standards within competition law to other regulatory fields. A competition lawyer schooled in the ‘two barriers’ approach to the competition law/regulation inter-relationship would not question the legitimate existence of regulatory norms ‘other than competition law policies’.^[142] Yet it is less obvious why competition law should effectively defer to standards of behaviour developed in other regulatory policy areas—a problem that becomes more acute in cases of regulatory gaming, which involve deference to vaguer understandings of what a regulatory regime ought to entail if working well. The objection is not that the policy goals reflected in the regulatory framework lack legitimacy or value. Instead, the question is why competition law internalises these goals by applying regulatory norms as the determinative standard of competition liability in such cases. Put another way, why should protection of the goals, values and policy choices that are reflected in a specific regulatory framework fall within the legitimate scope (and enforcement purview) of competition law in addition?

Before exploring this question, let us recap several core principles of the ‘orthodox’ jurisprudence. Formally, antitrust compliance is unrelated to regulatory compliance within EU law: a firm might breach the competition rules whilst respecting its regulatory obligations, and vice versa. This links to the absence of a fault requirement, meaning that good faith regulatory compliance does not immunise a firm from competition liability in respect of the same behaviour. Competition enforcers are limited to the pursuit of ‘competition issues’; a somewhat woolly limitation, but one which confirms that the competition rules are not a panacea for all public policy ills. Accordingly, to explain the non-compliance and regulatory gaming cases, we must consider why as a matter of competition law such behaviours come within the legitimate scope of the competition rules. To do so, we explore the case-law from three distinct perspectives—through the lenses of effectiveness, competition on the merits, and regulatory non-compliance and gaming as strategic action—asking whether each provides a plausible explanation for addressing the behaviour under competition law.

(i) Effectiveness of competition law as a regulatory enforcement mechanism

It is important to recognise why both non-compliance and regulatory gaming cases typically arise in practice: namely, the effectiveness of competition law as a regulatory enforcement mechanism, unconstrained by substantive or procedural limitation that may hinder enforcement of the regulatory framework directly. This is apparent from the background to many cases considered above. In Telecom Polska, the sector regulator had repeatedly enforced national telecommunications law against the defendant, but the penalties it imposed were insignificant and failed to deter non-compliance. In Slovak Telekom, the defendant had evaded its regulatory obligation to grant access to infrastructure for five years but received no pushback from the (captured or under-resourced) regulator. In Meta Platforms, the problem was a mismatch of jurisdiction: while the data privacy concerns were felt most acutely by German consumers and policymakers, power to enforce the GDPR lay with the Irish data protection authority, which dragged its feet. The deficiency of national patent systems is a key undercurrent in many competition cases in pharmaceutical markets.^[143]

Viewed pragmatically, it is easy to understand why competition law is used to reinforce regulatory norms: in effect, competition enforcement might get the job done more effectively than the regulation itself. Competition law brings much to the table as a mechanism to enforce regulatory rules. In particular, the broad sweep of its prohibitions means that competition law is sufficiently wide and flexible to catch such behaviour, including problematic conduct that might lie outside the formal parameters of the regulation itself. EU law also imposes demanding minimum standards to protect the independence, resourcing and sanctioning powers of competition authorities,^[144] which sectoral regulators often lack. The resistance within EU law to accept antitrust preclusion even in the presence of significant regulation supports the proposition that competition law can legitimately seek to resolve market problems that arise from malfunctioning regulation, providing a ‘complementary legal response[]’ in order to form ‘a coherent whole’ of market supervision.^[145]

Yet the bare fact that competition law can enforce regulatory norms more successfully provides an inadequate explanation for these cases. On the one hand, using competition law as a sticking-plaster on ineffective regulation gives little weight to the substance, and substantive limits, of the underlying regulatory framework. Cases like bPost and Meta Platforms demonstrate that, if competition law is applied to ‘regulated’ problems, this requires a degree of intentionality and coherence and must be done in a manner that recognises the ‘respective powers and competences’^[146] of each regime. Deploying competition law on the pragmatic basis that it can better achieve the goals of a regulatory framework fails to acknowledge the central role of the underpinning regulation in such cases, both in terms of how ‘the social problem involved’^[147] and the appropriate enforcement response should be framed.

On the other hand, justifying these cases on an effectiveness basis offers little insight into how such regulatory failures can also become, in the language of Orlen, ‘competition issues’.^[148] Regulation, as discussed, often incorporates a diversity of policy goals that extend beyond the protection of undistorted competition. Regulatory standards of acceptable behaviour typically seek to advance values outside the scope of competition law as such. To take an notable example, the GDPR states that ‘[t]he processing of personal data should be designed to serve mankind,’ and explains that the rules seek to balance a variety of fundamental rights including privacy and family life, freedom of thought, freedom of expression, freedom to conduct a business, fair trial rights, and cultural, religious and linguistic diversity.^[149] While it is difficult to argue against the importance of these varied considerations, it is clear that a data protection standard—such as that applied to impose antitrust liability in Meta Platforms—reflects a wider ranges of values than the mere protection of undistorted competition. If we want to bring the former within the legitimate scope of the latter, we need a more nuanced explanation as to when and why diverse regulatory concerns are also ‘competition issues’.

(ii) Regulatory compliance as a reflection of ‘competition on the merits’

Instead, we return to the concept of ‘competition on the merits’: the idea that competition law does not seek to constrain firms from competing, even aggressively, but merely limits market activity to the acceptable parameters of ‘normal competition’. The concept is prominent in recent jurisprudence^[150] and in the Commission’s Draft Guidelines on Article 102. The latter follow the approach in SEN^[151] by decoupling the notion of ‘conduct departing from competition on the merits’ from ‘capability to produce exclusionary effects,’ while indicating that both elements must be satisfied to establish exclusionary abuse of dominance.^[152] The Draft Guidelines nonetheless suggests that capability can be presumed for a variety of practices,^[153] making competition on the merits the key determinant of liability in many circumstances.

The proposition that competition law protects only ‘competition on the merits’ might justify the direct application of regulatory norms to the extent that the standard of ‘normal’ competition implies that market actors, inter alia, follow their regulatory obligations. Regulatory compliance is often costly for market actors, whether because the regulation curtails a firm’s scope to engage in profitable activity^[154] or requires it to incur costs it would prefer to avoid,^[155] or simply because of direct compliance costs.^[156] While escaping or minimising regulatory obligations can give individual undertakings a competitive advantage, doing so undermines the public interest values that the regulation advances and protects. Viewed in this manner, non-compliance or gaming of regulatory norms is inherently at odds with what we understand meritorious, and thus permissible, market competition to look like. Such an approach aligns with the ‘level-playing field’ notion that permeates EU law, which is premised on the idea that market actors should be subject to common rules and restrictions to ensure fair competition.^[157] Since the regulatory framework structures the ‘fair’ competitive game in the level-playing field of the marketplace, it is reasonable to require adherence under competition law even if some regulatory rules pursue values that are non-economic in nature.

At its simplest, the existing regulatory framework might be taken to reflect the lawful scope of a firm’s freedom in the marketplace. By incorporating regulatory standards into antitrust assessment, we do not ‘enforce’ these norms as such. Yet because a defendant can have no legitimate expectation to act contrary to its regulatory obligations, it cannot object where this behaviour is curtailed or punished through competition enforcement. Whereas in Slovak Telekom and Baltic Rail the Court denied that a regulatory violation might comprise an abuse in itself, it accepted that the presence of regulation may permit a less demanding legal standard for antitrust intervention.^[158] It justified this essentially on the basis that pre-existing regulation already circumscribed what the defendants could reasonably expect to be allowed to do in the marketplace. Competition law need not maintain the fiction that, say, a firm can validly refuse to grant access to its infrastructure unless an antitrust duty to deal arises exceptionally under Article 102, if sector-specific regulation has already imposed an unambiguous mandatory sharing obligation. A similar logic is discernible in the Commission’s approach to sustainability agreements in its Horizontal Cooperation Guidelines.^[159] Where ‘appropriate regulation’ exists to adequately address market failures, the Guidelines are sceptical of the acceptability of private coordination in tandem,^[160] ‘given that the legislator has already decided that each undertaking must individually comply with the obligation in question’.^[161] Yet private action may be more necessary to address ‘residual market failures that are not or not fully addressed by public policies and regulation’.^[162] In the latter case, undertakings have plausible claims for greater freedom of (private) action, since regulation has not yet delimited the legitimate scope of market activity. This reflects, in the language of Meta Platforms, ‘the reality of economic development’ in regulated markets.^[163]

Yet treating a regulatory framework as merely indicative of the limits to a firm’s economic freedom provides an incomplete explanation of the non-compliance and gaming case-law. First, it is less convincing in respect of behaviour that violates the spirit of a regulatory norm rather than its express provisions. It is one thing to assert that a defendant can have no legitimate expectation, for example, to refuse to comply with a mandatory sharing obligation imposed by sector-specific regulation; it is quite another to make this claim regarding behaviour that is, say, objectively ‘misleading’ but not strictly speaking unlawful.^[164] Arguably, the mere existence of collusion or dominance limits what firms can legitimately expect to do in markets, even absent abusive behaviour;^[165] and applying this logic, we might extend the proscriptive potential of regulation to a wider subset of behaviours at variance with its policy objectives more broadly.^[166] Yet this creates a legal certainty problem: how can firms know in advance what sorts of behaviours are likely to conflict with the broader ‘regulatory orbit’, if this is not clearly delimited by regulation? Second, this approach provides only a partial account of what happens in non-compliance and gaming cases: it tells us why defendants cannot object to competition enforcement covering the same ground, but not why such behaviour can legitimately be construed as a ‘competition issue’ as well as a regulatory breach.

A more muscular version of this argument is what we call the ‘dominance-plus’ approach. From this perspective, because non-compliance or regulatory gaming is inherently non-meritorious, any conduct of this sort is fair game under competition law provided that a threshold criterion for antitrust jurisdiction (most naturally a dominant position under Article 102) is also satisfied. While competition law cannot compel firms directly to comply with their regulatory obligations, it can punish non-compliance or gaming insofar as such conduct falls within the discrete prohibitions in the competition rules. It is the existence of significant market power (or the fact of coordination, if we extend the logic to a collusion-plus approach under Article 101^[167]) that brings the matter within the scope of competition law; the regulatory misconduct then provides the plus factor that crystallises the competition infringement. This was, effectively, the position of the Bundeskartellamt in Meta Platforms.^[168]

This approach has a degree of intuitive appeal. It is hard to argue that a firm acts meritoriously if it disregards or takes advantage of its regulatory obligations, though it is more debatable as to whether this is ‘normal’ market behaviour. Although the regulatory regime may reflect values beyond the task of protecting undistorted competition, the enforcement of such values in non-compliance and gaming cases can be explained as an example of competition law’s ‘sponge’-like quality, a term developed by Ezrachi to describe the capacity to absorb and reflect policy influences from other sources such as market regulation.^[169] Undeniably, this is an understanding of competition law’s role that reflects a more overtly ‘political’ vision of antitrust as a tool directly to regulate economic power, a perspective which typically gives less weight to more ‘legal’ questions of the permitted parameters of a dominant firm’s market freedom.^[170] Yet it is not obviously precluded by the Article 102 jurisprudence, which recognises the ‘special responsibility’ of dominant firms to protect effective competition,^[171] and which does not require misuse of the market power that underlies a dominant position in order to establish ‘abuse’.^[172]

Yet even accepting that dominant firms have a special responsibility to avoid distorting competition, objections can be raised against construing regulatory non-compliance or gaming of norms as ‘abuses’ in an antitrust sense. The case-law establishes that, even if ‘a regulatory obligation can be relevant for the assessment of abusive conduct,’ the presence of regulation ‘cannot relieve the Commission of the requirement of establishing that there is abuse’ on the facts.^[173] While Meta Platforms saw the most ambitious use of regulatory norms to determine liability, the Court still stopped short of treating regulatory non-compliance as conclusive proof of antitrust abuse, and it emphasised the differing ‘objectives and tasks’ pursued by competition law compared with other regulatory fields.^[174] In bPost, where the Court acknowledged that competition law might ‘legitimately’ tackle behaviour that is also subject to regulatory controls, it clearly envisaged that parallel proceedings would address diverse regulatory objectives.^[175] It is doubtful whether repackaging a regulatory violation as an antitrust one merely due to the presence of dominance (or indeed coordination) respects this limitation. The dominance-plus approach also sits uncomfortably with the principle that good regulatory compliance is generally irrelevant to the objective question of whether conduct attracts antitrust liability, provided that the regulatory regime allows sufficient scope for autonomous behaviour that can come within the reach of the competition rules.^[176]

This concern—that defendants might somehow be ‘damned if they do and damned if they don’t’ respect regulatory norms—highlights an additional objection, namely the risk of disproportionate punishments for what are, at their core, regulatory violations. bPost does not prohibit the parallel enforcement of regulatory norms, although it includes proportionality considerations in the determination of whether an exception to ne bis in idem is permissible in individual cases. But even where proceedings are limited to competition enforcement, the consequences for defendants are often much more disadvantageous than if prosecution was pursued under the regulation directly.^[177] Even if one has little sympathy for firms that disregard their regulatory obligations, reinterpreting a regulatory violation as a competition abuse can raise the stakes considerably, including the possibility of private damages claims alongside public enforcement.^[178] There is also the question of how to impose fair and sensible limits on the scope of liability from a dominance-plus perspective: might any regulatory breach—whether of tax law, zoning rules, labour protections, etc.—become an antitrust abuse merely due to the existence of dominance (or collusion)? Concerns about disproportionate liability are even more acute in regulatory gaming cases, where the defendant has not in fact violated the relevant regulatory norm, so that antitrust enforcement not only reinforces but extends regulatory liability.

Yet within the emerging Article 102 jurisprudence the existence of conduct departing competition on the merits is not (typically) the sole determinant of whether an abuse arises; additionally, at least in principle, the conduct must have the capability to produce exclusionary effects.^[179] An effective limiting principle for non-compliance and regulatory gaming cases could conceivably be derived by adding a robust requirement to demonstrate anticompetitive effects arising from the regulatory breach/gaming, alongside a ‘but for’ causation standard directly linking the behaviour to its harmful effects. Demonstrating actual (or at least potential^[180]) and appreciable anticompetitive effects would squarely locate the problem within the sphere of ‘competition issues’ that fall within the legitimate scope of competition law. Requiring a ‘but for’ causation standard, additionally, would enable us to explain why this regulatory abuse should also generate liability under competition law. Such an approach has much to recommend it in principle.

It is at this juncture, however, that we run up against the constraints of the existing case-law. Although the recent jurisprudence makes greater space for effects-based considerations, in practice it still falls short of both proposed elements in most instances. While the case-law is not a model of clarity,^[181] it is only with respect to the ‘by effect’ limb of Article 101(1) that we find anything close to such a robust effects requirement.^[182] Under the ‘by object’ limb,^[183] it is essentially unnecessary to examine effects where an arrangement discloses intrinsic harm to competition ‘by its very nature’.^[184] Under Article 102, which is the basis for most of the cases considered here, behaviour needs only to be capable of restricting competition,^[185] a standard that can be met despite evidence that the conduct ‘has not produced actual anti-competitive effects’.^[186] To the extent that the case-law imposes a causation requirement, it foresees merely a need to demonstrate a ‘correlation’ between the defendant’s conduct and the claimed (likelihood or capacity to cause) harm.^[187] There are sound competition policy reasons to defend a more prophylactic approach, in order to prohibit overtly restrictive arrangements^[188] or where the presence of a dominant firm implies a structural absence of competition.^[189] Yet the continuing distance between the rhetoric and reality of the effects-based approach means that a simple dominance-plus perspective may provide an over-inclusive account of when regulatory non-compliance and gaming of norms should also constitute competition issues.

(iii) (Non) compliance as a positive strategic action

Another way of thinking about these cases is to view non-compliance or regulatory gaming as a positive strategic action for competition law purposes. Here, the focus is the deliberate choice of defendants to neglect or ‘game’ their regulatory obligations as part of an anticompetitive course of conduct: the antitrust problem stems not merely from the fact the regulatory norm has been disregarded or disrupted, but also from the motivation for this behaviour. In effect, the regulatory framework is ‘weaponised’ to anticompetitive ends, and it is this dimension—rather than the regulatory element as such—that is pursued under competition law. The key aspect is thus the strategic behaviour of the defendant undertaking in terms of how it interacts with, takes advantage of, or disregards its regulatory obligations.

To provide an effective explanation of the non-compliance and regulatory gaming case, the question is whether (and in what circumstances) ‘strategic’ behaviour of this sort can breach the competition rules. Recent case-law provides support in several dimensions for an approach to liability that hinges on evidence of the anticompetitive intention defendants. The transformative Intel judgment emphasised the relevance of evidence of an anticompetitive ‘strategy’ to find abuse of dominance.^[190] The Court Google Shopping endorsed a focus on the ‘positive acts’ undertaken by the defendant,^[191] which served to distinguish its (abusive) actions from other (permissible) forms of self-preferencing. It also held that, when engaging in context-specific assessment, it is ‘clearly legitimate for the Commission to assess subjective matters, namely the motives underlying the business strategy in question’.^[192] The more contextual approach to ‘by object’ determination under Article 101 similarly extends consideration of the inherent ‘nature’ of coordination to the ‘aims which that conduct seeks to achieve from a competition standpoint’.^[193]

Yet intention-based requirements are controversial in competition law. Two broad objections can be advanced against conceptualising the non-compliance and gaming cases as instances of ‘strategic’ behaviour: one substantive and one practical. In substantive terms, it has been argued that in the presence of market power or coordination, certain forms of conduct should be objectively unacceptable due to their inherent capacity to distort competition, regardless of what a defendant seeks to achieve by its actions. This perspective is reflected in the jurisprudence, which rejects a determinative role for intention evidence within the framework of EU competition analysis. The perceived ‘objective aims’ of coordination are distinguished from the ‘subjective intention’ of coordinating parties, with the latter ‘not decisive for the purposes of the application of Article 101(1)’.^[194] Under Article 102, while ‘proof of an anti-competitive intention’ may constitute ‘an indication of the nature and objectives pursued by the strategy of that undertaking and, on that basis, may be taken into account,’ the recent case-law reiterates that ‘demonstration of the existence of such an intention is neither necessary nor sufficient, in itself, to establish the existence of an abuse’.^[195] In Google Shopping, the Court described ‘the existence of any anticompetitive intent’ as ‘only one of a number of facts which may be taken into account,’^[196] and emphasised that there is ‘no obligation to establish the existence of such intent…in order to render Article 102 applicable’.^[197]

The substantive objection is not trivial, and imposing an across-the-board obligation under Articles 101 or 102 to demonstrate subjective anticompetitive intention in addition to establishing behaviour that is objectively harmful to competition would clearly be inappropriate. Yet as demonstrated by the case-law on predatory pricing, intention can play a valuable, and determinative, role in establishing antitrust liability in certain circumstances. Under the Akzo test,^[198] anticompetitive intention serves to confirm the unacceptability in antitrust terms of behaviour that is objectively ambiguous from a competition policy perspective.^[199] Intention evidence can usefully perform a similar function in regulated conduct cases, marking the dividing line between behaviour that is purely a problem within the context of a regulatory regime and behaviour with a dual character that also merits antitrust intervention (i.e. that reveals a ‘competition issue’). Here again, we come up against a similar difficulty to that encountered when discussing the use of an effects-based limiting principle under the dominance-plus approach, namely the limits of the existing case-law. Yet it could be argued that recognising another intention-based exception where this possibility is already established in the jurisprudence is less radical than requiring a proper effects-and-causation analysis. In any event, these elements are not mutually exclusive.

The more procedural objection to making liability hinge on evidence of an anticompetitive strategy relates to the difficulties of demonstrating a ‘guilty mind’ where defendants are typically large corporate entities. On the one hand, there is the problem of determining what counts as the intention of the undertaking as a whole; on the other, there is the inherent tension, outlined above, of construing evidence of a desire to enhance an undertaking’s own market share as evidence of an objectively anticompetitive intention.^[200]

Yet, practically speaking, this has not prevented the accumulation of significant evidence of a defendant’s deliberate efforts to evade or defraud its regulatory obligations in many of the non-compliance and gaming cases discussed earlier.^[201] Indeed, the frustrating aspect of the Court’s insistence on the formal irrelevance of intention is that it is hard to reconcile with what occurred in these cases. For instance, whereas the General Court in AstraZeneca maintained that ‘proof of the deliberate nature of the conduct and of the bad faith of the undertaking in a dominant position is not required,’^[202] the higher Court of Justice stressed that such a hypothetical scenario was ‘radically different from…the present case.’^[203] The Teva decision, similarly, is imbued with the language of (bad) intention: the Commission found that the defendant ‘aimed at,’ ‘consciously,’ ‘intentionally’ and ‘clearly pursued’ its anticompetitive divisional patent applications.^[204] The decision formally maintained that there is ‘no requirement’ to establish anticompetitive intention,^[205] however, and cited Google Shopping to the effect that ‘the existence of an intention to compete on the merits, even if it were established, could not prove the absence of abuse’.^[206] Yet it is essentially impossible to envisage a plausible ‘good faith’ version of regulatory non-compliance or gaming as a competition law theory of harm. We have thus another example of the problem that so concerned Advocate General Wahl in his pivotal Opinion in Intel, with the Court saying one thing but doing quite another in practice.^[207] The policy question is whether to makes sense for EU competition law, in essence, to keep its options open by maintaining the fantasy that motivation is not a key element of the theory of harm in such cases; a question that pits legal principle against a desire to preserve the practical effectiveness of competition law to the greatest extent possible.

A focus on strategic anticompetitive intention in cases where the theory of harm is tied up with the question of regulatory compliance has a further aspect to commend itself: it could arguably provide a more satisfactory explanation for the problematic cases where a defendant is held to breach competition law despite its compliance with regulatory obligations that strongly influence its supposedly anticompetitive conduct. We identified two objections to this jurisprudence: the possible unfairness to defendants faced with the responsibility to correct policy choices made by a regulator, and the fact that it gives insufficient weight to how regulation affects ‘competition on the merits’ in the sector concerned. Yet these objections are less compelling if would-be ‘good faith’ compliance is not all that it seems. Where there is evidence that a defendant is acting in a strategically anticompetitive manner in its interaction with the regulatory framework, it is manifestly less unfair to pursue such behaviour under competition law. Evidence of anticompetitive intention can also provide an effective rejoinder to the claim that the regulatory regime demarcates the boundaries of ‘meritorious’ or permissible competition within a market. Of course, such an approach again departs from the orthodox ‘objectivity’ of competition assessment and may be criticised on this basis. Yet as a compromise between the two poles of deferring to and simply ignoring the coercive impact of regulation on private market behaviour, focusing on what the regulated defendant itself sought to achieve through its interaction with the regulatory framework arguably provides a reasonable limiting factor.

VI. Conclusion

This article began with a simple (perhaps simplistic) dichotomy between competition law, which promotes undistorted competition, and other market regulation, which has a greater or lesser capacity to distort competition in service of alternative policy goals. Yet Ezrachi has criticised the ‘pretence of purity’ that competition law adopts,^[208] and even if antitrust liability professes to be formally unrelated to compliance with other regulatory norms, the presence of regulation can have a profound effect on the task of competition assessment in practice.

Regulation and the diverse public interest values that it represents are a constant feature of modern markets. Context-specific competition enforcement necessarily absorbs and reflects these values to some degree, to the extent that it works within the ‘distortions’ of competition that follow from regulation. In the absence of deliberate efforts by policymakers to make competition impossible, the competition rules rarely cede jurisdiction to other regulatory norms, and whereas competition law may tolerate the presence of regulation that generates market power for certain regulated actors, it tends to be unforgiving of autonomous behaviour that misuses such market power. Yet it is rare for competition assessment to disregard the regulatory context; where such cases do arise, this article has argued for a more nuanced understanding of what ‘normal’ competition looks like in such circumstances.

Yet regulation is not always just ‘part of the scenery’ in competition cases. Efforts to escape or frustrate regulation can constitute standalone antitrust infringements if carried out by an entity falling within the scope of competition law, whether due to dominance or acting in concert. This article has argued that the apparent outsourcing of antitrust liability standards within the non-compliance and regulatory gaming case-law provides an effective vehicle by which to interrogate and better appreciate the interrelationship between competition law and other regulation. One way to understand these cases is to see the regulatory framework as setting the limits of fair competition in a marketplace, establishing the ‘rules of the game’ that all market actors can reasonably be expected to respect. While simple non-compliance or gaming of regulatory norms is arguably not a competition issue as such, it more plausibly becomes one where such conduct is the direct cause of appreciable anticompetitive effects. Another way to see these cases is as involving the ‘weaponisation’ of regulatory norms to anticompetitive ends as part of a deliberate strategy by regulated firms. From this perspective, the key antitrust consideration is what the defendant intends to achieve through its interaction with the regulatory framework. These perspectives are not mutually exclusive, moreover, and each finds some reflection in the existing case-law on non-compliance and regulatory gaming. Yet in many instances, the true explanation for punishing this conduct through competition enforcement is more pragmatic than principled: using competition law allows us to sidestep limitations or gaps within the regulatory framework itself. These cases thus involve an almost complete convergence between the goals of competition law and other market regulatory norms, and—for better or worse—take the legitimate scope of competition law to its outer limits.

1. * Law School, London School of Economics. Email: N.M.Dunne@lse.ac.uk. Thank you to LSE colleagues who gave helpful comments on an earlier draft of this presentation at a staff seminar, in particular Tarun Khaitan and Emmanuel Voyiakis; to two anonymous referees who similarly provided very useful feedback; and to Current Legal Problems editor Deni Mantzari for her advice, support and expert shepherding of this piece through the review process.

   See, e.g., Cases C-322/81, Michelin EU:C:1983:313, para. 10, C‑42/21 P, Baltic Rail EU:C:2023:12, para. 86 and C‑252/21, Meta Platforms EU:C:2023:537, para. 46. Also recital (11) of Regulation (EU) 2022/1925 on contestable and fair markets in the digital sector (Digital Markets Act), OJ L 265/1, 12.10.2022. ↑

2. As explained in section II, the term ‘regulation’ can be defined as broadly as ‘all forms of economic or social influence’ or as narrowly as ‘a specific set of commands’ imposed by the state: see text accompanying fns. 26 and 27. ↑

3. As the General Court recognised in Case T-168, GSK EU:T:2006:265, para. 104. ↑

4. See, for example, the OECD’s Competition Assessment Toolkit (2019), which is designed to assist governments in identifying and reducing barriers to competition and other market distortions which may stem from state regulation. ↑

5. Article 7 TFEU states that ‘[t]he Union shall ensure consistency between its policies and activities, taking all of its objectives into account and in accordance with the principle of conferral of powers’. There is, however, limited practice applying the provision: N. Nic Shuibhne, ‘Deconstructing and reconstructing Article 7 TFEU’ in F. Ippolito, M. E. Bartoloni and M. Condinanzi (eds.), The EU and the Proliferation of Integration Principles under the Lisbon Treaty (Routledge, 2018). ↑

6. See e.g. P. Ibáñez Colomo, The New EU Competition Law (Hart Publishing, 2023), Chapter 3. ↑

7. See e.g. E. Fox and E. Healey, ‘When the State Harms Competition—The Role for Competition Law’ 79 Antitrust law Journal 769 (2014). ↑

8. See T. Khaitan and S. Steel, ‘Theorising Areas of Law: A Taxonomy of Special Jurisprudence’ 28 Legal Theory 325 (2022), 329-34. ↑

9. This debate has several dimensions: a normative component, seeking to determine the optimal goals (e.g. E. Deutscher, ‘The competition-democracy nexus unpacked—competition law, republican liberty, and democracy’ 41 YEL 197 (2022)); an explicatory aspect, seeking to describe how the competition rules absorb and reflect different policy objectives (e.g. A. Ezrachi, ‘Sponge’ 5 JAE 49 (2017)); and an empirical aspect, seeking to identify the goals actually pursued in enforcement practice (see, e.g., M. Iacovides and K. Stylianou, ‘The new goals of EU competition law: sustainability, labour rights, and privacy’ 3 European Law Open 587 (2024)). ↑

10. See fn. 1. ↑

11. O. Odudu, ‘The Meaning of Undertaking within Article 81 EC’ 7 Cambridge Yearbook of European Legal Studies 211 (2005), 212. ↑

12. See Opinion of Advocate General Maduro in Case C-205/03 P, FENIN EU:C:2005:666, para. 26, on the distinction between ‘the power of the State’ which is subject to ‘democratic control,’ and the actions of ‘economic operators’ which fall within the purview of competition law. ↑

13. Article 3(3), TEU. ↑

14. Council Regulation 139/2004 of 20 January 2004 on the control of concentrations between undertakings (OJ L 24/1, 29.1.2004). ↑

15. See e.g. discussion in Case C-307/18, Generics (UK) EU:C:2020:52, paras. 87, 90 and 111 (referring to Article 101) and para. 152 (referring to Article 102). ↑

16. Case C-85/86, Hoffmann La Roche EU:C:1979:36, para. 91; also use of the term in C‑252/21, Meta Platforms EU:C:2023:537, para. 47. ↑

17. European Commission, Guidelines on the application of Article 102 of the Treaty on the Functioning of the European Union to abusive exclusionary conduct by dominant undertakings (hereafter ‘Draft Article 102 Guidelines’), August 2024, paras. 47-57. Discussing the role the concept plays within contemporary case-law, see P. Ibáñez Colomo, ‘Competition on the Merits’ 61 CMLRev 387 (2024). ↑

18. So, for example, the need to ‘meet competition’ is generally not considered an acceptable defence to prima facie restrictive behaviour, see e.g. Case T-671/19, Qualcomm EU:T:2024:626, paras. 598-99. ↑

19. The principal exception is the prohibition on hard-core cartel behaviour, where contextual analysis is ‘limited to what is strictly necessary’ to understand the anticompetitive nature of the conduct: Case C‑373/14 P, Toshiba EU:C:2016:26, para. 29. ↑

20. See e.g. Cases C-228/18, Budapest Bank EU:C:2020:265 and C-413/14 P, Intel EU:C:2017:632. ↑

21. Provided for formally by the Article 101(3) TFEU exception, with the possibility of ‘objective justification’ read into the Article 102 TFEU jurisprudence, see e.g. Case C‑209/10, Post Danmark EU:C:2012:172. ↑

22. For instance, specific competition rules exist for agriculture pursuant to Articles 42 and 43 TFEU. ↑

23. See discussion of the distinction between ex ante and ex post enforcement in Case C-449/21, Towercast EU:C:2023:207. ↑

24. Case C-376/20 P, CK Telecoms EU:C:2023:561, paras. 84-86. ↑

25. While there is scope to agree to behavioural remedies under the EUMR, this approach is disfavoured, at least in principle: Commission notice on remedies acceptable under Council Regulation 139/2004 (OJ C 267/1, 22.10.2008). ↑

26. R. Baldwin, M. Cave and M. Lodge, Understanding Regulation: Theory, Strategy and Practice (2^nd ed., Oxford University Press, 2011), p.3. ↑

27. Regulation need not adopt a classic top-down ‘command and control’ form to come within this understanding, but pure self-regulation by undertakings is treated simply as economic activity within the scope of the competition rules in EU law, see e.g. the professional rules in Case C-309/00 Wouters EU:C:2002:98 and the industry reorganisation efforts in Case C-209/07 BIDS EU:C:2008:643. For discussion of ‘decentred’ approaches to regulation, see J. Black, ‘Decentring Regulation: Understanding the Role of Regulation and Self-Regulation in a ‘Post-Regulatory’ World’ 54 Current Legal Problems 103 (2001). ↑

28. Indeed, viewed from a regulatory theory perspective, competition law can be seen as a variety of regulation that falls within the broad category of ‘market-harnessing controls’ (Baldwin et al., fn. 26 above, p.114-16). The competition jurisprudence, however, distinguishes the competition rules from ‘other’ regulation that may impact market dynamics, and that is the approach adopted within this piece. ↑

29. Digital Markets Act, Recital 11 (see also cases in fn. 1). ↑

30. For example, Directive 2019/944 on common rules for the internal market for electricity (OJ L 158/125, 14.6.2019), recital (2), included as its aims: ‘to deliver real choice for all Union final customers, [] new business opportunities, competitive prices, efficient investment signals and higher standards of service, and to contribute to security of supply and sustainability.’ ↑

31. Case C-280/08 P, Deutsche Telekom EU:C:2010:603. ↑

32. Opinion in Case Deutsche Telekom EU:C:2010:212, para. 21. ↑

33. Case C-280/08 P, Deutsche Telekom EU:C:2010:603, paras. 80-84; reaffirmed in Case C-220/24, Regia Autonomă Aeroportul Internaţional ‘Avram Iancu’ Cluj v Consiliul Concurenţei, EU:C:2025:124, para. 27. ↑

34. Indeed, in Case C-382/22 P, Cathay Pacific Airways v Commission EU:C:2026:129, para. 125, the Court of Justice stated that the State action defence ‘has been only partially accepted’ in its jurisprudence. ↑

35. C-280/08 P, Deutsche Telekom EU:C:2010:603, para.81. ↑

36. For a recent example where the State action defence actually succeeded on the facts, see Case T-136/19, Bulgarian Energy Holding EU:T:2023:669. ↑

37. Verizon Communications, Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 US 398 (2003). The complexities of the position within US law are discussed in H. Shelanski, ‘The Case for Rebalancing Antitrust and Regulation’ 109 Michigan Law Review 683. ↑

38. Case T-398/07, Spain v. Commission EU:T:2012:173, para. 55. ↑

39. Opinion in Case Deutsche Telekom EU:C:2010:212, para. 25. ↑

40. In Case C-117/20 bpost EU:C:2022:202, the Court recognised an ‘objective of general interest’ in ensuring the effective application of competition law and other market regulation in parallel, ‘since they are pursuing [] distinct legitimate objectives’ (para.50, emphasis added). ↑

41. See, e.g., Cases C-453/99 Courage and Crehan EU:C:2001:465, para. 26 and C-74/14 Eturas EU:C:2016:42, paras. 35-37. ↑

42. Case C-280/08 P, Deutsche Telekom EU:C:2010:603. ↑

43. Case C-209/07, BIDS EU:C:2008:643. ↑

44. As was the case in support of the Siemens/Alstom merger, which was nonetheless prohibited by the Commission (M.8677, 6 February 2019). ↑

45. See Ursula von der Leyen, European Commission Political Guidelines 2024-29: ‘I believe we need a new approach to competition policy,…more supportive of companies scaling up in global markets’. See also Barbara Moens, ‘EU to relax merger rules in bid to create ‘European champions’ Financial Times, 16 April 2026. ↑

46. Cases T‑321/05, AstraZeneca EU:T:2010:266, para. 356; and T‑814/17, Baltic Rail EU:T:2020:545, para. 189. ↑

47. Case C-209/07, BIDS EU:C:2008:643, para.21. ↑

48. Case C‑307/18, Generics (UK) and Others EU:C:2020:52. ↑

49. The Commission may take account of evidence of how the merged entity intends to act on the market going forward, but it places greater emphasis on whether that behaviour is anticipated to cause anticompetitive effects in fact: see e.g. Case M.11382—AGCO/TRIMBLE/JV (decision of 25.03.2024), for an example where the Commission concluded that distortive post-merger behaviour would not significantly impede effective competition in the circumstances. ↑

50. As the US Supreme Court said in Bell Atlantic Corp. v. Twombly 550 U.S. 544 (2007), ‘resisting competition is routine market conduct’. ↑

51. This is a core tenet of competition policy, with roots that can be traced back to Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations (1776). For an argument that Smith’s influence should extend beyond this ‘libertarian’ vision, see S. Makris, ‘A Smithian Political Economy Approach for the Competition Law of the 21st Century’ 88 Modern Law Review 712 (2025). ↑

52. Article 23, Regulation 1/2003. ↑

53. Commission Guidelines on the method of setting fines imposed pursuant to Article 23(2)(a) of Regulation No 1/2003 (OJ C 210/2, 1.9.2006). ↑

54. C‑457/10 P, AstraZeneca EU:C:2012:770, para. 132; reaffirmed in Case C-220/24, Aeroportul Cluj EU:C:2025:124, para. 28 ↑

55. Cases C-295/12 P, Telefónica EU:C:2014:2062, para.133 and C-220/24, Aeroportul Cluj EU:C:2025:124, para. 28. ↑

56. Case C-280/08 P, Deutsche Telekom EU:C:2010:603, para. 90, applying the principle developed in Case C‑344/98, Masterfoods EU:C:2000:689, para. 48. ↑

57. On the basis that only the Commission is empowered to find that there has been no breach of Article 102, whereas national regulators can only find breach or discontinue proceedings: Case C-375/09, Tele2 Polska EU:C:2011:270. ↑

58. Case C-280/08 P, Deutsche Telekom EU:C:2010:603, para. 89. ↑

59. Case C‑457/10 P, AstraZeneca EU:C:2012:770, para. 132. ↑

60. Case C-238/05 Asnef-Equifax EU:C:2006:734, para. 63. ↑

61. Guidelines on the assessment of non-horizontal mergers under the Council Regulation on the control of concentrations between undertakings (OJ C 265/6, 18.10.2008), para. 46. ↑

62. M.8124—Microsoft/LinkedIn, 6 December 2016. ↑

63. European Commission Press Release, ‘Mergers: Commission clears acquisition of Fitbit by Google, subject to conditions’ 17 December 2020. ↑

64. Case C-255/22 P Orlen EU:C:2024:790, para. 96. ↑

65. Ibid., para. 96. ↑

66. Ibid., para. 97. ↑

67. Ibid., para. 96. ↑

68. Ibid., para. 95 (emphasis added). ↑

69. To use the language of AG Rantos in his Opinion in Case C-255/22 P, Orlen EU:C:2024:466, para. 58. ↑

70. See fn. 9 above. ↑

71. Cases C-204/00 P etc., Aalborg Portland EU:C:2004:6, para. 338. ↑

72. Case C-117/20, bpost EU:C:2022:202, paras. 28-37. ↑

73. Ibid., para. 41. ↑

74. Ibid., para. 49. ↑

75. Ibid., para. 51. ↑

76. Case C-333/21, Superleague EU:C:2023:1011, para. 130 (discussing specifically Article 102 TFEU). ↑

77. Ibid., para. 166 (discussing specifically Article 101 TFEU). ↑

78. See text accompanying fn. 33 above. ↑

79. Case C-280/08 P, Deutsche Telekom EU:C:2010:603, para. 224. ↑

80. Ibid. ↑

81. To quote the defendant’s argument in Case C-280/08 P, Deutsche Telekom EU:C:2010:603, para. 205. ↑

82. Both the defendant and its rivals used revenues from call services to cross-subsidise their loss-making activity in the provision of line access services. The Commission, however, applied the margin squeeze test only to the defendant’s wholesale and retail-level line access activities. This approach was approved by the Court of Justice, which held that Commission was entitled to proceed as if tariff rebalancing had occurred as required by EU telecommunications law, even though the German regulator had failed to implement tariff rebalancing for social policy reasons (to protect access to telephone lines for poorer consumers). ↑

83. Case C‑165/19 P, Slovak Telekom EU:C:2021:239, para. 57; and C‑42/21 P, Baltic Rail EU:C:2023:12, para. 88. ↑

84. See Cases C-220/24, DB Station & Service EU:C:2022:832, para. 82, T-136/19, BEH EU:T:2023:669, paras. 784 & 961, and C-220/24, Aeroportul Cluj EU:C:2025:124, para. 31. ↑

85. The Bundeskartellamt applied only domestic competition law, a choice that has been criticised as potentially inconsistent with its obligations under Regulation 1/2003. By the time the case made it to the Court of Justice, the Court spoke in terms of the application of EU competition law. ↑

86. Regulation 2016/679 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data (OJ L 119/1, 04.05.2016). ↑

87. C‑252/21, Meta Platforms EU:C:2023:537, paras. 44-46. ↑

88. Ibid., para. 47 (emphasis added). ↑

89. The Court stated that, ‘in the context of the examination of an abuse of a dominant position by an undertaking on a particular market, it may be necessary for the competition authority…also to examine whether that undertaking’s conduct complies with rules other than those relating to competition law, such as’—but by implication not limited to—’the rules on the protection of personal data laid down by the GDPR’: ibid., para. 48 (emphasis added). ↑

90. Ibid., para. 50. ↑

91. Ibid., para. 51. ↑

92. See text accompanying fn. 75 above. ↑

93. C‑252/21, Meta Platforms EU:C:2023:537, para. 55. ↑

94. See e.g. Case C-205/03 P, FENIN EU:C:2006:453. ↑

95. See e.g. Case T-155/04, SELEX EU:T:2006:387. ↑

96. Case C-359/95 P Ladbroke Racing, para. 33. ↑

97. Case C-333/21, Superleague EU:C:2023:1011, para. 183. ↑

98. Case C-309/99, Wouters EU:C:2002:98, para. 9. ↑

99. See fns. 94 and 95 above. ↑

100. Contrast Case C‑333/21, Superleague EU:C:2023:1011, paras. 183-188, where the Court of Justice rejected the application of the Wouters exception to an organisation governed solely by private law (albeit not on that basis expressly). ↑

101. See fns. 34 and 35 above. ↑

102. Draft Article 102 Guidelines, paras. 167-71. ↑

103. Case C-233/23, Android Auto EU:C:2025:110, para. 75. See also C‑42/21 P, Baltic Rail EU:C:2023:12, which impliedly accepted that health and safety regulatory requirements could provide an objective justification for a de facto refusal to grant access to infrastructure, although the claimed necessity was deemed to be pretextual on the facts. ↑

104. Draft Article 102 Guidelines, para. 168. ↑

105. Ibid. ↑

106. As in e.g. Case C‑42/21 P, Baltic Rail EU:C:2023:12. ↑

107. As in e.g. Case C‑457/10 P, AstraZeneca EU:C:2012:770. ↑

108. Discussing the impact of privatisation specifically, see Case C‑245/24, Lukoil Bulgaria EU:C:2025:987. ↑

109. As in the E.ON/GDF pipeline cartel (Case COMP/39.401—E.ON/GDF, decision of 8 July 2009). ↑

110. As in Case C-331/21, EDP—Energias de Portugal EU:C:2023:812; in EDP, this fact was treated as evidence of the anticompetitive object of the arrangement contrary to Article 101(1) (see para. 102). ↑

111. As acknowledged in Case C‑252/21, Meta Platforms EU:C:2023:537, para. 47. ↑

112. See, e.g., Cases C-413/14 P, Intel EU:C:2017:632 and C-333/21, Superleague EU:C:2023:1011. ↑

113. Case C‑252/21, Meta Platforms EU:C:2023:537, para. 47. ↑

114. Case C‑48/22 P, Google Shopping EU:C:2024:726, para. 224. ↑

115. Ibid., para. 225 (the Court expressly rejected Google’s argument that the Commission should have conducted a counterfactual analysis to find causation, see para. 228). ↑

116. Including in the fields of airports, telecommunications, energy and rail transport. ↑

117. Case C‑165/19 P, Slovak Telekom EU:C:2021:239. ↑

118. See, e.g., Cases C‑501/06 P etc., GSK EU:C:2009:610, paras. 102-104. ↑

119. For a recent study of such missing case law, see M. Heim, ‘The curious case of the European Commission’s missing antitrust jurisprudence: lessons from abandoned Article 102 investigations’ 16 JECLAP 285 (2025). ↑

120. Case M.8124—Microsoft/LinkedIn, 6 December 2016, para. 177. ↑

121. See fn. 55. ↑

122. As in Cases 39.525—Telekomunikacja Polska (22 June 2011), Case C‑165/19 P, Slovak Telekom EU:C:2021:239 and C‑42/21 P, Baltic Rail EU:C:2023:12. ↑

123. S. Dogan & M. Lemley, ‘Antitrust Law and Regulatory Gaming’ 87 Texas Law Review 685 (2009). ↑

124. Case COMP/A.37.507/F3—AstraZeneca, 15 June 2005; upheld on appeal in Cases T‑321/05, AstraZeneca EU:T:2010:266 and C-457/10 P, AstraZeneca EU:C:2012:770. ↑

125. Case AT.40588—Teva, 31 October 2024. ↑

126. Case AT.40394—Aspen Pharma, 10 February 2021. ↑

127. Case AT.40134—AB InBev Beer Trade Restrictions (29 June 2016). ↑

128. Case C-179/16, F. Hoffmann-La Roche and Others EU:C:2018:25. ↑

129. Case AT.40178—Car Emissions, 8 July 2021. ↑

130. Including the Commission decisions in Cases AT.39226, Lundbeck and AT.39612, Servier and the preliminary ruling in Case C-307/18, Generics (UK) EU:C:2020:52. ↑

131. See e.g. M. Moore & D. Tambini (eds.), Regulating Big Tech. Policy Response to Digital Dominance, OUP (2022). ↑

132. See e.g. M. Gal & O. Aviv, ‘The Competitive Effects of the GDPR’ 16 JCLE 349 (2020) and C. B. Frey & G. Presidente, ‘Privacy regulation and firm performance: Estimating the GDPR effect globally’ 62 Economic Inquiry 1074 (2024). ↑

133. Recital 9, GDPR, which states that uneven implementation of the previous EU data protection regime could ‘distort competition’. ↑

134. See fn. 91. ↑

135. As in Cases C‑457/10 P, AstraZeneca EU:C:2012:770 and C-591/16 P, Lundbeck EU:C:2021:243, for example. ↑

136. As in the E.ON/GDF gas pipeline cartel, for example (see in particular, the judgment of the General Court on appeal, Case T-370/09 GDF Suez v Commission EU:T:2012:333. ↑

137. Ezrachi, fn. 9 above. ↑

138. I. Lianos, ‘Polycentric Competition Law’ 71 Current Legal Problems 161 (2018). ↑

139. For a more detailed critique of the inadequacy of this approach, see N. Dunne, Competition Law and Economics Regulation, CUP (2015), Chapter 4. ↑

140. In Deutsche Telekom, for instance, the Court dismissed the significant impact of the national regulatory framework on the defendant’s behaviour on the rather blithe ground that it could have sought to petition the regulator to adopt more cost-reflective price regulation (in spite of evidence that the regulator had deliberately chosen to prioritise social solidarity over efficiency in its regulatory policymaking). For further critique of this case from a fairness perspective, see Dunne, fn. 139, pp. 224-27. ↑

141. This was a significant theme within Deutsche Telekom: in such circumstances, the Court took a generous view of the Commission’s discretion to initiate competition law proceedings against the undertaking or infringement proceedings against the Member State maintaining the anticompetitive regulation. ↑

142. See fn. 69 above. ↑

143. As the General Court remarked in Lundbeck, a pay-to-delay case, ‘it is in the public interest to eliminate any obstacle to economic activity which may arise where a patent was granted in error’: Case T-472/13, Lundbeck EU:T:2016:449, paras. 119, 390 and 487. This is also an issue in the background in Case C‑457/10 P, AstraZeneca EU:C:2012:770 and Case AT.40588—Teva, 31 October 2024 amongst other cases. ↑

144. As provides by Article 4 (independence), Article 4 (resources) and Articles 10 and 13-16 (remedies and fines) of Directive (EU) 2019/1 to empower the competition authorities of the Member States to be more effective enforcers and to ensure the proper functioning of the internal market (ECN+ Directive), OJ L 11/3, 14.1.2019. ↑

145. See fn. 74. ↑

146. Case C‑252/21, Meta Platforms, para. 54. ↑

147. Case C-117/20 bpost EU:C:2022:202, para. 49. ↑

148. See fn. 67. ↑

149. GDPR, recital (4). ↑

150. See fn. 15. ↑

151. Case C-377/20, Servizio Elettrico Nazionale EU:C:2022:379, para. 61. ↑

152. Draft Article 102 Guidelines, para. 45. ↑

153. Ibid., para. 60. ↑

154. For example, data protection rules limit the extent to which firms may profitably combine data-sets containing personal information, while environmental protection rules limit the extent to which firms can use cheaper but more polluting technologies. ↑

155. For example, EU employment law prevents firms from dismissing pregnant employees and requires the provision of paid leave. ↑

156. On regulatory compliance costs generally, see OECD, OECD Regulatory Compliance Cost Assessment Guidance, OECD Publishing (2014). ↑

157. For discussion of the level-playing field notion generally, see e.g. M. Gillis, ‘Let’s Play?: An Examination of the ‘Level Playing Field’ in EU Free Trade Agreements’ 55 Journal of World Trade Law 715 (2021). ↑

158. Case C‑165/19 P, Slovak Telekom EU:C:2021:239, paras.58-60, and C‑42/21 P, Baltic Rail EU:C:2023:12, para.89. See also Case T-136/19, BEH EU:T:2023:669, para. 873. ↑

159. Commission Guidelines on the applicability of Article 101 to horizontal co-operation agreements (OJ C 259, 21.7.2023). ↑

160. Ibid., para. 520. ↑

161. Ibid., para. 564. ↑

162. Ibid., para. 520. ↑

163. C‑252/21, Meta Platforms EU:C:2023:537, para. 51. ↑

164. As in Cases C‑457/10 P, AstraZeneca EU:C:2012:770 and AT.40588—Teva, 31 October 2024 ↑

165. Although coordination or dominance does not attract antitrust liability itself, the jurisprudence emphasises that such firms have heightened obligations: coordinating firms must ensure that each still ‘determine[s] independently the policy which he intends to adopt on the…market’ (Case C-40/73, Suiker Unie EU:C:1975:174, para. 173) while dominant firms have a distinctive ‘special responsibility’ to avoid distorting competition (Case C-322/81, Michelin EU:C:1983:313, para. 10). ↑

166. The argument would be that such firms are expected to comply both with their regulatory obligations and to refrain from practices that might reasonably be considered at odds with the broad thrust of a regulatory scheme. An analogy might be drawn to the concept of potential competition, which is established where there are ‘real and concrete possibilities’ of new entry (see e.g. Case C‑307/18, Generics (UK) EU:C:2020:52, para. 36). This standard can be met even if there is considerable ambiguity as to whether new entrants can lawfully compete in a market given existing regulatory barriers to entry. In pay-to-delay cases, for instance, competition law does not tolerate agreements to exclude potential competitors even if there is a reasonable chance that patent law might facilitate this result: for instance, in Case T‑472/13, Lundbeck EU:T:2016:449, paras. 121-31, the court found potential competition where there was only a 50-60% chance that the would-be competitor could lawfully enter the market concerned. ↑

167. Coordination falling within Article 101 could also act as a jurisdictional trigger for the application of competition law; a ‘collusion-plus’ approach would then treat coordinated non-compliance or gaming efforts as a restriction of competition. An example of this arose in the Car Emissions cartel, see fn. 129. ↑

168. The German competition authority also drew on national case-law on consumer protection to support its approach to the concept of market abuse. ↑

169. Ezrachi, fn. 9. ↑

170. See e.g. A. Ayal, ‘The market for bigness: economic power and competition agencies’ duty to curtail it’ 1 JAE 221 (2013); L. Khan, ‘The Ideological Roots of America’s Market Power Problem’ 127 Yale L.J. F. 960 (2018); and A. Gerbrandy & P. Phoa, ‘The Power of Big Tech Corporations as Modern Bigness and a Vocabulary for Shaping Competition Law as Counter-power’ in M. Bennett, H. Brouwer, & R. Claassen (eds.), Wealth and Power: Philosophical Perspectives, Taylor & Francis (2022). ↑

171. See fn. 165. ↑

172. Case C-85/86, Hoffmann La Roche EU:C:1979:36, para. 91. ↑

173. Cases C‑165/19 P, Slovak Telekom EU:C:2021:239, para. 57, and C‑42/21 P, Baltic Rail EU:C:2023:12, para. 88. ↑

174. Case C‑252/21, Meta Platforms EU:C:2023:537, para.44. ↑

175. ‘[P]ublic authorities can legitimately choose complementary legal responses to certain conduct that is harmful to society through different procedures forming a coherent whole so as to address different aspects of the social problem involved’: Case C-117/20, bpost EU:C:2022:202, para. 49. ↑

176. Exemplified by Case C-280/08 P, Deutsche Telekom EU:C:2010:603. ↑

177. Under the ECN+ Directives, national competition authorities must have the ability to impose behavioural and structural remedies on defendants (Art.10), plus fines of up to at least 10% of the annual worldwide turnover of the undertakings concerned (Art.15) for competition law violations. These provisions mirror the powers of the Commission under Regulation 1/2003. ↑

178. See Directive 2014/104/EU on certain rules governing actions for damages under national law for infringements of the competition law provisions of the Member States and of the European Union (OJ L 349/1, 5.12.2014). ↑

179. See e.g. Case C-413/14 P, Intel EU:C:2017:632, para. 138. ↑

180. See e.g. the discussion in Case C-307/18, Generics (UK) EU:C:2020:52. ↑

181. Surveying the case-law, see P. Ibanez Colomo, ‘Anticompetitive Effects in EU Competition Law’ 17 JCLE 209 (2020). ↑

182. This requires it to be demonstrates that coordination ‘has as its actual or potential effect the prevention, restriction or distortion of competition, which must be appreciable’ (Superleague, para.169). Thus it is sufficient that the conduct merely limits potential (as opposed to existing) competition, although this does require it to be established that ‘real and concrete possibilities’ for new market entry exist (Cases C‑307/18, Generics (UK) EU:C:2020:52, para. 36). ↑

183. Relied upon, inter alia, in Roche & Novartis (fn. 128), Car Emissions (fn. 129) and various ‘pay-to-delay’ cases. ↑

184. Case C-333/21, Superleague EU:C:2023:1011, para. 165. ↑

185. Case C-377/20, Servizio Elettrico Nazionale EU:C:2022:379, para. 50. ↑

186. Ibid., para. 54. ↑

187. See fn. 115 above. ↑

188. See e.g. Case C-333/21, Superleague EU:C:2023:1011, para. 163. ↑

189. See e.g. discussion in Case C-377/20, Servizio Elettrico Nazionale EU:C:2022:379, paras. 53-54. ↑

190. Case C-413/14 P, Intel EU:C:2017:632, para. 139; followed in Case C-680/20, Unilever EU:C:2023:33. ↑

191. Case C‑48/22 P, Google Shopping EU:C:2024:726, para. 240. ↑

192. Ibid., para. 254. ↑

193. Case C-333/21, Superleague EU:C:2023:1011, para. 167. ↑

194. Ibid. ↑

195. Case C-680/20, Unilever EU:C:2023:33, para. 45 (emphasis added). ↑

196. Case C‑48/22 P, Google Shopping EU:C:2024:726, para. 255. ↑

197. Ibid., para. 256. ↑

198. Named for Case C-62/86 Akzo v Commission EU:C:1991:286. ↑

199. The Akzo test is an off-shoot of the Areeda-Turner test, which was developed to provide an objective means to determine when low prices—typically considered desirable from a competition policy perspective—should nonetheless be found abusive by virtue of their exclusionary effects (see P. E. Areeda & D. F. Turner, ‘Predatory pricing and related practices under section 2 of the Sherman Act’ 88 Harvard Law Review 697 [1975]). The Areeda-Turner test posits that prices below average variable cost (AVC, a proxy for margin cost) should deemed conclusively abusive on the basis that no rationale firm would continue to produce if it cannot obtain prices at or above AVC, so that its choice to do so can be taken to disclose sufficient evidence of anticompetitive intention. The Akzo case extended the Areeda-Turner test to prices above AVC but below average total cost where there is evidence that such prices are ‘determined as part of a plan for eliminating a competitor’ (C-62/86, para.72), on the basis that such prices have the capacity to exclude as-efficient competitors, while the actual intention evidence serves to counter any potentially innocent explanations for the pricing behaviour. ↑

200. See fn. 51 above and accompanying text. ↑

201. Including AstraZeneca, Teva, Aspen Pharma, Telekom Polska, Lundbeck and the quite fantastical case of Baltic Rail, where the defendant opted to destroy 19km of its own rail infrastructure rather than share it with a freight rail competitor under the EU rail liberalisation framework. ↑

202. See fn. 46 above. ↑

203. Case C‑457/10 P, AstraZeneca EU:C:2012:770, para. 99. ↑

204. Case AT.40588—Teva, 31 October 2024. ↑

205. Ibid., para. 1033. ↑

206. Ibid., para. 1033, citing Case T-612/17, Google Shopping EU:T:2021:763, para. 257. ↑

207. Opinion in Case C-413/14 P, Intel v Commission EU:C:2016:788, para. 66. ↑

208. Ezrachi, fn. 9, p.50. ↑

1. Introduction

What roles do regulatory norms play in EU competition law assessment? The question matters both for the consistency of the supervisory framework for economic activity in the EU, and because it helps us to better understand the legitimate concern of the competition rules, a key theme in contemporary competition policy debates. Competition law (also known as antitrust) aims to promote and protect effective competitive behaviour by economic actors – the jurisprudence uses the term ‘undistorted’^[1] competition to describe the ultimate policy objective. Regulation, a broader and more ambiguous concept,^[2] often applies to and constrains the same market conduct, though typically in pursuit of alternative or additional policy objectives. Regulation can enhance opportunities for effective competition or limit the scope for abusive market behaviour; but it can also ‘distort’^[3] competition in ways that raise concerns from a competition policy perspective, whether by diminishing overall competitive dynamics or by facilitating or incentivising anticompetitive conduct.^[4] EU law has long endorsed the concurrent application of the competition rules to regulated firm behaviour, an approach that maximises the effectiveness of competition policy. Yet the jurisprudence struggles with a problem of internal coherence, seeking to protect the distinctiveness and primacy of the competition rules while also finding space for the increasing centrality of contextual analysis in contemporary competition enforcement. Moreover, applying competition law to behaviour that has already been affected by other regulatory norms creates its own challenges, including issues of legitimacy, consistency and proportionality. These questions are the focus of this piece.

In principle, the competition rules apply only to the autonomous behaviour of firms, and such conduct – whether taking the form of agreements, dominant firm practices or mergers – is prohibited only where it is demonstrated that anticompetitive effects – whether actual, potential or presumed – are likely to result. Yet where competition law is applied to regulated behaviour, there is a higher likelihood that suspect conduct has been influenced or even directed by the regulatory obligations and that any resulting market distortions can be attributed, wholly or partly, to the impact of the regulation. In such contexts, competition law assessment must consider not only whether and how the defendant’s behaviour has caused anticompetitive harm, but also the extent to which that behaviour and its effects were determined by the underlying regulatory framework. The greater the overlap between the scope of the regulatory obligations and the competition law theory of harm, the more complex this task becomes. Since competition law and regulation must co-exist as legitimate and socially valuable mechanisms of market control, there are arguments, both practical and legal,^[5] for ensuring that these separate regimes are interpreted and applied in a consistent, ideally mutually reinforcing, manner. But because this question typically arises where competition law is applied in markets already subject to ex ante regulation, there can be tension between preserving the effectiveness of competition law and its underlying goal of protecting undistorted competition, while simultaneously respecting other public interest values pursued by ‘distortive’ regulation.

A great deal has already been written about the relationship between competition law and other regulation, including much of my own work. This article seeks to make an original contribution by focusing on how substantive regulatory norms are accommodated within competition law analysis and what this tells us about competition law’s place within the wider market regulatory sphere. We leave aside the distinct questions of when competition law itself exhibits certain ‘regulatory’ characteristics,^[6] and of the appropriate competition policy response to anticompetitive State regulation.^[7] In doing so, the article adopts an essentially nonnormative, conceptual approach, which aims to work effectively within the confines of the existing jurisprudence rather than making more aspirational claims for wholesale change.^[8] Our analysis is not directly concerned with the goals of competition law, although there are plenty of articles of this sort about.^[9] But when considering how competition law interacts with other regulatory fields that pursue a more diverse range of policy objectives, it is impossible to avoid the question of what competition law itself seeks to achieve. Above, we described the goal of competition law broadly as the protection of ‘undistorted’ competition, an approach that, although not without its critics, finds support in recent jurisprudence and legislation.^[10] The article thus seeks to provide a more granular exploration of what we mean by undistorted competition, in a context where markets can never really be free of distortions, whether stemming from private or public sources. By considering the extent to which other regulatory norms can provide a conduit by which to reflect a wider range of public interest values within competition enforcement, while also recognising the legal and practical limits on this approach, we seek to obtain a better understanding of what Odudu termed ‘the legitimate scope’^[11] of competition law.

The article is structured as follows. Section II introduces and distinguishes competition law from other regulatory instruments of market supervision. Section III sets out and analyses the ‘orthodox’ understanding of the inter-relationship, which prioritises the independence and effectiveness of competition law as a mechanism of market control. Section IV discusses the more ‘contextual’ contemporary approach, considering the ways in which regulation may provide a relevant element of context within competition law analysis. Section V steps back, to consider the normatively-oriented question of how the ‘outsourcing’ of liability standards in competition law to other regulatory fields can be explained. It examines a variety of perspectives: the effectiveness of competition law as a regulatory enforcement mechanism; regulatory norms as a constraint on the acceptable parameters of competition on the merits; and regulatory (non-)compliance as a positive anticompetitive action. Section VI brings these disparate strands of analysis together and concludes.

2. Distinguishing competition law from ‘regulation’ as instruments of market supervision

A necessary starting point is to distinguish competition law from other forms of market regulation. Competition law comprises a set of legal rules that aim to ensure open, undistorted and fair competition in the internal market, by controlling the acquisition and exercise of market power (as distinct, inter alia, from public power).^[12] Within the framework of the EU treaties, the competition rules operate as important flanking provisions to secure the objective of ‘a highly competitive social market economy’.^[13] Yet most competition cases do not belabour this point. Instead, enforcement tends to be a relatively technical exercise aimed at identifying and proscribing various agreements (under Article 101 of the Treaty on the Functioning of the European Union, or TFEU), practices of dominant undertakings (under Article 102 TFEU) and mergers (under the EU Merger Control Regulation, or EUMR)^[14] that are deemed restrictive of competition in some legally relevant sense. A key theme within recent case-law is the notion of ‘competition on the merits’:^[15] although competition law is aimed at anticompetitive firm behaviour, the rules do not seek to prevent firms, even where they are dominant or acting in concert, from competing ‘on the merits’ (what is also termed ‘normal competition’^[16]). What amounts to normal or meritorious competition in an individual case is a question of competition law. It is thus determined by applying the legal tests and principles developed in the competition jurisprudence,^[17] instead of deferring to the behavioural or regulatory norms within a sector.^[18]

Contemporary EU competition law has several salient aspects for our purposes. First, competition assessment is, in almost every instance,^[19] a highly context-specific exercise: conduct is held to restrict competition within its market circumstances,^[20] while also taking account of any countervailing pro-competitive efficiencies that might be generated.^[21] Second, and somewhat by contrast, competition law is largely sector-indifferent in its application. With a few exceptions,^[22] the competition rules apply across all sectors, supervising vastly different industry practices under the same legal provisions. Finally, the competition rules governing anticompetitive agreements and dominant firm practices are largely ex post in application, serving to punish infringing behaviour after-the-fact.^[23] The EUMR, conversely, is prospective in analysis,^[24] but the object of the exercise is not (generally)^[25] to regulate the behaviour of newly-merged firms going forward, but rather to catch and proscribe potentially harmful concentrations before these occur. In this article, we focus on the impact of pre-existing regulation on the application of Articles 101 and 102 TFEU to market behaviour, while considering insights from merger control where appropriate.

The term regulation is less precise in its meaning. It can be defined as broadly as ‘all forms of economic or social influence’ that may stem from public or private sources, or as a narrowly as ‘a specific set of commands’ imposed by the State and pertaining to closely identified activities.^[26] For the purposes of this article, we adopt a relatively loose definition, referring to State-imposed (or at least State-endorsed)^[27] norms that, directly or indirectly, control or constrain market activity by economic actors, broadly understood. This encompasses both sector-specific regulation, which applies to market participants that are active only within particular sectors (such as telecommunications, energy, transport or digital platform services), and to regulatory regimes of more general application (such as data protection, health and safety, or environmental law).

Applying this definition, it can readily be seen that most markets today are subject to multiple layers of regulation that, to borrow the language of the competition jurisprudence, may ‘distort’ competitive outcomes to a greater or lesser extent. In some instances, it is obvious how overlap with competition law arises: for example, regulation may grant an economic actor special or exclusive rights that give it a position of market dominance, or regulation may mandate pro-competitive behaviour (such as imposing access requirements) or proscribe direct abuses of market power (through, for example, pricing controls). In other instances, the potential for interaction with competition law arises more obliquely, for example where regulation constrains a firm’s ability to maximise its profits or provides opportunities for exploitative or exclusionary behaviour.

Given the breadth of our understanding of regulation, and by implication the variety of regulatory models that it encompasses, it is impossible to neatly distinguish competition law from other forms of market regulation that may apply to the same conduct.^[28] Two distinctions of relevance might, however, be advanced. First, although the precise ‘goals’ of EU competition law are debated, broadly speaking the rules aim at a straightforward objective, namely ‘the protection of undistorted competition’.^[29] While a regulatory regime may seek to enhance effective competition, this need not be its only policy objective, and most of the examples of regulation that overlap with competition law within the case-law have pursued other regulatory goals additionally or alternatively,^[30] some of which may sit uncomfortably with the efficiency-focused perspective of the competition rules. Second, one of the necessary trade-offs that comes from the wide applicability of competition law is that its prohibitions are sketched in broad and abstract terms. This creates flexibility and scope for progressive development of the law to address emerging market problems, but such advantages arise at the expense of ex ante precision and certainty. Some regulatory regimes operate in a similar manner, yet regulation can be more prescriptive, identifying with greater exactness the specific regulated actors or the content of mandatory norms. Specification can assist enforcement, enabling the supervision of complex market problems. Yet where regulated entities are eager to resist regulation, a precisely defined and limited set of rules may facilitate strategic behaviour to escape or exploit regulatory controls, a problem considered in Section 5.

The broad scope of the competition rules coupled with the pervasiveness of regulation in modern markets leads to overlapping jurisdiction and potential conflict. The jurisprudence demarcating the inter-relationship of competition law and other regulation within the EU legal system is complex. One strand of case-law emphasises the independence and effectiveness of competition law as a mechanism of market control, reflecting an ‘orthodox’ understanding of the inter-relationship that prioritises but also effectively siloes competition law. Another strand embraces the more contextual approach that has come to the fore in the general competition case-law, recognising the extent to which regulation may feed into and even determine the substance of competition law assessment. While these jurisprudential currents are not inherently at odds, they reflect notably different visions of this inter-relationship. There are also points of inconsistency and even conflict within the case-law, which remain underexplored in the existing jurisprudence. In what follows, we set out first the apparently resolute ‘orthodox’ position (Section 3) and then the more nuanced contextual approach (Section 4). Doing so identifies several cases in which it becomes more difficult to reconcile the orthodox and contextual accounts, which are explored in Section 5.

3. The interaction of competition law and regulation: the orthodox view

Within EU law, the orthodox position is that the competition rules are essentially distinct from and unaffected by other regulatory norms that may apply to the same market behaviour. This approach is exemplified by Deutsche Telekom, in which the defendant was held to have breached Article 102 in circumstances where it had complied with its sector-specific regulatory obligations, where the sector regulator had either set or endorsed the pricing practices that comprised the abuse, and where the national regulatory framework pursued a legitimate public policy objective (social solidarity) by endorsing a pricing structure that the Commission subsequently held to be inefficient.^[31] An analogy of ‘two barriers’ to lawful market participation was suggested by the Advocate General in Deutsche Telekom: economic actors must comply with the overarching obligations imposed by the competition rules, while simultaneously discharging any parallel regulatory obligations.^[32] Typically, moreover, these legal requirements are enforced by different regulators, who each decide independently whether the norms falling within their jurisdictions have been breached. This orthodoxy has several important implications.

A. No ouster of competition law in the presence of regulation

First, the existence of regulation within a market does not generally oust the applicability of competition law to the same behaviour, even if the regulatory regime has a significant impact on the conduct or underlying level of competitiveness within the market.^[33] The principal exception, the ‘State action’ doctrine, is narrowly construed,^[34] applying only where anticompetitive conduct is either required by regulation or where the regulatory framework removes all scope for competitive (and, by implication, anticompetitive) behaviour.^[35] The logic behind this defence is that competition law regulates anticompetitive behaviour by firms and not merely uncompetitive market structures; but its narrowness in practice suggests a deep reluctance to decouple the abstract notion of competition from the concrete actions of competitors.^[36]

This can be contrasted with the approach in US antitrust law, which more readily embraces a rule of antitrust preclusion where ‘a regulatory structure designed to deter and remedy anticompetitive harm’ can be identified.^[37] There are at least three reasons which explain the distinctive position to be found in EU law. First, the competition rules are primary EU law, meaning that there is a formal obstacle within the EU’s hierarchical legal structure to allowing ‘special’ regulatory rules found in secondary or national legislation to take precedence over the ‘general’ competition rules.^[38] Second, even if regulation takes account, inter alia, of the goal of effective competition, the assumption is that it almost always pursues ‘objectives which differ from those of [EU] competition policy’.^[39] The policy objectives of one set of rules cannot be subsumed into the other; with the result that, to the extent we wish to pursue both sets of goals simultaneously, both sets of rules must remain in operation.^[40] Finally, EU competition law incorporates a remarkably strong principle of effectiveness, which has aided the progressive development and expansion of these rules in many areas.^[41] Limiting any rule of antitrust preclusion to only the most exceptional circumstances where regulation removes all scope for (anti)competitive conduct serves to protect the effectiveness of competition law by ensuring that these rules continue to apply in the largest number of circumstances.

This means, for instance, that a dominant firm can be held to have engaged in an abusive margin squeeze (which entails manipulating the relationship between wholesale and retail prices for a product) where both price levels have been set or approved by a sector-specific regulator.^[42] Similarly, industry efforts to reorganise a sector suffering from an overcapacity crisis can be treated effectively as a cartel, even if undertaken at the behest of and in accordance with a plan devised by national government.^[43] And under the EUMR, strong governmental support has not conventionally provided a compelling counterweight where the Commission concludes that a proposed concentration is likely to significantly impede effective competition,^[44] although this position may prove more difficult to sustain in future.^[45]

The fact that, in EU competition law, defendants can breach the rules even where their conduct or the anticompetitive effects that follow are heavily influenced by State-imposed regulation is made possible by the lack of a fault or ‘bad’ intention requirement. Thus, under Article 102, ‘proof of the deliberate nature of the conduct and of the bad faith of the undertaking in a dominant position is not required for the purposes of identifying an abuse’.^[46] Under Article 101, it is ‘irrelevant’ to the assessment of whether a restriction by object exists that the coordinating parties acted in pursuit of ‘legitimate’ (ie not deliberately anticompetitive) objectives.^[47] The ‘by effects’ limb of that provision, additionally, considers only the impact in fact of an agreement on the market, rather than the reasons motivating its adoption.^[48] Intention also plays a very limited role in determining the essentially factual question of how a merger is likely to affect competition if it proceeds.^[49]

The absence of a fault requirement can be explained by the fact that the competition rules are aimed at behaviour that harms competition; yet the neoclassical economics that underpin the conventional competition law framework assumes that all firms ultimately intend to ‘harm’ competition in a literal sense by increasing their market share and profits at the expensive of competitors.^[50] Indeed, such self-interested behaviour has long been considered an essentially virtuous aim to the extent that it drives the competitive dynamics necessary to maximise efficiency.^[51] The upshot is that competition law takes an explicitly ‘objective’ approach, at least in principle, to the assessment of whether behaviour amounts to normal, meritorious – and so permissible—competition. To the extent that an absence of fault or bad intention is relevant within EU competition enforcement, it goes primarily to the penalty imposed: breaches must be committed intentionally or negligently to attract fines under Regulation 1/2003,^[52] while the influence of regulation on (otherwise autonomous) firm behaviour can be reflected in a discount on the overall fine.^[53]

B. Competition law and regulatory compliance are unrelated

Second, the question of whether behaviour violates competition law is formally ‘unrelated to its compliance or non-compliance with other legal rules’,^[54] regardless of how strong the link between the regulatory framework and the allegedly abusive behaviour. This means, on the one hand, that behaviour that conforms to a relevant regulatory standard might still breach competition law, if it amounts to an anticompetitive agreement or abuse of dominance.^[55] This position has been defended within EU law on the bases, inter alia, that decisions of (national) regulators should not bind the Commission,^[56] even if the regulator purports to apply and find compliance with the competition rules in parallel;^[57] and that the absence of a fault requirement, discussed above, means that good faith regulatory compliance is insufficient to avoid competition liability.^[58] The general principle that regulatory compliance is ‘unrelated to’ competition law liability would seem to infer that regulatory non-compliance is also largely irrelevant from an antitrust perspective, though Meta Platforms (discussed in Section 4.A) nuances this assumption.

Accordingly, the case-law recognises that ‘in the majority of cases, abuses of dominant positions consist of behaviour which is otherwise lawful under branches of law other than competition law’.^[59] The jurisprudence is more complicated under Article 101, insofar as the sanction of nullity in Article 101(2) means that agreements that violate the provision are void as a matter of contract law too. But even in this context, the Court considers that compliance with other regulatory norms is, generally, ‘not, as such, a matter for competition law’,^[60] and so does not determine the question of whether an arrangement has the object or effect of restricting competition. Under the EUMR, when assessing whether a merged entity is likely to engage in harmful behaviour, the Commission takes account of any pre-existing regulation that might effectively inhibit the merged entity from doing so.^[61] It has also sought, in certain instances, to repackage regulatory compliance as a dimension of quality, to the extent that users place economic value on the underlying public interest protected (such as the level of data privacy offered by a product).^[62] Yet the Commission is clear that, if the objection to a concentration is a fear the merged entity will fail to comply with other regulatory rules in future, ‘[s]uch concerns are not within the remit of merger control and there are regulatory tools better placed to address them’.^[63]

C. Constrained powers of Commission when acting as competition regulator

This links to a third implication of the orthodox separation of competition law and other regulation within EU law, namely the limits of the Commission’s powers when ‘acting as competition regulator’.^[64] In the recent case of Orlen, the Court of Justice emphasised that, when exercising its competition powers, the Commission is restricted to imposing remedies or penalties that ‘may respond to the competition concerns [it has identified] on the market concerned’.^[65] Thus the Commission cannot legitimately use competition enforcement to advance other regulatory goals or interests,^[66] unless these can be absorbed into the ‘competition issues’ at stake.^[67] The Court acknowledged that the Commission’s competition enforcement activity ‘could not lead to a result which is contrary to’ other regulatory norms within EU law,^[68] and in particular, it cannot make binding remedies that would conflict with other (EU-level, at least) regulation. But that it quite different from saying that the Commission might positively impose ‘obligations … motivated by policies other than competition law policies’^[69] through the avenue of competition enforcement; this is something, the Court was clear, the Commission cannot do.

The attitude of the Court in Orlen potentially poses a not-insignificant significant hurdle to realising more ambitious applications of the competition rules,^[70] by establishing that competition law enforcement can pursue only ‘competition issues’. Of course, there is a risk of circularity, insofar as what we consider to legitimately constitute a ‘competition issue’ may vary depending upon how broadly we treat the legitimate objectives of competition law. But the judgment clarifies a point of importance to this article: competition enforcement is not an instrumental tool by which to advance a grab bag of ‘other’ regulatory objectives but must confine itself to responding to competition-related concerns.

D. Ne bis in idem and parallel enforcement of competition law and regulation

Finally, this orthodox understanding of the relationship between competition law and other regulatory norms – as discrete sources of legal obligations for market actors, applied by distinct regulators pursuing different public interest goals – creates potential for overlapping jurisdiction and enforcement. Compliance with regulatory obligations does not, as explained, certify that the regulated behaviour also complies with competition law. Yet overlapping jurisdiction does not always generate conflict, and it is entirely possible that the same behaviour might be found to violate both the competition law and the regulatory standards.

For many years, EU competition law recognised a distinctive principle of double jeopardy or ne bis in idem, which precluded the parallel application of multiple sanctions only if it was demonstrated that the regulatory regime pursued the same legal interest as the competition rules, in addition to unity of facts and offender.^[71] This unique approach reinforced the orthodox view of competition law and regulation as distinct and unrelated tools for market supervision. More recently, the Grand Chamber in bpost has realigned competition law with other areas of EU law by shifting to an idem factum standard requiring only the same facts and offender.^[72] The Court nonetheless recognised that duplicate proceedings and penalties might be permitted where they are a necessary and proportionate response to achieve public interest objectives or to protect the rights and freedoms of others.^[73] Of particular relevance is the Court’s strong assertion, when considering the proportionality of parallel proceedings, that ‘public authorities can legitimately choose complementary legal responses to certain conduct that is harmful to society through different procedures forming a coherent whole so as to address different aspects of the social problem involved, provided that the accumulated legal responses do not represent an excessive burden for the individual concerned’.^[74] Yet parallel jurisdiction cannot be exercised wholly unilaterally if it is to comprise a proportionate response. Instead, the Court in bpost stressed the need for sufficient ‘coordination’ between the competition regulator and other prosecuting authority to ensure a truly ‘complementary’ response.^[75]

4. The interaction of competition law and regulation: the contextual view

One shortcoming of the ‘two barriers’ understanding of the interaction of competition law and other regulation is that it sets up a dichotomy between these instruments of market supervision that does not reflect the reality of how most markets work today. In its competition case-law, the Court of Justice increasingly emphasises that the determination of whether conduct amounts to a breach of the rules requires a detailed assessment of context, meaning that decision-makers must take account of ‘all of the relevant factual circumstances’^[76] and the ‘economic and legal context’^[77] in which allegedly anticompetitive conduct occurs. The underlying regulatory landscape, insofar as it may shape competition dynamics and/or influence the defendant’s own behaviour, is an obvious contextual element that should be considered. Although EU law rejects the proposition that the presence of specific regulation should oust the more general competition rules as a matter of law, the former may play a significant role in determining whether competition law is breached on the facts.

A. Relevance of regulatory norms in principle: from Deutsche Telekom to Meta Platforms

Indeed, even as the Court in Deutsche Telekom sidelined the State action defence and discounted the significance of good faith regulatory compliance in competition assessment,^[78] it recognised the relevance of regulatory context insofar as it ‘contributes to the determination of the competitive conditions under which an undertaking … carries on its business in the relevant markets’.^[79] The Court elaborated three areas where regulatory context might be relevant: ‘defining the relevant markets, assessing the abusive nature of such conduct or setting the amount of the fines’.^[80] In reality, the invocation of regulatory context in Deutsche Telekom was somewhat disingenuous, enabling the Court to endorse a cost–price calculation mechanism that was ‘at odds with the realities of the market’^[81] for the defendant and its rivals.^[82] Despite this, in both Slovak Telekom and Baltic Rail, the Court drew upon this statement from Deutsche Telekom to conclude that the regulatory context justified application of a less exacting legal standard under Article 102, reiterating that ‘a regulatory obligation can be relevant for the assessment of abusive conduct’.^[83] This language is echoed in later cases.^[84] It is the 2023 judgment of the Grand Chamber in Meta Platforms, however, that provides the most explicit support for a potentially pivotal role for regulatory norms within competition law assessment.

The Meta Platforms case began with an infringement decision of the German competition authority against the owner of the Facebook social network, which aimed at the latter’s user-data-gathering policies. In essence, the decision construed Meta’s policies for use of Facebook services as an abuse of market power,^[85] using as a proxy for unfairness in competition law terms the fact that these policies were non-compliant with relevant data protection standards as reflected in the EU’s General Data Protection Regulation (GDPR).^[86] The Court of Justice, called upon to advise the fairly unconvinced national court, began by reiterating the orthodoxy of the distinctiveness of competition law from other regulation, including the different legal interests protected and enforcement by different regulators.^[87] Yet it quickly turned to acknowledge the applicability of regulatory standards and the role of compliance within the nominally distinct task of competition assessment.

The Court went much further than in previous cases, where the regulatory context had been described as potentially ‘relevant’ to this determination; in Meta Platforms, the Court held that ‘compliance or non-compliance … may, depending on the circumstances, be a vital clue’ as to the permissibility of behaviour under competition law.^[88] Notably, although the questions referred pertained only to data protection, the Court’s response was pitched in broader terms.^[89] Though the Court did not elaborate at length on how regulatory compliance might offer a ‘vital clue’ to competition liability, it noted the ‘great importance’ of having access to user data for many digital economy business models,^[90] so that it might undermine the effectiveness of competition enforcement to ‘disregard the reality of this economic development’.^[91] Echoing bpost,^[92] the Court then considered how national authorities should collaborate to minimise ‘the risk of divergences’,^[93] an approach that it grounded in the general EU law duty of sincere cooperation in Article 4(3) TEU. The judgment thus offers a more nuanced account of the inter-relationship between competition law and other regulation, by acknowledging that the latter may influence or even determine the application of the former in practice, depending upon its effects within a relevant market.

B. Relevance of regulatory norms in practice

The impact of the regulatory context on competition law assessment can be grouped into three broad categories: where regulation effectively excludes competition law liability; where it changes the structural competition dynamics within a market; and where it influences the behaviour of economic actors in a manner that prevents – or conversely provokes – competition law infringements.

(i) Regulation as exemption of competition law liability

Most radically, though also most exceptionally, regulation might remove the scope for applying the competition rules to market behaviour or at least preclude a finding of liability. The competition rules apply to ‘undertakings’, a concept that has been interpreted to exclude certain (typically highly regulated) market actors that pursue objectives related to social solidarity or that exercise quasi-public powers.^[94] The State action defence, noted above, provides an effective means to avoid Articles 101 and 102, but only where the applicable regulatory regime removes all capacity for ‘autonomous conduct’, competitive or otherwise, by regulated undertakings.^[95] Similarly, under Article 106(2) undertakings that are ‘entrusted’ by Member States with the operation of services of general economic interest can be exempted from application of competition law, but only to the extent that being subject to such obligations would ‘obstruct the performance, in law or in fact’ of this task. These three disparate exceptions share a common theme: although the would-be defendant is engaged in what is, strictly speaking, ‘economic’ activity on the market, the nature or impact of the underlying regulatory framework removes its actions from the purview of the competition rules as a matter of law, regardless of their effects on competition in fact.

Alternatively, the operation of the regulatory framework may provide a good defence that enables a defendant to rebut the claim that its conduct is harmful from a competition perspective. The Article 101 jurisprudence recognises the ‘Wouters exception’, which enables certain prima facie restrictive arrangements to escape prohibition ‘by effect’ on the basis that the coordination pursues ‘legitimate objectives in the public interest which are not per se anticompetitive’, and subject to indispensability and proportionality requirements.^[96] Formally, the exception does not hinge upon the impact of regulation on the activities under scrutiny; that is, the concept of ‘legitimate objectives in the public interest’ does not require that defendants are empowered to achieve those objectives under public regulation. But it is notable that, on the facts of Wouters, the organisation concerned (the Dutch bar association) was expressly tasked under domestic legislation to adopt the rules under scrutiny, ‘in the interests of the proper practice of the [legal] profession’.^[97] In this sense, a parallel can be drawn to the exceptions to the ‘undertaking’ concept that have similarly developed in the jurisprudence.^[98] In each instance, although the activity concerned is ‘economic’ in nature the predominant interests at stake are not, and typically (indeed, perhaps necessarily in practice) this will be reflected in the underlying regulatory framework.^[99]

Under Article 102, it is difficult to successfully argue that regulatory obligations deprive a dominant undertaking’s conduct of its ‘autonomous’ nature so as to preclude antitrust scrutiny.^[100] But there may be greater scope, where behaviour is driven by such considerations, for advancing a claim of ‘objective necessity’, which can serve as an objective justification, or defence, to a prima facie finding of abuse.^[101] This potential is seen most clearly in the case-law on refusal to deal, where the jurisprudence explicitly recognises that the ‘applicable regulatory framework’ may impose constraints on dominant firm behaviour in a manner that essentially legitimates what would otherwise be an abusive refusal to grant access to the dominant firm’s property.^[102] But objective necessity claims are narrowly construed in practice, and defendants have the burden of establishing that their behaviour is both necessary to achieve the public interest aim and that the actual or potential exclusionary effects resulting from the conduct are proportionate to the allegedly necessary aim.^[103] The objective necessity defence moreover does not apply where a dominant defendant claims, ‘rightly or wrongly’, that its behaviour was necessary to prevent other market actors from breaching their regulatory obligations,^[104] the idea being that private actors should leave the task of enforcement to the professionals, ie the applicable public regulator.

(ii) Regulation as a structural feature of market competition

Most of the time the underlying regulatory framework does not provide anything so dramatic as a wholesale dispensation from competition scrutiny. It is more usual for regulation to feed into antitrust assessment either by heightening or dampening competition dynamics (ie the possibilities for effective competition within a market) or by altering (whether by constraining or directing) the behaviour of economic actors in ways that make anticompetitive conduct more or less likely.

Regulation, first, can function structurally as a barrier to entry or expansion by competitors, typically enhancing the power of incumbent market actors and making anticompetitive effects more likely or plausible. The case-law recognises a wide variety of such circumstances, including statutory monopolies and intellectual property rights.^[105] Consideration of barriers to entry is a key component of the initial assessment of dominance under Article 102, and can also be highly informative of whether a merger is likely to create or strengthen a dominant position, so as to violate the ‘significant impediment to effective competition’ standard under the EUMR. Such structural effects may endure even after regulation: so, for instance, entities that previously benefited from special or exclusive rights under regulation may continue to experience de facto benefits after such legislation is removed,^[106] firms may maintain anticompetitive arrangements after the regulatory justification for doing so has disappeared,^[107] or firms may even seek subsequently to reimplement regulatory barriers through private arrangements after these have been dismantled by market liberalisation.^[108]

Regulatory barriers to entry may also be relevant to the assessment of likely anticompetitive effects that follow from behaviour.^[109] It was noted above that contemporary competition assessment is highly attuned to the specific context in which conduct occurs; increasingly, the case-law also emphasises that competition law aims at behaviour that has at least potential anticompetitive effects.^[110] The regulatory framework may therefore feed into the evaluation of whether coordination violates Article 101(1) or that single firm conduct constitutes an abuse of existing market power, insofar as the regulatory context may influence ‘the consequences of a certain practice in the market or for consumers’.^[111]

A more complex question is whether the anticompetitive effects must be caused by the impugned behaviour. In Google Shopping, the Court of Justice described the establishment of a causal link between a firm’s behaviour and the claimed actual or potential anticompetitive effects as one of the ‘essential constituent elements of an infringement of competition law’.^[112] This would appear to preclude a finding of breach of Articles 101 or 102 where the harmful effects on competition are attributable more or less entirely to the impact of the relevant regulation, even if defendant undertaking(s) have engaged in what looks like wholly unmeritorious behaviour from a competition perspective. Accordingly, while there is a high threshold to prove that regulation robs firm behaviour of its autonomous character, there may be greater scope to argue that the pervasive distortive effects of regulation mean the necessary causal link cannot be found. Yet Google Shopping does not require an exacting causation standard, with the Court accepting that it is sufficient merely to demonstrate a ‘correlation’ between suspect practices and the evolution of the market concerned, provided that a causal relationship can be confirmed by ‘additional information’ such as the position of market participants.^[113] The question in regulated markets is therefore whether it is possible to identify a sufficient diminution in effective competition arising from the suspect behaviour, over and above any limitations or distortions arising from the regulatory restrictions.

Conversely, regulation may facilitate new entry which enhances competition overall. This is, for instance, a principal objective of much of the market liberalisation legislation introduced in the EU since the 1990s.^[114] There are a number of common regulatory techniques which are deployed frequently to increase opportunities for competition: such as requiring the removal of existing monopoly rights that protect (typically dominant) incumbent operators from new entry; imposing structural separation on vertically-integrated incumbents to reduce incentives to engage in exclusionary behaviour; and requiring incumbents to share existing infrastructure or even data with potential new rivals, to enable or assist entry. Regulation aimed at market-opening may be supplemented by regulation that controls the behaviour of former monopolists in liberalised sectors, a point developed further below. In such circumstances, antitrust takes account of the altered market dynamics effectuated by the presence of the regulation, so that a defendant cannot, for instance, argue that its behaviour should be assessed as if unencumbered by its regulatory burdens.^[115] On the other hand, the regulatory framework may make it more plausible for a defendant to claim that its superficially restrictive behaviour generates countervailing efficiencies.^[116]

(iii) Regulation and its impact on market behaviour

Regulation may also influence a firm’s market behaviour in ways that have implications for competition assessment: either because it directly controls market behaviour or because it incentivises conduct that generates (anti)competitive effects. Here, several scenarios can be envisaged.

Most constructively, regulation may guide a firm to competition law-compliant behaviour. Where appropriately calibrated and effectively implemented, a regulatory framework may prevent regulated firms from engaging in the sorts of behaviour that would otherwise cause competition problems: by limiting a firm’s capacity to charge prices that are excessively high or have exclusionary effect, for instance, or by preventing use of ‘unfair trading conditions’ in contractual relations. This outcome is most likely to manifest itself obliquely, in an absence of antitrust problems in a sector.^[117] To the extent that context-specific competition assessment takes place prospectively, the general starting point is that the market behaviour of economic actors is shaped and constrained by other regulatory norms that apply within that sector and to that conduct.^[118] This is most salient in merger control, where it is assumed that firms comply with their regulatory obligations, and that their scope for ‘free’ market conduct is curtailed accordingly.

More contentiously, regulation may guide a firm to behaviour that is essentially at odds with competition law, a scenario most likely to arise where the regulatory regime endorses ‘inefficient’ market outcomes in pursuit of other policy goals. In this case we may encounter the possibility flagged in Section 3, namely prosecution under the competition rules of behaviour that is not only fully compliant with, but has also been largely influenced by, a regulatory framework that deliberately prioritises other socially valuable goals.^[119] This is an essentially unsatisfactory outcome, which is considered further below.

The first two scenarios each envisage effective compliance with regulatory obligations, whether this leads to competitive market outcomes or not. But firms do not always respect their regulatory obligations or make good faith efforts to do so. Within the competition jurisprudence, it is possible to identify several scenarios whereby an antitrust defendant’s liability is essentially contingent upon its interaction with another regulatory framework. It is these cases, in which regulatory norms function as the ‘hook’ for additional antitrust liability, which raise the most interesting but also the most difficult questions regarding the impact of regulatory norms on competition law.

Most straightforwardly, a regulated firm may simply disregard the constraints that are placed on its behaviour. In Meta Platforms, a digital firm’s data-gathering policy was deemed to amount to an abuse of its market power because the policy failed to respect the relevant data protection standard. A similar approach has been adopted in the Commission’s enforcement activity in liberalised utilities sectors, where failure to abide by regulatory duties to share infrastructure is also construed as an abuse of dominance by the incumbent operator.^[120] Here, the regulatory violation – which may or may not have been prosecuted by the sector regulator in parallel – is usually obvious; the more interesting question is how and why the behaviour might additionally become the focus of an antitrust investigation.

Beyond the problem of mere non-compliance, the competition case-law discloses a variety of instances where liability hinges on a firm’s disruptive interaction with a regulatory framework. In essence, these cases involve so-called ‘regulatory gaming’, namely ‘[b]ehaviour that abuses a neutral or procompetitive regulatory structure and wields it as a tool to accomplish [anticompetitive] results’.^[121] What distinguishes the non-compliance cases from the regulatory gaming ones is that, generally speaking, in the latter the firm’s behaviour is unproblematic (or at least unprosecutable) under the relevant regulatory regime. Instead, antitrust liability derives from the defendant’s efforts to avoid the regulatory constraints while remaining within the strict letter of the law (what we term ‘bad faith compliance’), or to exploit opportunities created by the existing regulation to anticompetitive ends (‘abusive compliance’).

Many examples of bad faith compliance relate to the highly regulated pharmaceutical sector. In AstraZeneca,^[122] the first EU competition case that directly embraced regulatory gaming as a theory of harm, the defendant took advantage of loopholes in the rules on pharmaceutical marketing authorisation and weaknesses in the administration of national patent offices to exclude would-be generic competitors. In Teva,^[123] the defendant strategically lodged and withdrew divisional patent applications in a manner permitted by the regulatory framework but which was clearly intended to frustrate new entry. Most remarkably, in Aspen Pharma the defendant successfully implemented an excessive pricing campaign in a context where every Member State had domestic price regulation, by exploiting design flaws which allowed it to manipulate the regulatory price-setting processes.^[124] For our purposes, what is most notable about these cases is how they make the leap from attributing antitrust liability for non-compliance with the express rules of a regulatory framework to non-compliance with the broader spirit or underlying policy objectives of those rules.

In abusive compliance cases, the regulatory context is primarily of relevance insofar as it provides an explanation for strategic behaviour that otherwise lacks an anticompetitive rationale. This is exemplified by AB InBev,^[125] where the defendant changed its packaging for beer sold in France and the Netherlands to prevent cross-border resales in Belgium, where it charged higher wholesale prices. This change – to no longer include information in both Dutch and French on beer sold in countries other than Belgium – only made sense as an exclusionary strategy in view of Belgian legislation on food labelling, which meant that larger retailers only purchased products labelled in both languages. In Roche & Novartis,^[126] the defendants engaged in collusive lobbying efforts to persuade the Commission to change its advice on the off-label prescribing of a certain drug, which enhanced the profits of both cartelists. Similarly, in Car Emissions,^[127] car manufacturers agreed to design new vehicles to the minimum regulatory standard for nitrogen oxide cleaning, an arrangement that the Commission prosecuted as a secret cartel. What is notable is that the cartelists explicitly agreed to adhere to their existing regulatory obligations, albeit knowing that these were likely to become more demanding in future; in doing so, however, they effectively agreed not to compete on cleaning technology as a dimension of competition, which the Commission construed as a limitation both of consumer choice and future innovation. Pharmaceutical pay-to-delay agreements, whereby drug companies pay off potential new entrants in the guise of patent litigation settlements, can also be seen as abusive compliance.^[128]

The real world is complicated, of course, and some regulations may be equivocal in terms of their overall effects on competition, so that a neat dichotomy between regulation that limits competition versus that which enhances competition often cannot be drawn. The GDPR provides an example of such mixed effects. On the one hand, by imposing limits on what market actors can do with personal data, it constrains the market power of Big Tech companies, which have disproportionately plentiful access to such commercially valuable information.^[129] On the other, GDPR compliance is an expensive business, and so the very existence of these rules, which apply to all data controllers regardless of size, may constitute a barrier to entry.^[130] The GDPR says little about its intended effects on competition beyond a vague reference to levelling the playing field,^[131] yet it has a significant impact on many digital business models, as recognised in Meta Platforms.^[132]

C. Understanding the role of regulatory norms

Regulation accordingly feeds into competition assessment in numerous (occasionally contradictory) dimensions, from reducing the scope to identify autonomous market behaviour to enhancing the risk that firm conduct generates harmful effects. Where competition law is applied in regulated markets, the competition rules do not generally endorse or apply the regulatory norms as such. But by accepting a market context where these norms reflect the accepted ‘rules of the game’ for market participants, competition law essentially reinforces the more diverse public interest values that are reflected within other regulatory regimes. Occasionally, this means deferring entirely to alternative values by allowing the regulatory regime to displace the application of competition law or to provide at least a good defence. More usually, these norms are absorbed within the contextual analysis, and the case-law increasingly recognises the legitimacy of aligning the thrust of competition analysis with pre-existing regulatory obligations (provided, of course, that these are not fundamentally at odds). There is obvious potential for tension, where for instance competition law turns its focus on market power that has been created by regulation as, say, a reward for innovation or to ensure the universal delivery of public services. But even here, competition analysis typically finds a way to accommodate potentially conflicting values, by distinguishing between the lawful scope of intellectual property rights and efforts to extend those rights beyond what is envisaged by IP law,^[133] for instance, or by recognising the fundamentally different position of liberalised and non-liberalised markets.^[134] Even if the goal of competition law is merely to achieve and protect undistorted competition, the space given to other regulatory norms within the contextual approach allows for the accommodation of complementary (and sometimes conflicting) values. These cases thus illustrate the tendency towards a ‘sponge’^[135]-like or ‘polycentric’^[136] quality within competition law, to the extent that a diversity of values can be accommodated – if not necessarily pursued directly – within contemporary enforcement.

The above discussion nonetheless identified two broad areas where this narrative of an essentially passive-but-tolerant inter-relationship between competition law and other regulatory norms does not provide an effective explanation. The first scenario arises where a (typically national) regulatory framework favours inefficient market outcomes in pursuit of non-economic objectives, albeit leaving sufficient scope for independent conduct to attract antitrust attention. The unsatisfactory position in the current law has been noted: in line with the orthodox approach, unless the narrow State action defence applies, distortions generated by the regulatory framework can be reflected only by discounting the penalty that might be imposed.^[137] This position can be critiqued from two dimensions. It creates unfairness for, or at least imposes a heavy burden on, defendants, who seem to have a positive responsibility to seek to align their regulatory obligations with the perceived demands of competition law.^[138] But even leaving the question of fairness aside, when viewed through the lens of contextual analysis this approach gives little weight to the obviously significant impact of the regulatory framework on competition dynamics in the market concerned. Instead, there is a strong argument that the (typically, non-economic) considerations factored into the regulatory framework should inform, though not necessarily determine, the competition enforcer’s understanding of what ‘competition on the merits’ looks like in a sector. While this may be insufficient to provide a good defence where there is further evidence that the defendant’s conduct is objectively non-meritorious in competition law terms, it would at least provide some recognition of the public interests that motivate the regulatory framework. To the extent that the latter are pretextual or at least insufficient to justify the restrictions on competition entailed, the more appropriate competition policy response is to attack the regulation directly.^[139]

A second category of cases that require more in-depth consideration are those involving non-compliance and regulatory gaming, which pose a conceptual dilemma from a competition policy perspective. Through such behaviour, a firm may engage in conduct that is harmful to effective market competition, which seems to fall within the legitimate scope of the competition rules. Yet insofar as the firm’s ‘unmeritorious’ behaviour revolves principally around its interactions with another regulatory framework, one might ask what makes this an appropriate subject for antitrust intervention, as opposed to the more obviously and directly applicable regulatory regime. In cases of non-compliance, where the behaviour violates the regulatory norm, the question is how to justify competition enforcement as opposed to (or in addition to) direct regulatory enforcement. In cases of bad faith compliance, the question is why competition law might take it upon itself to redesign or fill in perceived gaps within a regulatory framework rather than deferring to the regulation itself. Cases of abusive compliance are less intimately entwined with the obligations imposed by a regulatory framework, making these easier to explain as examples of ‘regulation as context’, discussed above. Yet such cases almost unavoidably involve value judgements regarding the nature of the defendant’s interaction with the regulatory regime, and so raise a similar question to other categories: how can we explain the role played by regulatory norms within the competition law assessment in such cases? This is the focus of the next section of this article.

5. Understanding the role of regulatory norms in competition assessment: explaining the non-compliance and regulatory gaming case-law

The crux of concern is that these cases involve, in essence, the ‘outsourcing’ of liability standards within competition law to other regulatory fields. A competition lawyer schooled in the ‘two barriers’ approach to the competition law/regulation inter-relationship would not question the legitimate existence of regulatory norms ‘other than competition law policies’.^[140] Yet it is less obvious why competition law should effectively defer to standards of behaviour developed in other regulatory policy areas – a problem that becomes more acute in cases of regulatory gaming, which involve deference to vaguer understandings of what a regulatory regime ought to entail if working well. The objection is not that the policy goals reflected in the regulatory framework lack legitimacy or value. Instead, the question is why competition law internalises these goals by applying regulatory norms as the determinative standard of competition liability in such cases. Put another way, why should protection of the goals, values and policy choices that are reflected in a specific regulatory framework fall within the legitimate scope (and enforcement purview) of competition law in addition?

Before exploring this question, let us recap several core principles of the ‘orthodox’ jurisprudence. Formally, antitrust compliance is unrelated to regulatory compliance within EU law: a firm might breach the competition rules while respecting its regulatory obligations, and vice versa. This links to the absence of a fault requirement, meaning that good faith regulatory compliance does not immunise a firm from competition liability in respect of the same behaviour. Competition enforcers are limited to the pursuit of ‘competition issues’; a somewhat woolly limitation, but one which confirms that the competition rules are not a panacea for all public policy ills. Accordingly, to explain the non-compliance and regulatory gaming cases, we must consider why as a matter of competition law such behaviours come within the legitimate scope of the competition rules. To do so, we explore the case-law from three distinct perspectives – through the lenses of effectiveness, competition on the merits, and regulatory non-compliance and gaming as strategic action – asking whether each provides a plausible explanation for addressing the behaviour under competition law.

A. Effectiveness of competition law as a regulatory enforcement mechanism

It is important to recognise why both non-compliance and regulatory gaming cases typically arise in practice: namely, the effectiveness of competition law as a regulatory enforcement mechanism, unconstrained by substantive or procedural limitation that may hinder enforcement of the regulatory framework directly. This is apparent from the background to many cases considered above. In Telecom Polska, the sector regulator had repeatedly enforced national telecommunications law against the defendant, but the penalties it imposed were insignificant and failed to deter non-compliance. In Slovak Telekom, the defendant had evaded its regulatory obligation to grant access to infrastructure for five years but received no pushback from the (captured or under-resourced) regulator. In Meta Platforms, the problem was a mismatch of jurisdiction: while the data privacy concerns were felt most acutely by German consumers and policymakers, power to enforce the GDPR lay with the Irish data protection authority, which dragged its feet. The deficiency of national patent systems is a key undercurrent in many competition cases in pharmaceutical markets.^[141]

Viewed pragmatically, it is easy to understand why competition law is used to reinforce regulatory norms: in effect, competition enforcement might get the job done more effectively than the regulation itself. Competition law brings much to the table as a mechanism to enforce regulatory rules. In particular, the broad sweep of its prohibitions means that competition law is sufficiently wide and flexible to catch such behaviour, including problematic conduct that might lie outside the formal parameters of the regulation itself. EU law also imposes demanding minimum standards to protect the independence, resourcing and sanctioning powers of competition authorities,^[142] which sectoral regulators often lack. The resistance within EU law to accept antitrust preclusion even in the presence of significant regulation supports the proposition that competition law can legitimately seek to resolve market problems that arise from malfunctioning regulation, providing a ‘complementary legal response[]’ in order to form ‘a coherent whole’ of market supervision.^[143]

Yet the bare fact that competition law can enforce regulatory norms more successfully provides an inadequate explanation for these cases. On the one hand, using competition law as a sticking-plaster on ineffective regulation gives little weight to the substance, and substantive limits, of the underlying regulatory framework. Cases like bpost and Meta Platforms demonstrate that, if competition law is applied to ‘regulated’ problems, this requires a degree of intentionality and coherence and must be done in a manner that recognises the ‘respective powers and competences’^[144] of each regime. Deploying competition law on the pragmatic basis that it can better achieve the goals of a regulatory framework fails to acknowledge the central role of the underpinning regulation in such cases, both in terms of how ‘the social problem involved’^[145] and the appropriate enforcement response should be framed.

On the other hand, justifying these cases on an effectiveness basis offers little insight into how such regulatory failures can also become, in the language of Orlen, ‘competition issues’.^[146] Regulation, as discussed, often incorporates a diversity of policy goals that extend beyond the protection of undistorted competition. Regulatory standards of acceptable behaviour typically seek to advance values outside the scope of competition law as such. To take an notable example, the GDPR states that ‘[t]he processing of personal data should be designed to serve mankind’, and explains that the rules seek to balance a variety of fundamental rights including privacy and family life, freedom of thought, freedom of expression, freedom to conduct a business, fair trial rights, and cultural, religious and linguistic diversity.^[147] While it is difficult to argue against the importance of these varied considerations, it is clear that a data protection standard – such as that applied to impose antitrust liability in Meta Platforms – reflects a wider ranges of values than the mere protection of undistorted competition. If we want to bring the former within the legitimate scope of the latter, we need a more nuanced explanation as to when and why diverse regulatory concerns are also ‘competition issues’.

B. Regulatory compliance as a reflection of ‘competition on the merits’

Instead, we return to the concept of ‘competition on the merits’: the idea that competition law does not seek to constrain firms from competing, even aggressively, but merely limits market activity to the acceptable parameters of ‘normal competition’. The concept is prominent in recent jurisprudence and in the Commission’s Draft Guidelines on Article 102.^[148] The latter follow the approach in SEN,^[149] by decoupling the notion of ‘conduct departing from competition on the merits’ from ‘capability to produce exclusionary effects’, while indicating that both elements must be satisfied to establish exclusionary abuse of dominance.^[150] The Draft Guidelines nonetheless suggests that capability can be presumed for a variety of practices,^[151] making competition on the merits the key determinant of liability in many circumstances.

The proposition that competition law protects only ‘competition on the merits’ might justify the direct application of regulatory norms to the extent that the standard of ‘normal’ competition implies that market actors, inter alia, follow their regulatory obligations. Regulatory compliance is often costly for market actors, whether because the regulation curtails a firm’s scope to engage in profitable activity^[152] or requires it to incur costs it would prefer to avoid,^[153] or simply because of direct compliance costs.^[154] While escaping or minimising regulatory obligations can give individual undertakings a competitive advantage, doing so undermines the public interest values that the regulation advances and protects. Viewed in this manner, non-compliance or gaming of regulatory norms is inherently at odds with what we understand meritorious, and thus permissible, market competition to look like. Such an approach aligns with the ‘level playing field’ notion that permeates EU law, which is premised on the idea that market actors should be subject to common rules and restrictions to ensure fair competition.^[155] Since the regulatory framework structures the ‘fair’ competitive game in the level playing field of the marketplace, it is reasonable to require adherence under competition law even if some regulatory rules pursue values that are non-economic in nature.

At its simplest, the existing regulatory framework might be taken to reflect the lawful scope of a firm’s freedom in the marketplace. By incorporating regulatory standards into antitrust assessment, we do not ‘enforce’ these norms as such. Yet because a defendant can have no legitimate expectation to act contrary to its regulatory obligations, it cannot object where this behaviour is curtailed or punished through competition enforcement. Whereas in Slovak Telekom and Baltic Rail the Court denied that a regulatory violation might comprise an abuse in itself, it accepted that the presence of regulation may permit a less demanding legal standard for antitrust intervention.^[156] It justified this essentially on the basis that pre-existing regulation already circumscribed what the defendants could reasonably expect to be allowed to do in the marketplace. Competition law need not maintain the fiction that, say, a firm can validly refuse to grant access to its infrastructure unless an antitrust duty to deal arises exceptionally under Article 102, if sector-specific regulation has already imposed an unambiguous mandatory sharing obligation. A similar logic is discernible in the Commission’s approach to sustainability agreements in its Horizontal Cooperation Guidelines.^[157] Where ‘appropriate regulation’ exists to adequately address market failures, the Guidelines are sceptical of the acceptability of private coordination in tandem,^[158] ‘given that the legislator has already decided that each undertaking must individually comply with the obligation in question’.^[159] Yet private action may be more necessary to address ‘residual market failures that are not or not fully addressed by public policies and regulation’.^[160] In the latter case, undertakings have plausible claims for greater freedom of (private) action, since regulation has not yet delimited the legitimate scope of market activity. This reflects, in the language of Meta Platforms, ‘the reality of economic development’ in regulated markets.^[161]

Yet treating a regulatory framework as merely indicative of the limits to a firm’s economic freedom provides an incomplete explanation of the non-compliance and gaming case-law. First, it is less convincing in respect of behaviour that violates the spirit of a regulatory norm rather than its express provisions. It is one thing to assert that a defendant can have no legitimate expectation, for example, to refuse to comply with a mandatory sharing obligation imposed by sector-specific regulation; it is quite another to make this claim regarding behaviour that is, say, objectively ‘misleading’ but not strictly speaking unlawful.^[162] Arguably, the mere existence of collusion or dominance limits what firms can legitimately expect to do in markets, even absent abusive behaviour;^[163] and applying this logic, we might extend the proscriptive potential of regulation to a wider subset of behaviours at variance with its policy objectives more broadly.^[164] Yet this creates a legal certainty problem: how can firms know in advance what sorts of behaviours are likely to conflict with the broader ‘regulatory orbit’, if this is not clearly delimited by regulation? Second, this approach provides only a partial account of what happens in non-compliance and gaming cases: it tells us why defendants cannot object to competition enforcement covering the same ground, but not why such behaviour can legitimately be construed as a ‘competition issue’ as well as a regulatory breach.

A more muscular version of this argument is what we call the ‘dominance-plus’ approach. From this perspective, because non-compliance or regulatory gaming is inherently non-meritorious, any conduct of this sort is fair game under competition law provided that a threshold criterion for antitrust jurisdiction (most naturally a dominant position under Article 102) is also satisfied. While competition law cannot compel firms directly to comply with their regulatory obligations, it can punish non-compliance or gaming insofar as such conduct falls within the discrete prohibitions in the competition rules. It is the existence of significant market power (or the fact of coordination, if we extend the logic to a collusion-plus approach under Article 101^[165]) that brings the matter within the scope of competition law; the regulatory misconduct then provides the plus factor that crystallises the competition infringement. This was, effectively, the position of the Bundeskartellamt in Meta Platforms.^[166]

This approach has a degree of intuitive appeal. It is hard to argue that a firm acts meritoriously if it disregards or takes advantage of its regulatory obligations, though it is more debatable as to whether this is ‘normal’ market behaviour. Although the regulatory regime may reflect values beyond the task of protecting undistorted competition, the enforcement of such values in non-compliance and gaming cases can be explained as an example of competition law’s ‘sponge’-like quality, a term developed by Ezrachi to describe the capacity to absorb and reflect policy influences from other sources such as market regulation.^[167] Undeniably, this is an understanding of competition law’s role that reflects a more overtly ‘political’ vision of antitrust as a tool directly to regulate economic power, a perspective which typically gives less weight to more ‘legal’ questions of the permitted parameters of a dominant firm’s market freedom.^[168] Yet it is not obviously precluded by the Article 102 jurisprudence, which recognises the ‘special responsibility’ of dominant firms to protect effective competition,^[169] and which does not require misuse of the market power that underlies a dominant position in order to establish ‘abuse’.^[170]

Yet even accepting that dominant firms have a special responsibility to avoid distorting competition, objections can be raised against construing regulatory non-compliance or gaming of norms as ‘abuses’ in an antitrust sense. The case-law establishes that, even if ‘a regulatory obligation can be relevant for the assessment of abusive conduct’, the presence of regulation ‘cannot relieve the Commission of the requirement of establishing that there is abuse’ on the facts.^[171] While Meta Platforms saw the most ambitious use of regulatory norms to determine liability, the Court still stopped short of treating regulatory non-compliance as conclusive proof of antitrust abuse, and it emphasised the differing ‘objectives and tasks’ pursued by competition law compared with other regulatory fields.^[172] In bpost, where the Court acknowledged that competition law might ‘legitimately’ tackle behaviour that is also subject to regulatory controls, it clearly envisaged that parallel proceedings would address diverse regulatory objectives.^[173] It is doubtful whether repackaging a regulatory violation as an antitrust one merely due to the presence of dominance (or indeed coordination) respects this limitation. The dominance-plus approach also sits uncomfortably with the principle that good regulatory compliance is generally irrelevant to the objective question of whether conduct attracts antitrust liability, provided that the regulatory regime allows sufficient scope for autonomous behaviour that can come within the reach of the competition rules.^[174]

This concern – that defendants might somehow be damned if they do and damned if they don’t respect regulatory norms – highlights an additional objection, namely the risk of disproportionate punishments for what are, at their core, regulatory violations. bpost does not prohibit the parallel enforcement of regulatory norms, although it includes proportionality considerations in the determination of whether an exception to ne bis in idem is permissible in individual cases. But even where proceedings are limited to competition enforcement, the consequences for defendants are often much more disadvantageous than if prosecution was pursued under the regulation directly.^[175] Even if one has little sympathy for firms that disregard their regulatory obligations, reinterpreting a regulatory violation as a competition abuse can raise the stakes considerably, including the possibility of private damages claims alongside public enforcement.^[176] There is also the question of how to impose fair and sensible limits on the scope of liability from a dominance-plus perspective: might any regulatory breach – whether of tax law, zoning rules, labour protections etc – become an antitrust abuse merely due to the existence of dominance (or collusion)? Concerns about disproportionate liability are even more acute in regulatory gaming cases, where the defendant has not in fact violated the relevant regulatory norm, so that antitrust enforcement not only reinforces but extends regulatory liability.

Yet within the emerging Article 102 jurisprudence the existence of conduct departing competition on the merits is not (typically) the sole determinant of whether an abuse arises; additionally, at least in principle, the conduct must have the capability to produce exclusionary effects.^[177] An effective limiting principle for non-compliance and regulatory gaming cases could conceivably be derived by adding a robust requirement to demonstrate anticompetitive effects arising from the regulatory breach/gaming, alongside a ‘but for’ causation standard directly linking the behaviour to its harmful effects. Demonstrating actual (or at least potential)^[178] and appreciable anticompetitive effects would squarely locate the problem within the sphere of ‘competition issues’ that fall within the legitimate scope of competition law. Requiring a ‘but for’ causation standard, additionally, would enable us to explain why this regulatory abuse should also generate liability under competition law. Such an approach has much to recommend it in principle.

It is at this juncture, however, that we run up against the constraints of the existing case-law. Although the recent jurisprudence makes greater space for effects-based considerations, in practice it still falls short of both proposed elements in most instances. While the case-law is not a model of clarity,^[179] it is only with respect to the ‘by effect’ limb of Article 101(1) that we find anything close to such a robust effects requirement.^[180] Under the ‘by object’ limb,^[181] it is essentially unnecessary to examine effects where an arrangement discloses intrinsic harm to competition ‘by its very nature’.^[182] Under Article 102, which is the basis for most of the cases considered here, behaviour needs only to be capable of restricting competition,^[183] a standard that can be met despite evidence that the conduct ‘has not produced actual anti-competitive effects’.^[184] To the extent that the case-law imposes a causation requirement, it foresees merely a need to demonstrate a ‘correlation’ between the defendant’s conduct and the claimed (likelihood or capacity to cause) harm.^[185] There are sound competition policy reasons to defend a more prophylactic approach, in order to prohibit overtly restrictive arrangements,^[186] or where the presence of a dominant firm implies a structural absence of competition.^[187] Yet the continuing distance between the rhetoric and reality of the effects-based approach means that a simple dominance-plus perspective may provide an over-inclusive account of when regulatory non-compliance and gaming of norms should also constitute competition issues.

C. (Non) compliance as a positive strategic action

Another way of thinking about these cases is to view non-compliance or regulatory gaming as a positive strategic action for competition law purposes. Here, the focus is the deliberate choice of defendants to neglect or ‘game’ their regulatory obligations as part of an anticompetitive course of conduct: the antitrust problem stems not merely from the fact the regulatory norm has been disregarded or disrupted, but also from the motivation for this behaviour. In effect, the regulatory framework is ‘weaponised’ to anticompetitive ends, and it is this dimension – rather than the regulatory element as such—that is pursued under competition law. The key aspect is thus the strategic behaviour of the defendant undertaking in terms of how it interacts with, takes advantage of, or disregards its regulatory obligations.

To provide an effective explanation of the non-compliance and regulatory gaming case, the question is whether (and in what circumstances) ‘strategic’ behaviour of this sort can breach the competition rules. Recent case-law provides support in several dimensions for an approach to liability that hinges on evidence of the anticompetitive intention of defendants. The transformative Intel judgment emphasised the relevance of evidence of an anticompetitive ‘strategy’ to find abuse of dominance.^[188] The Court in Google Shopping endorsed a focus on the ‘positive acts’ undertaken by the defendant,^[189] which served to distinguish its (abusive) actions from other (permissible) forms of self-preferencing. It also held that, when engaging in context-specific assessment, it is ‘clearly legitimate for the Commission to assess subjective matters, namely the motives underlying the business strategy in question’.^[190] The more contextual approach to ‘by object’ determination under Article 101 similarly extends consideration of the inherent ‘nature’ of coordination to the ‘aims which that conduct seeks to achieve from a competition standpoint’.^[191]

Yet intention-based requirements are controversial in competition law. Two broad objections can be advanced against conceptualising the non-compliance and gaming cases as instances of ‘strategic’ behaviour: one substantive and one practical. In substantive terms, it has been argued that in the presence of market power or coordination, certain forms of conduct should be objectively unacceptable due to their inherent capacity to distort competition, regardless of what a defendant seeks to achieve by its actions. This perspective is reflected in the jurisprudence, which rejects a determinative role for intention evidence within the framework of EU competition analysis. The perceived ‘objective aims’ of coordination are distinguished from the ‘subjective intention’ of coordinating parties, with the latter ‘not decisive for the purposes of the application of Article 101(1)’.^[192] Under Article 102, while ‘proof of an anti-competitive intention’ may constitute ‘an indication of the nature and objectives pursued by the strategy of that undertaking and, on that basis, may be taken into account’, the recent case-law reiterates that ‘demonstration of the existence of such an intention is neither necessary nor sufficient, in itself, to establish the existence of an abuse’.^[193] In Google Shopping, the Court described ‘the existence of any anticompetitive intent’ as ‘only one of a number of facts which may be taken into account’,^[194] and emphasised that there is ‘no obligation to establish the existence of such intent … in order to render Article 102 applicable’.^[195]

The substantive objection is not trivial, and imposing an across-the-board obligation under Articles 101 or 102 to demonstrate subjective anticompetitive intention in addition to establishing behaviour that is objectively harmful to competition would clearly be inappropriate. Yet as demonstrated by the case-law on predatory pricing, intention can play a valuable, and determinative, role in establishing antitrust liability in certain circumstances. Under the Akzo test,^[196] anticompetitive intention serves to confirm the unacceptability in antitrust terms of behaviour that is objectively ambiguous from a competition policy perspective.^[197] Intention evidence can usefully perform a similar function in regulated conduct cases, marking the dividing line between behaviour that is purely a problem within the context of a regulatory regime and behaviour with a dual character that also merits antitrust intervention (ie that reveals a ‘competition issue’). Here, again, we come up against a similar difficulty to that encountered when discussing the use of an effects-based limiting principle under the dominance-plus approach, namely the limits of the existing case-law. Yet it could be argued that recognising another intention-based exception where this possibility is already established in the jurisprudence is less radical than requiring a proper effects-and-causation analysis. In any event, these elements are not mutually exclusive.

The more procedural objection to making liability hinge on evidence of an anticompetitive strategy relates to the difficulties of demonstrating a ‘guilty mind’ where defendants are typically large corporate entities. On the one hand, there is the problem of determining what counts as the intention of the undertaking as a whole; on the other, there is the inherent tension, outlined above, of construing evidence of a desire to enhance an undertaking’s own market share as evidence of an objectively anticompetitive intention.^[198]

Yet, practically speaking, this has not prevented the accumulation of significant evidence of a defendant’s deliberate efforts to evade or defraud its regulatory obligations in many of the non-compliance and gaming cases discussed earlier.^[199] Indeed, the frustrating aspect of the Court’s insistence on the formal irrelevance of intention is that it is hard to reconcile with what occurred in these cases. For instance, whereas the General Court in AstraZeneca maintained that ‘proof of the deliberate nature of the conduct and of the bad faith of the undertaking in a dominant position is not required’,^[200] the higher Court of Justice stressed that such a hypothetical scenario was ‘radically different from … the present case’.^[201] The Teva decision, similarly, is imbued with the language of (bad) intention: the Commission found that the defendant ‘aimed at’, ‘consciously’, ‘intentionally’ and ‘clearly pursued’ its anticompetitive divisional patent applications.^[202] The decision formally maintained that there is ‘no requirement’ to establish anticompetitive intention,^[203] however, and cited Google Shopping to the effect that ‘the existence of an intention to compete on the merits, even if it were established, could not prove the absence of abuse’.^[204] Yet it is essentially impossible to envisage a plausible ‘good faith’ version of regulatory non-compliance or gaming as a competition law theory of harm. We have thus another example of the problem that so concerned Advocate General Wahl in his pivotal Opinion in Intel, with the Court saying one thing but doing quite another in practice.^[205] The policy question is whether it makes sense for EU competition law, in essence, to keep its options open by maintaining the fantasy that motivation is not a key element of the theory of harm in such cases; a question that pits legal principle against a desire to preserve the practical effectiveness of competition law to the greatest extent possible.

A focus on strategic anticompetitive intention in cases where the theory of harm is tied up with the question of regulatory compliance has a further aspect to commend itself: it could arguably provide a more satisfactory explanation for the problematic cases where a defendant is held to breach competition law despite its compliance with regulatory obligations that strongly influence its supposedly anticompetitive conduct. We identified two objections to this jurisprudence: the possible unfairness to defendants faced with the responsibility to correct policy choices made by a regulator, and the fact that it gives insufficient weight to how regulation affects ‘competition on the merits’ in the sector concerned. Yet these objections are less compelling if would-be ‘good faith’ compliance is not all that it seems. Where there is evidence that a defendant is acting in a strategically anticompetitive manner in its interaction with the regulatory framework, it is manifestly less unfair to pursue such behaviour under competition law. Evidence of anticompetitive intention can also provide an effective rejoinder to the claim that the regulatory regime demarcates the boundaries of ‘meritorious’ or permissible competition within a market. Of course, such an approach again departs from the orthodox ‘objectivity’ of competition assessment and may be criticised on this basis. Yet as a compromise between the two poles of deferring to and simply ignoring the coercive impact of regulation on private market behaviour, focusing on what the regulated defendant itself sought to achieve through its interaction with the regulatory framework arguably provides a reasonable limiting factor.

6. Conclusion

This article began with a simple (perhaps simplistic) dichotomy between competition law, which promotes undistorted competition, and other market regulation, which has a greater or lesser capacity to distort competition in service of alternative policy goals. Yet Ezrachi has criticised the ‘pretence of purity’ that competition law adopts,^[206] and even if antitrust liability professes to be formally unrelated to compliance with other regulatory norms, the presence of regulation can have a profound effect on the task of competition assessment in practice.

Regulation and the diverse public interest values that it represents are a constant feature of modern markets. Context-specific competition enforcement necessarily absorbs and reflects these values to some degree, to the extent that it works within the ‘distortions’ of competition that follow from regulation. In the absence of deliberate efforts by policymakers to make competition impossible, the competition rules rarely cede jurisdiction to other regulatory norms, and whereas competition law may tolerate the presence of regulation that generates market power for certain regulated actors, it tends to be unforgiving of autonomous behaviour that misuses such market power. Yet it is rare for competition assessment to disregard the regulatory context; where such cases do arise, this article has argued for a more nuanced understanding of what ‘normal’ competition looks like in such circumstances.

Yet regulation is not always just ‘part of the scenery’ in competition cases. Efforts to escape or frustrate regulation can constitute standalone antitrust infringements if carried out by an entity falling within the scope of competition law, whether due to dominance or acting in concert. This article has argued that the apparent outsourcing of antitrust liability standards within the non-compliance and regulatory gaming case-law provides an effective vehicle by which to interrogate and better appreciate the inter-relationship between competition law and other regulation. One way to understand these cases is to see the regulatory framework as setting the limits of fair competition in a marketplace, establishing the ‘rules of the game’ that all market actors can reasonably be expected to respect. While simple non-compliance or gaming of regulatory norms is arguably not a competition issue as such, it more plausibly becomes one where such conduct is the direct cause of appreciable anticompetitive effects. Another way to see these cases is as involving the ‘weaponisation’ of regulatory norms to anticompetitive ends as part of a deliberate strategy by regulated firms. From this perspective, the key antitrust consideration is what the defendant intends to achieve through its interaction with the regulatory framework. These perspectives are not mutually exclusive, moreover, and each finds some reflection in the existing case-law on non-compliance and regulatory gaming. Yet in many instances, the true explanation for punishing this conduct through competition enforcement is more pragmatic than principled: using competition law allows us to sidestep limitations or gaps within the regulatory framework itself. These cases thus involve an almost complete convergence between the goals of competition law and other market regulatory norms, and – for better or worse – take the legitimate scope of competition law to its outer limits.

1. * Law School, London School of Economics. Email: n.m.dunne@lse.ac.uk. Thank you to LSE colleagues who gave helpful comments on an earlier draft of this presentation at a staff seminar, in particular Tarun Khaitan and Emmanuel Voyiakis; to two anonymous referees who similarly provided very useful feedback; and to Current Legal Problems editor Deni Mantzari for her advice, support and expert shepherding of this piece through the review process.

   See, eg, Cases C-322/81 Michelin EU:C:1983:313, para 10, C‑42/21 P Baltic Rail EU:C:2023:12, para 86 and C‑252/21 Meta Platforms EU:C:2023:537, para 46. Also recital 11 of Regulation (EU) 2022/1925 on contestable and fair markets in the digital sector (Digital Markets Act) [2022] OJ L265/1. ↑

2. As explained in Section 2, the term ‘regulation’ can be defined as broadly as ‘all forms of economic or social influence’ or as narrowly as ‘a specific set of commands’ imposed by the State: see text accompanying nn 26 and 27. ↑

3. As the General Court recognised in Case T-168 GSK EU:T:2006:265, para 104. ↑

4. See, for example, the OECD’s Competition Assessment Toolkit (2019), which is designed to assist governments in identifying and reducing barriers to competition and other market distortions which may stem from State regulation. ↑

5. Article 7 TFEU states that ‘[t]he Union shall ensure consistency between its policies and activities, taking all of its objectives into account and in accordance with the principle of conferral of powers’. There is, however, limited practice applying the provision: N Nic Shuibhne, ‘Deconstructing and Reconstructing Article 7 TFEU’ in F Ippolito, ME Bartoloni and M Condinanzi (eds), The EU and the Proliferation of Integration Principles under the Lisbon Treaty (Routledge 2018). ↑

6. See eg P Ibáñez Colomo, The New EU Competition Law (Hart Publishing 2023) ch 3. ↑

7. See eg E Fox and E Healey, ‘When the State Harms Competition – The Role for Competition Law’ (2014) 79 Antitrust law Journal 769. ↑

8. See T Khaitan and S Steel, ‘Theorising Areas of Law: A Taxonomy of Special Jurisprudence’ (2022) 28 Legal Theory 325, 329–34. ↑

9. This debate has several dimensions: a normative component, seeking to determine the optimal goals (eg E Deutscher, ‘The Competition–Democracy Nexus Unpacked – Competition Law, Republican Liberty, and Democracy’ (2022) 41 YEL 197); an explicatory aspect, seeking to describe how the competition rules absorb and reflect different policy objectives (eg A Ezrachi, ‘Sponge’ (2017) 5 JAE 49); and an empirical aspect, seeking to identify the goals actually pursued in enforcement practice (see, eg, M Iacovides and K Stylianou, ‘The New Goals of EU Competition Law: Sustainability, Labour Rights, and Privacy’ (2024) 3 European Law Open 587). ↑

10. See n 1. ↑

11. O Odudu, ‘The Meaning of Undertaking within Article 81 EC’ (2005) 7 Cambridge Yearbook of European Legal Studies 211, 212. ↑

12. See Opinion of Advocate General Maduro in Case C-205/03 P FENIN EU:C:2005:666, para 26, on the distinction between ‘the power of the State’, which is subject to ‘democratic control’, and the actions of ‘economic operators’, which fall within the purview of competition law. ↑

13. Article 3(3) TEU. ↑

14. Council Regulation 139/2004 of 20 January 2004 on the control of concentrations between undertakings [2004] OJ L24/1. ↑

15. See eg discussion in Case C-307/18 Generics (UK) EU:C:2020:52, paras 87, 90 and 111 (referring to Article 101) and para 152 (referring to Article 102). ↑

16. Case C-85/86 Hoffmann La Roche EU:C:1979:36, para 91; also use of the term in C‑252/21, Meta Platforms EU:C:2023:537, para 47. ↑

17. European Commission, Guidelines on the application of Article 102 of the Treaty on the Functioning of the European Union to abusive exclusionary conduct by dominant undertakings (hereafter ‘Draft Article 102 Guidelines’), August 2024, paras 47–57. Discussing the role the concept plays within contemporary case-law, see P Ibáñez Colomo, ‘Competition on the Merits’ (2024) 61 CMLRev 387. ↑

18. So, for example, the need to ‘meet competition’ is generally not considered an acceptable defence to prima facie restrictive behaviour, see eg Case T-671/19 Qualcomm EU:T:2024:626, paras 598–99. ↑

19. The principal exception is the prohibition on hard-core cartel behaviour, where contextual analysis is ‘limited to what is strictly necessary’ to understand the anticompetitive nature of the conduct: Case C‑373/14 P Toshiba EU:C:2016:26, para 29. ↑

20. See eg Cases C-228/18 Budapest Bank EU:C:2020:265 and C-413/14 P Intel EU:C:2017:632. ↑

21. Provided for formally by the Article 101(3) TFEU exception, with the possibility of ‘objective justification’ read into the Article 102 TFEU jurisprudence, see eg Case C‑209/10 Post Danmark EU:C:2012:172. ↑

22. For instance, specific competition rules exist for agriculture pursuant to Articles 42 and 43 TFEU. ↑

23. See discussion of the distinction between ex ante and ex post enforcement in Case C-449/21 Towercast EU:C:2023:207. ↑

24. Case C-376/20 P CK Telecoms EU:C:2023:561, paras 84–86. ↑

25. While there is scope to agree to behavioural remedies under the EUMR, this approach is disfavoured, at least in principle: Commission notice on remedies acceptable under Council Regulation 139/2004 [2008] OJ C267/1. ↑

26. R Baldwin, M Cave and M Lodge, Understanding Regulation: Theory, Strategy and Practice (2nd edn, OUP 2011) 3. ↑

27. Regulation need not adopt a classic top-down ‘command and control’ form to come within this understanding, but pure self-regulation by undertakings is treated simply as economic activity within the scope of the competition rules in EU law, see eg the professional rules in Case C-309/00 Wouters EU:C:2002:98 and the industry reorganisation efforts in Case C-209/07 BIDS EU:C:2008:643. For discussion of ‘decentred’ approaches to regulation, see J Black, ‘Decentring Regulation: Understanding the Role of Regulation and Self-Regulation in a “Post-Regulatory” World’ (2001) 54 Current Legal Problems 103. ↑

28. Indeed, viewed from a regulatory theory perspective, competition law can be seen as a variety of regulation that falls within the broad category of ‘market-harnessing controls’ (Baldwin and others (n 26) 114–16). The competition jurisprudence, however, distinguishes the competition rules from ‘other’ regulation that may impact market dynamics, and that is the approach adopted within this piece. ↑

29. Digital Markets Act, Recital 11 (see also cases in n 1). ↑

30. For example, Directive 2019/944 on common rules for the internal market for electricity (OJ L158/125), recital (2), included as its aims: ‘to deliver real choice for all Union final customers, [] new business opportunities, competitive prices, efficient investment signals and higher standards of service, and to contribute to security of supply and sustainability.’ ↑

31. Case C-280/08 P Deutsche Telekom EU:C:2010:603. ↑

32. Opinion in Case Deutsche Telekom EU:C:2010:212, para 21. ↑

33. Case C-280/08 P Deutsche Telekom EU:C:2010:603, paras 80–84; reaffirmed in Case C-220/24 Regia Autonomă Aeroportul Internaţional ‘Avram Iancu’ Cluj v Consiliul Concurenţei EU:C:2025:124, para 27. ↑

34. Indeed, in Case C-382/22 P Cathay Pacific Airways v Commission EU:C:2026:129, para 125, the Court of Justice stated that the State action defence ‘has been only partially accepted’ in its jurisprudence. ↑

35. C-280/08 P Deutsche Telekom EU:C:2010:603, para 81. ↑

36. For a recent example where the State action defence actually succeeded on the facts, see Case T-136/19 Bulgarian Energy Holding EU:T:2023:669. ↑

37. Verizon Communications v Law Offices of Curtis V. Trinko, LLP, 540 US 398 (2003). The complexities of the position within US law are discussed in H Shelanski, ‘The Case for Rebalancing Antitrust and Regulation’ (2011) 109 Michigan Law Review 683. ↑

38. Case T-398/07 Spain v Commission EU:T:2012:173, para 55. ↑

39. Opinion in Case Deutsche Telekom EU:C:2010:212, para 25. ↑

40. In Case C-117/20 bpost EU:C:2022:202, the Court recognised an ‘objective of general interest’ in ensuring the effective application of competition law and other market regulation in parallel, ‘since they are pursuing [] distinct legitimate objectives’ (para 50, emphasis added). ↑

41. See, eg, Cases C-453/99 Courage and Crehan EU:C:2001:465, para 26 and C-74/14 Eturas EU:C:2016:42, paras 35–37. ↑

42. Case C-280/08 P Deutsche Telekom EU:C:2010:603. ↑

43. Case C-209/07 BIDS EU:C:2008:643. ↑

44. As was the case in support of the Siemens/Alstom merger, which was nonetheless prohibited by the Commission (M.8677, 6 February 2019). ↑

45. See Ursula von der Leyen, European Commission Political Guidelines 2024–29: ‘I believe we need a new approach to competition policy, … more supportive of companies scaling up in global markets’. See also Barbara Moens, ‘EU to Relax Merger Rules in Bid to Create ‘European Champions’ Financial Times (16 April 2026). ↑

46. Cases T‑321/05 AstraZeneca EU:T:2010:266, para 356 and T‑814/17 Baltic Rail EU:T:2020:545, para 189. ↑

47. Case C-209/07 BIDS EU:C:2008:643, para 21. ↑

48. Case C‑307/18 Generics (UK) and Others EU:C:2020:52. ↑

49. The Commission may take account of evidence of how the merged entity intends to act on the market going forward, but it places greater emphasis on whether that behaviour is anticipated to cause anticompetitive effects in fact: see eg Case M.11382—AGCO/TRIMBLE/JV (decision of 25.03.2024), for an example where the Commission concluded that distortive post-merger behaviour would not significantly impede effective competition in the circumstances. ↑

50. As the US Supreme Court said in Bell Atlantic v Twombly 550 U.S. 544 (2007), ‘resisting competition is routine market conduct’. ↑

51. This is a core tenet of competition policy, with roots that can be traced back to Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations (1776). For an argument that Smith’s influence should extend beyond this ‘libertarian’ vision, see S Makris, ‘A Smithian Political Economy Approach for the Competition Law of the 21st Century’ (2025) 88 Modern Law Review 712. ↑

52. Article 23, Regulation 1/2003. ↑

53. Commission Guidelines on the method of setting fines imposed pursuant to Article 23(2)(a) of Regulation No 1/2003 [2006] OJ C210/2. ↑

54. C‑457/10 P AstraZeneca EU:C:2012:770, para 132; reaffirmed in Case C-220/24 Aeroportul Cluj EU:C:2025:124, para 28. ↑

55. Cases C-295/12 P Telefónica EU:C:2014:2062, para 133 and C-220/24 Aeroportul Cluj EU:C:2025:124, para 28. ↑

56. Case C-280/08 P Deutsche Telekom EU:C:2010:603, para 90, applying the principle developed in Case C‑344/98 Masterfoods EU:C:2000:689, para 48. ↑

57. On the basis that only the Commission is empowered to find that there has been no breach of Article 102, whereas national regulators can only find breach or discontinue proceedings: Case C-375/09 Tele2 Polska EU:C:2011:270. ↑

58. Case C-280/08 P Deutsche Telekom EU:C:2010:603, para 89. ↑

59. Case C‑457/10 P AstraZeneca EU:C:2012:770, para 132. ↑

60. Case C-238/05 Asnef-Equifax EU:C:2006:734, para 63. ↑

61. Guidelines on the assessment of non-horizontal mergers under the Council Regulation on the control of concentrations between undertakings [2008] OJ C265/6, para 46. ↑

62. M.8124—Microsoft/LinkedIn, 6 December 2016. ↑

63. European Commission, ‘Mergers: Commission Clears Acquisition of Fitbit by Google, Subject to Conditions’ (Press Release, 17 December 2020). ↑

64. Case C-255/22 P Orlen EU:C:2024:790, para 96. ↑

65. ibid, para 96. ↑

66. ibid, para 97. ↑

67. ibid, para 96. ↑

68. ibid, para 95 (emphasis added). ↑

69. To use the language of AG Rantos in his Opinion in Case C-255/22 P Orlen EU:C:2024:466, para 58. ↑

70. See n 9. ↑

71. Cases C-204/00 P etc Aalborg Portland EU:C:2004:6, para 338. ↑

72. Case C-117/20 bpost EU:C:2022:202, paras 28–37. ↑

73. ibid, para 41. ↑

74. ibid, para 49. ↑

75. ibid, para 51. ↑

76. Case C-333/21 Superleague EU:C:2023:1011, para 130 (discussing specifically Article 102 TFEU). ↑

77. ibid, para 166 (discussing specifically Article 101 TFEU). ↑

78. See text accompanying n 33. ↑

79. Case C-280/08 P Deutsche Telekom EU:C:2010:603, para 224. ↑

80. ibid. ↑

81. To quote the defendant’s argument in Case C-280/08 P Deutsche Telekom EU:C:2010:603, para 205. ↑

82. Both the defendant and its rivals used revenues from call services to cross-subsidise their loss-making activity in the provision of line access services. The Commission, however, applied the margin squeeze test only to the defendant’s wholesale and retail-level line access activities. This approach was approved by the Court of Justice, which held that the Commission was entitled to proceed as if tariff rebalancing had occurred as required by EU telecommunications law, even though the German regulator had failed to implement tariff rebalancing for social policy reasons (to protect access to telephone lines for poorer consumers). ↑

83. Cases C‑165/19 P Slovak Telekom EU:C:2021:239, para 57 and C‑42/21 P Baltic Rail EU:C:2023:12, para 88. ↑

84. See Cases C-220/24 DB Station & Service EU:C:2022:832, para 82, T-136/19 BEH EU:T:2023:669, paras 784 and 961, and C-220/24 Aeroportul Cluj EU:C:2025:124, para 31. ↑

85. The Bundeskartellamt applied only domestic competition law, a choice that has been criticised as potentially inconsistent with its obligations under Regulation 1/2003. By the time the case made it to the Court of Justice, the Court spoke in terms of the application of EU competition law. ↑

86. Regulation 2016/679 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data [2016] OJ L119/1. ↑

87. Case C‑252/21 Meta Platforms EU:C:2023:537, paras 44–46. ↑

88. ibid, para 47 (emphasis added). ↑

89. The Court stated that, ‘in the context of the examination of an abuse of a dominant position by an undertaking on a particular market, it may be necessary for the competition authority … also to examine whether that undertaking’s conduct complies with rules other than those relating to competition law, such as’ – but by implication not limited to – ‘the rules on the protection of personal data laid down by the GDPR’: ibid, para 48 (emphasis added). ↑

90. ibid, para 50. ↑

91. ibid, para 51. ↑

92. See text accompanying n 75. ↑

93. Case C‑252/21 Meta Platforms EU:C:2023:537, para 55. ↑

94. As in eg Cases C-205/03 P FENIN EU:C:2006:453 and T-155/04 SELEX EU:T:2006:387, respectively. ↑

95. Case C-359/95 P Ladbroke Racing, para 33. ↑

96. Case C-333/21 Superleague EU:C:2023:1011, para 183. ↑

97. Case C-309/99 Wouters EU:C:2002:98, para 9. ↑

98. See n 94 above. ↑

99. Contrast Case C‑333/21 Superleague EU:C:2023:1011, paras 183–188, where the Court of Justice rejected the application of the Wouters exception to an organisation governed solely by private law (albeit not on that basis expressly). ↑

100. See nn 34 and 35 above. ↑

101. Draft Article 102 Guidelines, paras 167–71. ↑

102. Case C-233/23 Android Auto EU:C:2025:110, para 75. See also Case C‑42/21 P Baltic Rail EU:C:2023:12, which impliedly accepted that health and safety regulatory requirements could provide an objective justification for a de facto refusal to grant access to infrastructure, although the claimed necessity was deemed to be pretextual on the facts. ↑

103. Draft Article 102 Guidelines, para 168. ↑

104. ibid. ↑

105. As in eg Cases C‑42/21 P Baltic Rail EU:C:2023:12 and C‑457/10 P AstraZeneca EU:C:2012:770, respectively. ↑

106. Discussing the impact of privatisation specifically, see Case C‑245/24 Lukoil Bulgaria EU:C:2025:987. ↑

107. As in the E.ON/GDF pipeline cartel (Case COMP/39.401—E.ON/GDF, decision of 8 July 2009). ↑

108. As in Case C-331/21 EDP—Energias de Portugal EU:C:2023:812; in EDP, this fact was treated as evidence of the anticompetitive object of the arrangement contrary to Article 101(1) (see para 102). ↑

109. As acknowledged in Case C‑252/21 Meta Platforms EU:C:2023:537, para 47. ↑

110. See, eg, Cases C-413/14 P Intel EU:C:2017:632 and C-333/21 Superleague EU:C:2023:1011. ↑

111. Case C‑252/21 Meta Platforms EU:C:2023:537, para 47. ↑

112. Case C‑48/22 P Google Shopping EU:C:2024:726, para 224. ↑

113. ibid, para 225 (the Court expressly rejected Google’s argument that the Commission should have conducted a counterfactual analysis to find causation, see para 228). ↑

114. Including in the fields of airports, telecommunications, energy and rail transport. ↑

115. Case C‑165/19 P Slovak Telekom EU:C:2021:239. ↑

116. See, eg, Cases C‑501/06 P etc, GSK EU:C:2009:610, paras 102–04. ↑

117. For a recent study of such missing case-law, see M Heim, ‘The Curious Case of the European Commission’s Missing Antitrust Jurisprudence: Lessons from Abandoned Article 102 Investigations’ (2025) 16 JECLAP 285. ↑

118. Case M.8124—Microsoft/LinkedIn, 6 December 2016, para 177. ↑

119. See n 55. ↑

120. As in Cases 39.525—Telekomunikacja Polska (22 June 2011), C‑165/19 P Slovak Telekom EU:C:2021:239 and C‑42/21 P Baltic Rail EU:C:2023:12. ↑

121. S Dogan and M Lemley, ‘Antitrust Law and Regulatory Gaming’ (2009) 87 Texas Law Review 685. ↑

122. Case COMP/A.37.507/F3—AstraZeneca, 15 June 2005; upheld on appeal in Cases T‑321/05 AstraZeneca EU:T:2010:266 and C-457/10 P AstraZeneca EU:C:2012:770. ↑

123. Case AT.40588—Teva, 31 October 2024. ↑

124. Case AT.40394—Aspen Pharma, 10 February 2021. ↑

125. Case AT.40134—AB InBev Beer Trade Restrictions (29 June 2016). ↑

126. Case C-179/16 F. Hoffmann-La Roche and Others EU:C:2018:25. ↑

127. Case AT.40178—Car Emissions, 8 July 2021. ↑

128. Including the Commission decisions in Cases AT.39226 Lundbeck and AT.39612 Servier and the preliminary ruling in Case C-307/18 Generics (UK) EU:C:2020:52. ↑

129. See eg M Moore & D Tambini (eds), Regulating Big Tech. Policy Response to Digital Dominance (OUP 2022). ↑

130. See eg M Gal and O Aviv, ‘The Competitive Effects of the GDPR’ (2020) 16 JCLE 349 and CB Frey & G Presidente, ‘Privacy Regulation and Firm Performance: Estimating the GDPR Effect Globally’ (2024) 62 Economic Inquiry 1074. ↑

131. Recital 9, GDPR, which states that uneven implementation of the previous EU data protection regime could ‘distort competition’. ↑

132. See n 91. ↑

133. As in Cases C‑457/10 P AstraZeneca EU:C:2012:770 and C-591/16 P Lundbeck EU:C:2021:243, for example. ↑

134. As in the E.ON/GDF gas pipeline cartel, for example (see in particular, the judgment of the General Court on appeal, Case T-370/09 GDF Suez v Commission EU:T:2012:333. ↑

135. Ezrachi (n 9). ↑

136. I Lianos, ‘Polycentric Competition Law’ (2018) 71 Current Legal Problems 161. ↑

137. For a more detailed critique of the adequacy of this approach, see N Dunne, Competition Law and Economics Regulation (CUP 2015) ch 4. ↑

138. In Deutsche Telekom, for instance, the Court dismissed the significant impact of the national regulatory framework on the defendant’s behaviour on the rather blithe ground that it could have sought to petition the regulator to adopt more cost-reflective price regulation (in spite of evidence that the regulator had deliberately chosen to prioritise social solidarity over efficiency in its regulatory policymaking). For further critique of this case from a fairness perspective, see Dunne (n 137) 224–27. ↑

139. This was a significant theme within Deutsche Telekom: in such circumstances, the Court took a generous view of the Commission’s discretion to initiate competition law proceedings against the undertaking or infringement proceedings against the Member State maintaining the anticompetitive regulation. ↑

140. See n 69. ↑

141. As the General Court remarked in Lundbeck, a pay-to-delay case, ‘it is in the public interest to eliminate any obstacle to economic activity which may arise where a patent was granted in error’: Case T-472/13 Lundbeck EU:T:2016:449, paras 119, 390 and 487. This is also an issue in the background in Case C‑457/10 P AstraZeneca EU:C:2012:770 and Case AT.40588—Teva, 31 October 2024 amongst other cases. ↑

142. As provided by Article 4 (independence), Article 5 (resources) and Articles 10 and 13–16 (remedies and fines) of Directive (EU) 2019/1 to empower the competition authorities of the Member States to be more effective enforcers and to ensure the proper functioning of the internal market (ECN+ Directive) [2019] OJ L11/3. ↑

143. See n 74. ↑

144. Case C‑252/21 Meta Platforms, para 54. ↑

145. Case C-117/20 bpost EU:C:2022:202, para 49. ↑

146. See n 67. ↑

147. GDPR, recital 4. ↑

148. See fn. 15. ↑

149. Case C-377/20 Servizio Elettrico Nazionale EU:C:2022:379, para 61. ↑

150. Draft Article 102 Guidelines, para 45. ↑

151. ibid, para 60. ↑

152. For example, data protection rules limit the extent to which firms may profitably combine data-sets containing personal information, while environmental protection rules limit the extent to which firms can use cheaper but more polluting technologies. ↑

153. For example, EU employment law prevents firms from dismissing pregnant employees and requires the provision of paid leave. ↑

154. On regulatory compliance costs generally, see OECD, OECD Regulatory Compliance Cost Assessment Guidance (OECD Publishing 2014). ↑

155. For discussion of the level-playing-field notion generally, see eg M Gillis, ‘Let’s Play?: An Examination of the “Level Playing Field” in EU Free Trade Agreements’ (2021) 55 Journal of World Trade Law 715. ↑

156. Case C‑165/19 P Slovak Telekom EU:C:2021:239, paras58–60, and C‑42/21 P Baltic Rail EU:C:2023:12, para 89. See also Case T-136/19 BEH EU:T:2023:669, para 873. ↑

157. Commission Guidelines on the applicability of Article 101 to horizontal co-operation agreements [2023] OJ C259. ↑

158. ibid, para 520. ↑

159. ibid, para 564. ↑

160. ibid, para 520. ↑

161. C‑252/21 Meta Platforms EU:C:2023:537, para 51. ↑

162. As in Cases C‑457/10 P AstraZeneca EU:C:2012:770 and AT.40588—Teva, 31 October 2024. ↑

163. Although coordination or dominance does not attract antitrust liability itself, the jurisprudence emphasises that such firms have heightened obligations: coordinating firms must ensure that each still ‘determine[s] independently the policy which he intends to adopt on the … market’ (Case C-40/73 Suiker Unie EU:C:1975:174, para 173) while dominant firms have a distinctive ‘special responsibility’ to avoid distorting competition (Case C-322/81 Michelin EU:C:1983:313, para 10). ↑

164. The argument would be that such firms are expected to comply both with their regulatory obligations and to refrain from practices that might reasonably be considered at odds with the broad thrust of a regulatory scheme. An analogy might be drawn to the concept of potential competition, which is established where there are ‘real and concrete possibilities’ of new entry (see eg Case C‑307/18 Generics (UK) EU:C:2020:52, para 36). This standard can be met even if there is considerable ambiguity as to whether new entrants can lawfully compete in a market given existing regulatory barriers to entry. In pay-to-delay cases, for instance, competition law does not tolerate agreements to exclude potential competitors even if there is a reasonable chance that patent law might facilitate this result: for instance, in Case T‑472/13 Lundbeck EU:T:2016:449, paras 121–31, the court found potential competition where there was only a 50–60 per cent chance that the would-be competitor could lawfully enter the market concerned. ↑

165. Coordination falling within Article 101 could also act as a jurisdictional trigger for the application of competition law; a ‘collusion-plus’ approach would then treat coordinated non-compliance or gaming efforts as a restriction of competition. An example of this arose in the Car Emissions cartel, see n 127. ↑

166. The German competition authority also drew on national case-law on consumer protection to support its approach to the concept of market abuse. ↑

167. Ezrachi (n 9). ↑

168. See eg A Ayal, ‘The Market for Bigness: Economic Power and Competition Agencies’ Duty to Curtail It’ (2013) 1 JAE 221; L Khan, ‘The Ideological Roots of America’s Market Power Problem’ (2018) 127 Yale LJF 960; and A Gerbrandy and P Phoa, ‘The Power of Big Tech Corporations as Modern Bigness and a Vocabulary for Shaping Competition Law as Counter-power’ in M Bennett, H Brouwer and R Claassen (eds), Wealth and Power: Philosophical Perspectives (Taylor & Francis 2022). ↑

169. See n 163. ↑

170. Case C-85/86 Hoffmann La Roche EU:C:1979:36, para 91. ↑

171. Cases C‑165/19 P Slovak Telekom EU:C:2021:239, para 57 and C‑42/21 P Baltic Rail EU:C:2023:12, para 88. ↑

172. Case C‑252/21 Meta Platforms EU:C:2023:537, para 44. ↑

173. ‘[P]ublic authorities can legitimately choose complementary legal responses to certain conduct that is harmful to society through different procedures forming a coherent whole so as to address different aspects of the social problem involved’: Case C-117/20 bpost EU:C:2022:202, para 49. ↑

174. Exemplified by Case C-280/08 P Deutsche Telekom EU:C:2010:603. ↑

175. Under the ECN+ Directives, national competition authorities must have the ability to impose behavioural and structural remedies on defendants (Article 10), plus fines of up to at least 10 per cent of the annual worldwide turnover of the undertakings concerned (Article 15) for competition law violations. These provisions mirror the powers of the Commission under Regulation 1/2003. ↑

176. See Directive 2014/104/EU on certain rules governing actions for damages under national law for infringements of the competition law provisions of the Member States and of the European Union [2014] OJ L349/1. ↑

177. See eg Case C-413/14 P Intel EU:C:2017:632, para 138. ↑

178. See eg the discussion in Case C-307/18 Generics (UK) EU:C:2020:52. ↑

179. Surveying the case-law, see P Ibanez Colomo, ‘Anticompetitive Effects in EU Competition Law’ (2020) 17 JCLE 209. ↑

180. This requires it to be demonstrated that coordination ‘has as its actual or potential effect the prevention, restriction or distortion of competition, which must be appreciable’ (Superleague, para 169). Thus it is sufficient that the conduct merely limits potential (as opposed to existing) competition, although this does require it to be established that ‘real and concrete possibilities’ for new market entry exist (Case C‑307/18 Generics (UK) EU:C:2020:52, para 36). ↑

181. Relied upon, inter alia, in Roche & Novartis (n 126), Car Emissions (n 127) and various ‘pay-to-delay’ cases. ↑

182. Case C-333/21 Superleague EU:C:2023:1011, para 165. ↑

183. Case C-377/20 Servizio Elettrico Nazionale EU:C:2022:379, para 50. ↑

184. ibid, para 54. ↑

185. See n 113. ↑

186. See eg Case C-333/21 Superleague EU:C:2023:1011, para 163. ↑

187. See eg discussion in Case C-377/20 Servizio Elettrico Nazionale EU:C:2022:379, paras 53–54. ↑

188. Case C-413/14 P Intel EU:C:2017:632, para 139; followed in Case C-680/20 Unilever EU:C:2023:33. ↑

189. Case C‑48/22 P Google Shopping EU:C:2024:726, para 240. ↑

190. ibid, para 254. ↑

191. Case C-333/21 Superleague EU:C:2023:1011, para 167. ↑

192. ibid. ↑

193. Case C-680/20 Unilever EU:C:2023:33, para 45 (emphasis added). ↑

194. Case C‑48/22 P Google Shopping EU:C:2024:726, para 255. ↑

195. ibid, para 256. ↑

196. Named for Case C-62/86 Akzo v Commission EU:C:1991:286. ↑

197. The Akzo test is an off-shoot of the Areeda-Turner test, which was developed to provide an objective means to determine when low prices – typically considered desirable from a competition policy perspective – should nonetheless be found abusive by virtue of their exclusionary effects (see PE Areeda and DF Turner, ‘Predatory Pricing and Related Practices under Section 2 of the Sherman Act’ (1975) 88 Harvard Law Review 697). The Areeda-Turner test posits that prices below average variable cost (AVC, a proxy for margin cost) should be deemed conclusively abusive on the basis that no rationale firm would continue to produce if it cannot obtain prices at or above AVC, so that its choice to do so can be taken to disclose sufficient evidence of anticompetitive intention. The Akzo case extended the Areeda-Turner test to prices above AVC but below average total cost where there is evidence that such prices are ‘determined as part of a plan for eliminating a competitor’ (C-62/86, para 72), on the basis that such prices have the capacity to exclude as-efficient competitors, while the actual intention evidence serves to counter any potentially innocent explanations for the pricing behaviour. ↑

198. See n 51 and accompanying text. ↑

199. Including AstraZeneca, Teva, Aspen Pharma, Telekom Polska, Lundbeck and the quite fantastical case of Baltic Rail, where the defendant opted to destroy 19km of its own rail infrastructure rather than share it with a freight rail competitor under the EU rail liberalisation framework. ↑

200. See n 46. ↑

201. Case C‑457/10 P AstraZeneca EU:C:2012:770, para 99. ↑

202. Case AT.40588—Teva, 31 October 2024. ↑

203. ibid, para 1033. ↑

204. ibid, para 1033, citing Case T-612/17 Google Shopping EU:T:2021:763, para 257. ↑

205. Opinion in Case C-413/14 P Intel v Commission EU:C:2016:788, para 66. ↑

206. Ezrachi (n 9) 50. ↑

1. Introduction

What roles do regulatory norms play in EU competition law assessment? The question matters both for the consistency of the supervisory framework for economic activity in the EU, and because it helps us to better understand the legitimate concern of the competition rules, a key theme in contemporary competition policy debates. Competition law (also known as antitrust) aims to promote and protect effective competitive behaviour by economic actors – the jurisprudence uses the term ‘undistorted’^[1] competition to describe the ultimate policy objective. Regulation, a broader and more ambiguous concept,^[2] often applies to and constrains the same market conduct, though typically in pursuit of alternative or additional policy objectives. Regulation can enhance opportunities for effective competition or limit the scope for abusive market behaviour; but it can also ‘distort’^[3] competition in ways that raise concerns from a competition policy perspective, whether by diminishing overall competitive dynamics or by facilitating or incentivising anticompetitive conduct.^[4] EU law has long endorsed the concurrent application of the competition rules to regulated firm behaviour, an approach that maximises the effectiveness of competition policy. Yet the jurisprudence struggles with a problem of internal coherence, seeking to protect the distinctiveness and primacy of the competition rules while also finding space for the increasing centrality of contextual analysis in contemporary competition enforcement. Moreover, applying competition law to behaviour that has already been affected by other regulatory norms creates its own challenges, including issues of legitimacy, consistency and proportionality. These questions are the focus of this piece.

In principle, the competition rules apply only to the autonomous behaviour of firms, and such conduct – whether taking the form of agreements, dominant firm practices or mergers – is prohibited only where it is demonstrated that anticompetitive effects – whether actual, potential or presumed – are likely to result. Yet where competition law is applied to regulated behaviour, there is a higher likelihood that suspect conduct has been influenced or even directed by the regulatory obligations and that any resulting market distortions can be attributed, wholly or partly, to the impact of the regulation. In such contexts, competition law assessment must consider not only whether and how the defendant’s behaviour has caused anticompetitive harm, but also the extent to which that behaviour and its effects were determined by the underlying regulatory framework. The greater the overlap between the scope of the regulatory obligations and the competition law theory of harm, the more complex this task becomes. Since competition law and regulation must co-exist as legitimate and socially valuable mechanisms of market control, there are arguments, both practical and legal,^[5] for ensuring that these separate regimes are interpreted and applied in a consistent, ideally mutually reinforcing, manner. But because this question typically arises where competition law is applied in markets already subject to ex ante regulation, there can be tension between preserving the effectiveness of competition law and its underlying goal of protecting undistorted competition, while simultaneously respecting other public interest values pursued by ‘distortive’ regulation.

A great deal has already been written about the relationship between competition law and other regulation, including much of my own work. This article seeks to make an original contribution by focusing on how substantive regulatory norms are accommodated within competition law analysis and what this tells us about competition law’s place within the wider market regulatory sphere. We leave aside the distinct questions of when competition law itself exhibits certain ‘regulatory’ characteristics,^[6] and of the appropriate competition policy response to anticompetitive State regulation.^[7] In doing so, the article adopts an essentially nonnormative, conceptual approach, which aims to work effectively within the confines of the existing jurisprudence rather than making more aspirational claims for wholesale change.^[8] Our analysis is not directly concerned with the goals of competition law, although there are plenty of articles of this sort about.^[9] But when considering how competition law interacts with other regulatory fields that pursue a more diverse range of policy objectives, it is impossible to avoid the question of what competition law itself seeks to achieve. Above, we described the goal of competition law broadly as the protection of ‘undistorted’ competition, an approach that, although not without its critics, finds support in recent jurisprudence and legislation.^[10] The article thus seeks to provide a more granular exploration of what we mean by undistorted competition, in a context where markets can never really be free of distortions, whether stemming from private or public sources. By considering the extent to which other regulatory norms can provide a conduit by which to reflect a wider range of public interest values within competition enforcement, while also recognising the legal and practical limits on this approach, we seek to obtain a better understanding of what Odudu termed ‘the legitimate scope’^[11] of competition law.

The article is structured as follows. Section II introduces and distinguishes competition law from other regulatory instruments of market supervision. Section III sets out and analyses the ‘orthodox’ understanding of the inter-relationship, which prioritises the independence and effectiveness of competition law as a mechanism of market control. Section IV discusses the more ‘contextual’ contemporary approach, considering the ways in which regulation may provide a relevant element of context within competition law analysis. Section V steps back, to consider the normatively-oriented question of how the ‘outsourcing’ of liability standards in competition law to other regulatory fields can be explained. It examines a variety of perspectives: the effectiveness of competition law as a regulatory enforcement mechanism; regulatory norms as a constraint on the acceptable parameters of competition on the merits; and regulatory (non-)compliance as a positive anticompetitive action. Section VI brings these disparate strands of analysis together and concludes.

2. Distinguishing competition law from ‘regulation’ as instruments of market supervision

A necessary starting point is to distinguish competition law from other forms of market regulation. Competition law comprises a set of legal rules that aim to ensure open, undistorted and fair competition in the internal market, by controlling the acquisition and exercise of market power (as distinct, inter alia, from public power).^[12] Within the framework of the EU treaties, the competition rules operate as important flanking provisions to secure the objective of ‘a highly competitive social market economy’.^[13] Yet most competition cases do not belabour this point. Instead, enforcement tends to be a relatively technical exercise aimed at identifying and proscribing various agreements (under Article 101 of the Treaty on the Functioning of the European Union, or TFEU), practices of dominant undertakings (under Article 102 TFEU) and mergers (under the EU Merger Control Regulation, or EUMR)^[14] that are deemed restrictive of competition in some legally relevant sense. A key theme within recent case-law is the notion of ‘competition on the merits’:^[15] although competition law is aimed at anticompetitive firm behaviour, the rules do not seek to prevent firms, even where they are dominant or acting in concert, from competing ‘on the merits’ (what is also termed ‘normal competition’^[16]). What amounts to normal or meritorious competition in an individual case is a question of competition law. It is thus determined by applying the legal tests and principles developed in the competition jurisprudence,^[17] instead of deferring to the behavioural or regulatory norms within a sector.^[18]

Contemporary EU competition law has several salient aspects for our purposes. First, competition assessment is, in almost every instance,^[19] a highly context-specific exercise: conduct is held to restrict competition within its market circumstances,^[20] while also taking account of any countervailing pro-competitive efficiencies that might be generated.^[21] Second, and somewhat by contrast, competition law is largely sector-indifferent in its application. With a few exceptions,^[22] the competition rules apply across all sectors, supervising vastly different industry practices under the same legal provisions. Finally, the competition rules governing anticompetitive agreements and dominant firm practices are largely ex post in application, serving to punish infringing behaviour after-the-fact.^[23] The EUMR, conversely, is prospective in analysis,^[24] but the object of the exercise is not (generally)^[25] to regulate the behaviour of newly-merged firms going forward, but rather to catch and proscribe potentially harmful concentrations before these occur. In this article, we focus on the impact of pre-existing regulation on the application of Articles 101 and 102 TFEU to market behaviour, while considering insights from merger control where appropriate.

The term regulation is less precise in its meaning. It can be defined as broadly as ‘all forms of economic or social influence’ that may stem from public or private sources, or as a narrowly as ‘a specific set of commands’ imposed by the State and pertaining to closely identified activities.^[26] For the purposes of this article, we adopt a relatively loose definition, referring to State-imposed (or at least State-endorsed)^[27] norms that, directly or indirectly, control or constrain market activity by economic actors, broadly understood. This encompasses both sector-specific regulation, which applies to market participants that are active only within particular sectors (such as telecommunications, energy, transport or digital platform services), and to regulatory regimes of more general application (such as data protection, health and safety, or environmental law).

Applying this definition, it can readily be seen that most markets today are subject to multiple layers of regulation that, to borrow the language of the competition jurisprudence, may ‘distort’ competitive outcomes to a greater or lesser extent. In some instances, it is obvious how overlap with competition law arises: for example, regulation may grant an economic actor special or exclusive rights that give it a position of market dominance, or regulation may mandate pro-competitive behaviour (such as imposing access requirements) or proscribe direct abuses of market power (through, for example, pricing controls). In other instances, the potential for interaction with competition law arises more obliquely, for example where regulation constrains a firm’s ability to maximise its profits or provides opportunities for exploitative or exclusionary behaviour.

Given the breadth of our understanding of regulation, and by implication the variety of regulatory models that it encompasses, it is impossible to neatly distinguish competition law from other forms of market regulation that may apply to the same conduct.^[28] Two distinctions of relevance might, however, be advanced. First, although the precise ‘goals’ of EU competition law are debated, broadly speaking the rules aim at a straightforward objective, namely ‘the protection of undistorted competition’.^[29] While a regulatory regime may seek to enhance effective competition, this need not be its only policy objective, and most of the examples of regulation that overlap with competition law within the case-law have pursued other regulatory goals additionally or alternatively,^[30] some of which may sit uncomfortably with the efficiency-focused perspective of the competition rules. Second, one of the necessary trade-offs that comes from the wide applicability of competition law is that its prohibitions are sketched in broad and abstract terms. This creates flexibility and scope for progressive development of the law to address emerging market problems, but such advantages arise at the expense of ex ante precision and certainty. Some regulatory regimes operate in a similar manner, yet regulation can be more prescriptive, identifying with greater exactness the specific regulated actors or the content of mandatory norms. Specification can assist enforcement, enabling the supervision of complex market problems. Yet where regulated entities are eager to resist regulation, a precisely defined and limited set of rules may facilitate strategic behaviour to escape or exploit regulatory controls, a problem considered in Section 5.

The broad scope of the competition rules coupled with the pervasiveness of regulation in modern markets leads to overlapping jurisdiction and potential conflict. The jurisprudence demarcating the inter-relationship of competition law and other regulation within the EU legal system is complex. One strand of case-law emphasises the independence and effectiveness of competition law as a mechanism of market control, reflecting an ‘orthodox’ understanding of the inter-relationship that prioritises but also effectively siloes competition law. Another strand embraces the more contextual approach that has come to the fore in the general competition case-law, recognising the extent to which regulation may feed into and even determine the substance of competition law assessment. While these jurisprudential currents are not inherently at odds, they reflect notably different visions of this inter-relationship. There are also points of inconsistency and even conflict within the case-law, which remain underexplored in the existing jurisprudence. In what follows, we set out first the apparently resolute ‘orthodox’ position (Section 3) and then the more nuanced contextual approach (Section 4). Doing so identifies several cases in which it becomes more difficult to reconcile the orthodox and contextual accounts, which are explored in Section 5.

3. The interaction of competition law and regulation: the orthodox view

Within EU law, the orthodox position is that the competition rules are essentially distinct from and unaffected by other regulatory norms that may apply to the same market behaviour. This approach is exemplified by Deutsche Telekom, in which the defendant was held to have breached Article 102 in circumstances where it had complied with its sector-specific regulatory obligations, where the sector regulator had either set or endorsed the pricing practices that comprised the abuse, and where the national regulatory framework pursued a legitimate public policy objective (social solidarity) by endorsing a pricing structure that the Commission subsequently held to be inefficient.^[31] An analogy of ‘two barriers’ to lawful market participation was suggested by the Advocate General in Deutsche Telekom: economic actors must comply with the overarching obligations imposed by the competition rules, while simultaneously discharging any parallel regulatory obligations.^[32] Typically, moreover, these legal requirements are enforced by different regulators, who each decide independently whether the norms falling within their jurisdictions have been breached. This orthodoxy has several important implications.

A. No ouster of competition law in the presence of regulation

First, the existence of regulation within a market does not generally oust the applicability of competition law to the same behaviour, even if the regulatory regime has a significant impact on the conduct or underlying level of competitiveness within the market.^[33] The principal exception, the ‘State action’ doctrine, is narrowly construed,^[34] applying only where anticompetitive conduct is either required by regulation or where the regulatory framework removes all scope for competitive (and, by implication, anticompetitive) behaviour.^[35] The logic behind this defence is that competition law regulates anticompetitive behaviour by firms and not merely uncompetitive market structures; but its narrowness in practice suggests a deep reluctance to decouple the abstract notion of competition from the concrete actions of competitors.^[36]

This can be contrasted with the approach in US antitrust law, which more readily embraces a rule of antitrust preclusion where ‘a regulatory structure designed to deter and remedy anticompetitive harm’ can be identified.^[37] There are at least three reasons which explain the distinctive position to be found in EU law. First, the competition rules are primary EU law, meaning that there is a formal obstacle within the EU’s hierarchical legal structure to allowing ‘special’ regulatory rules found in secondary or national legislation to take precedence over the ‘general’ competition rules.^[38] Second, even if regulation takes account, inter alia, of the goal of effective competition, the assumption is that it almost always pursues ‘objectives which differ from those of [EU] competition policy’.^[39] The policy objectives of one set of rules cannot be subsumed into the other; with the result that, to the extent we wish to pursue both sets of goals simultaneously, both sets of rules must remain in operation.^[40] Finally, EU competition law incorporates a remarkably strong principle of effectiveness, which has aided the progressive development and expansion of these rules in many areas.^[41] Limiting any rule of antitrust preclusion to only the most exceptional circumstances where regulation removes all scope for (anti)competitive conduct serves to protect the effectiveness of competition law by ensuring that these rules continue to apply in the largest number of circumstances.

This means, for instance, that a dominant firm can be held to have engaged in an abusive margin squeeze (which entails manipulating the relationship between wholesale and retail prices for a product) where both price levels have been set or approved by a sector-specific regulator.^[42] Similarly, industry efforts to reorganise a sector suffering from an overcapacity crisis can be treated effectively as a cartel, even if undertaken at the behest of and in accordance with a plan devised by national government.^[43] And under the EUMR, strong governmental support has not conventionally provided a compelling counterweight where the Commission concludes that a proposed concentration is likely to significantly impede effective competition,^[44] although this position may prove more difficult to sustain in future.^[45]

The fact that, in EU competition law, defendants can breach the rules even where their conduct or the anticompetitive effects that follow are heavily influenced by State-imposed regulation is made possible by the lack of a fault or ‘bad’ intention requirement. Thus, under Article 102, ‘proof of the deliberate nature of the conduct and of the bad faith of the undertaking in a dominant position is not required for the purposes of identifying an abuse’.^[46] Under Article 101, it is ‘irrelevant’ to the assessment of whether a restriction by object exists that the coordinating parties acted in pursuit of ‘legitimate’ (ie not deliberately anticompetitive) objectives.^[47] The ‘by effects’ limb of that provision, additionally, considers only the impact in fact of an agreement on the market, rather than the reasons motivating its adoption.^[48] Intention also plays a very limited role in determining the essentially factual question of how a merger is likely to affect competition if it proceeds.^[49]

The absence of a fault requirement can be explained by the fact that the competition rules are aimed at behaviour that harms competition; yet the neoclassical economics that underpin the conventional competition law framework assumes that all firms ultimately intend to ‘harm’ competition in a literal sense by increasing their market share and profits at the expensive of competitors.^[50] Indeed, such self-interested behaviour has long been considered an essentially virtuous aim to the extent that it drives the competitive dynamics necessary to maximise efficiency.^[51] The upshot is that competition law takes an explicitly ‘objective’ approach, at least in principle, to the assessment of whether behaviour amounts to normal, meritorious – and so permissible—competition. To the extent that an absence of fault or bad intention is relevant within EU competition enforcement, it goes primarily to the penalty imposed: breaches must be committed intentionally or negligently to attract fines under Regulation 1/2003,^[52] while the influence of regulation on (otherwise autonomous) firm behaviour can be reflected in a discount on the overall fine.^[53]

B. Competition law and regulatory compliance are unrelated

Second, the question of whether behaviour violates competition law is formally ‘unrelated to its compliance or non-compliance with other legal rules’,^[54] regardless of how strong the link between the regulatory framework and the allegedly abusive behaviour. This means, on the one hand, that behaviour that conforms to a relevant regulatory standard might still breach competition law, if it amounts to an anticompetitive agreement or abuse of dominance.^[55] This position has been defended within EU law on the bases, inter alia, that decisions of (national) regulators should not bind the Commission,^[56] even if the regulator purports to apply and find compliance with the competition rules in parallel;^[57] and that the absence of a fault requirement, discussed above, means that good faith regulatory compliance is insufficient to avoid competition liability.^[58] The general principle that regulatory compliance is ‘unrelated to’ competition law liability would seem to infer that regulatory non-compliance is also largely irrelevant from an antitrust perspective, though Meta Platforms (discussed in Section 4.A) nuances this assumption.

Accordingly, the case-law recognises that ‘in the majority of cases, abuses of dominant positions consist of behaviour which is otherwise lawful under branches of law other than competition law’.^[59] The jurisprudence is more complicated under Article 101, insofar as the sanction of nullity in Article 101(2) means that agreements that violate the provision are void as a matter of contract law too. But even in this context, the Court considers that compliance with other regulatory norms is, generally, ‘not, as such, a matter for competition law’,^[60] and so does not determine the question of whether an arrangement has the object or effect of restricting competition. Under the EUMR, when assessing whether a merged entity is likely to engage in harmful behaviour, the Commission takes account of any pre-existing regulation that might effectively inhibit the merged entity from doing so.^[61] It has also sought, in certain instances, to repackage regulatory compliance as a dimension of quality, to the extent that users place economic value on the underlying public interest protected (such as the level of data privacy offered by a product).^[62] Yet the Commission is clear that, if the objection to a concentration is a fear the merged entity will fail to comply with other regulatory rules in future, ‘[s]uch concerns are not within the remit of merger control and there are regulatory tools better placed to address them’.^[63]

C. Constrained powers of Commission when acting as competition regulator

This links to a third implication of the orthodox separation of competition law and other regulation within EU law, namely the limits of the Commission’s powers when ‘acting as competition regulator’.^[64] In the recent case of Orlen, the Court of Justice emphasised that, when exercising its competition powers, the Commission is restricted to imposing remedies or penalties that ‘may respond to the competition concerns [it has identified] on the market concerned’.^[65] Thus the Commission cannot legitimately use competition enforcement to advance other regulatory goals or interests,^[66] unless these can be absorbed into the ‘competition issues’ at stake.^[67] The Court acknowledged that the Commission’s competition enforcement activity ‘could not lead to a result which is contrary to’ other regulatory norms within EU law,^[68] and in particular, it cannot make binding remedies that would conflict with other (EU-level, at least) regulation. But that it quite different from saying that the Commission might positively impose ‘obligations … motivated by policies other than competition law policies’^[69] through the avenue of competition enforcement; this is something, the Court was clear, the Commission cannot do.

The attitude of the Court in Orlen potentially poses a not-insignificant significant hurdle to realising more ambitious applications of the competition rules,^[70] by establishing that competition law enforcement can pursue only ‘competition issues’. Of course, there is a risk of circularity, insofar as what we consider to legitimately constitute a ‘competition issue’ may vary depending upon how broadly we treat the legitimate objectives of competition law. But the judgment clarifies a point of importance to this article: competition enforcement is not an instrumental tool by which to advance a grab bag of ‘other’ regulatory objectives but must confine itself to responding to competition-related concerns.

D. Ne bis in idem and parallel enforcement of competition law and regulation

Finally, this orthodox understanding of the relationship between competition law and other regulatory norms – as discrete sources of legal obligations for market actors, applied by distinct regulators pursuing different public interest goals – creates potential for overlapping jurisdiction and enforcement. Compliance with regulatory obligations does not, as explained, certify that the regulated behaviour also complies with competition law. Yet overlapping jurisdiction does not always generate conflict, and it is entirely possible that the same behaviour might be found to violate both the competition law and the regulatory standards.

For many years, EU competition law recognised a distinctive principle of double jeopardy or ne bis in idem, which precluded the parallel application of multiple sanctions only if it was demonstrated that the regulatory regime pursued the same legal interest as the competition rules, in addition to unity of facts and offender.^[71] This unique approach reinforced the orthodox view of competition law and regulation as distinct and unrelated tools for market supervision. More recently, the Grand Chamber in bpost has realigned competition law with other areas of EU law by shifting to an idem factum standard requiring only the same facts and offender.^[72] The Court nonetheless recognised that duplicate proceedings and penalties might be permitted where they are a necessary and proportionate response to achieve public interest objectives or to protect the rights and freedoms of others.^[73] Of particular relevance is the Court’s strong assertion, when considering the proportionality of parallel proceedings, that ‘public authorities can legitimately choose complementary legal responses to certain conduct that is harmful to society through different procedures forming a coherent whole so as to address different aspects of the social problem involved, provided that the accumulated legal responses do not represent an excessive burden for the individual concerned’.^[74] Yet parallel jurisdiction cannot be exercised wholly unilaterally if it is to comprise a proportionate response. Instead, the Court in bpost stressed the need for sufficient ‘coordination’ between the competition regulator and other prosecuting authority to ensure a truly ‘complementary’ response.^[75]

4. The interaction of competition law and regulation: the contextual view

One shortcoming of the ‘two barriers’ understanding of the interaction of competition law and other regulation is that it sets up a dichotomy between these instruments of market supervision that does not reflect the reality of how most markets work today. In its competition case-law, the Court of Justice increasingly emphasises that the determination of whether conduct amounts to a breach of the rules requires a detailed assessment of context, meaning that decision-makers must take account of ‘all of the relevant factual circumstances’^[76] and the ‘economic and legal context’^[77] in which allegedly anticompetitive conduct occurs. The underlying regulatory landscape, insofar as it may shape competition dynamics and/or influence the defendant’s own behaviour, is an obvious contextual element that should be considered. Although EU law rejects the proposition that the presence of specific regulation should oust the more general competition rules as a matter of law, the former may play a significant role in determining whether competition law is breached on the facts.

A. Relevance of regulatory norms in principle: from Deutsche Telekom to Meta Platforms

Indeed, even as the Court in Deutsche Telekom sidelined the State action defence and discounted the significance of good faith regulatory compliance in competition assessment,^[78] it recognised the relevance of regulatory context insofar as it ‘contributes to the determination of the competitive conditions under which an undertaking … carries on its business in the relevant markets’.^[79] The Court elaborated three areas where regulatory context might be relevant: ‘defining the relevant markets, assessing the abusive nature of such conduct or setting the amount of the fines’.^[80] In reality, the invocation of regulatory context in Deutsche Telekom was somewhat disingenuous, enabling the Court to endorse a cost–price calculation mechanism that was ‘at odds with the realities of the market’^[81] for the defendant and its rivals.^[82] Despite this, in both Slovak Telekom and Baltic Rail, the Court drew upon this statement from Deutsche Telekom to conclude that the regulatory context justified application of a less exacting legal standard under Article 102, reiterating that ‘a regulatory obligation can be relevant for the assessment of abusive conduct’.^[83] This language is echoed in later cases.^[84] It is the 2023 judgment of the Grand Chamber in Meta Platforms, however, that provides the most explicit support for a potentially pivotal role for regulatory norms within competition law assessment.

The Meta Platforms case began with an infringement decision of the German competition authority against the owner of the Facebook social network, which aimed at the latter’s user-data-gathering policies. In essence, the decision construed Meta’s policies for use of Facebook services as an abuse of market power,^[85] using as a proxy for unfairness in competition law terms the fact that these policies were non-compliant with relevant data protection standards as reflected in the EU’s General Data Protection Regulation (GDPR).^[86] The Court of Justice, called upon to advise the fairly unconvinced national court, began by reiterating the orthodoxy of the distinctiveness of competition law from other regulation, including the different legal interests protected and enforcement by different regulators.^[87] Yet it quickly turned to acknowledge the applicability of regulatory standards and the role of compliance within the nominally distinct task of competition assessment.

The Court went much further than in previous cases, where the regulatory context had been described as potentially ‘relevant’ to this determination; in Meta Platforms, the Court held that ‘compliance or non-compliance … may, depending on the circumstances, be a vital clue’ as to the permissibility of behaviour under competition law.^[88] Notably, although the questions referred pertained only to data protection, the Court’s response was pitched in broader terms.^[89] Though the Court did not elaborate at length on how regulatory compliance might offer a ‘vital clue’ to competition liability, it noted the ‘great importance’ of having access to user data for many digital economy business models,^[90] so that it might undermine the effectiveness of competition enforcement to ‘disregard the reality of this economic development’.^[91] Echoing bpost,^[92] the Court then considered how national authorities should collaborate to minimise ‘the risk of divergences’,^[93] an approach that it grounded in the general EU law duty of sincere cooperation in Article 4(3) TEU. The judgment thus offers a more nuanced account of the inter-relationship between competition law and other regulation, by acknowledging that the latter may influence or even determine the application of the former in practice, depending upon its effects within a relevant market.

B. Relevance of regulatory norms in practice

The impact of the regulatory context on competition law assessment can be grouped into three broad categories: where regulation effectively excludes competition law liability; where it changes the structural competition dynamics within a market; and where it influences the behaviour of economic actors in a manner that prevents – or conversely provokes – competition law infringements.

(i) Regulation as exemption of competition law liability

Most radically, though also most exceptionally, regulation might remove the scope for applying the competition rules to market behaviour or at least preclude a finding of liability. The competition rules apply to ‘undertakings’, a concept that has been interpreted to exclude certain (typically highly regulated) market actors that pursue objectives related to social solidarity or that exercise quasi-public powers.^[94] The State action defence, noted above, provides an effective means to avoid Articles 101 and 102, but only where the applicable regulatory regime removes all capacity for ‘autonomous conduct’, competitive or otherwise, by regulated undertakings.^[95] Similarly, under Article 106(2) undertakings that are ‘entrusted’ by Member States with the operation of services of general economic interest can be exempted from application of competition law, but only to the extent that being subject to such obligations would ‘obstruct the performance, in law or in fact’ of this task. These three disparate exceptions share a common theme: although the would-be defendant is engaged in what is, strictly speaking, ‘economic’ activity on the market, the nature or impact of the underlying regulatory framework removes its actions from the purview of the competition rules as a matter of law, regardless of their effects on competition in fact.

Alternatively, the operation of the regulatory framework may provide a good defence that enables a defendant to rebut the claim that its conduct is harmful from a competition perspective. The Article 101 jurisprudence recognises the ‘Wouters exception’, which enables certain prima facie restrictive arrangements to escape prohibition ‘by effect’ on the basis that the coordination pursues ‘legitimate objectives in the public interest which are not per se anticompetitive’, and subject to indispensability and proportionality requirements.^[96] Formally, the exception does not hinge upon the impact of regulation on the activities under scrutiny; that is, the concept of ‘legitimate objectives in the public interest’ does not require that defendants are empowered to achieve those objectives under public regulation. But it is notable that, on the facts of Wouters, the organisation concerned (the Dutch bar association) was expressly tasked under domestic legislation to adopt the rules under scrutiny, ‘in the interests of the proper practice of the [legal] profession’.^[97] In this sense, a parallel can be drawn to the exceptions to the ‘undertaking’ concept that have similarly developed in the jurisprudence.^[98] In each instance, although the activity concerned is ‘economic’ in nature the predominant interests at stake are not, and typically (indeed, perhaps necessarily in practice) this will be reflected in the underlying regulatory framework.^[99]

Under Article 102, it is difficult to successfully argue that regulatory obligations deprive a dominant undertaking’s conduct of its ‘autonomous’ nature so as to preclude antitrust scrutiny.^[100] But there may be greater scope, where behaviour is driven by such considerations, for advancing a claim of ‘objective necessity’, which can serve as an objective justification, or defence, to a prima facie finding of abuse.^[101] This potential is seen most clearly in the case-law on refusal to deal, where the jurisprudence explicitly recognises that the ‘applicable regulatory framework’ may impose constraints on dominant firm behaviour in a manner that essentially legitimates what would otherwise be an abusive refusal to grant access to the dominant firm’s property.^[102] But objective necessity claims are narrowly construed in practice, and defendants have the burden of establishing that their behaviour is both necessary to achieve the public interest aim and that the actual or potential exclusionary effects resulting from the conduct are proportionate to the allegedly necessary aim.^[103] The objective necessity defence moreover does not apply where a dominant defendant claims, ‘rightly or wrongly’, that its behaviour was necessary to prevent other market actors from breaching their regulatory obligations,^[104] the idea being that private actors should leave the task of enforcement to the professionals, ie the applicable public regulator.

(ii) Regulation as a structural feature of market competition

Most of the time the underlying regulatory framework does not provide anything so dramatic as a wholesale dispensation from competition scrutiny. It is more usual for regulation to feed into antitrust assessment either by heightening or dampening competition dynamics (ie the possibilities for effective competition within a market) or by altering (whether by constraining or directing) the behaviour of economic actors in ways that make anticompetitive conduct more or less likely.

Regulation, first, can function structurally as a barrier to entry or expansion by competitors, typically enhancing the power of incumbent market actors and making anticompetitive effects more likely or plausible. The case-law recognises a wide variety of such circumstances, including statutory monopolies and intellectual property rights.^[105] Consideration of barriers to entry is a key component of the initial assessment of dominance under Article 102, and can also be highly informative of whether a merger is likely to create or strengthen a dominant position, so as to violate the ‘significant impediment to effective competition’ standard under the EUMR. Such structural effects may endure even after regulation: so, for instance, entities that previously benefited from special or exclusive rights under regulation may continue to experience de facto benefits after such legislation is removed,^[106] firms may maintain anticompetitive arrangements after the regulatory justification for doing so has disappeared,^[107] or firms may even seek subsequently to reimplement regulatory barriers through private arrangements after these have been dismantled by market liberalisation.^[108]

Regulatory barriers to entry may also be relevant to the assessment of likely anticompetitive effects that follow from behaviour.^[109] It was noted above that contemporary competition assessment is highly attuned to the specific context in which conduct occurs; increasingly, the case-law also emphasises that competition law aims at behaviour that has at least potential anticompetitive effects.^[110] The regulatory framework may therefore feed into the evaluation of whether coordination violates Article 101(1) or that single firm conduct constitutes an abuse of existing market power, insofar as the regulatory context may influence ‘the consequences of a certain practice in the market or for consumers’.^[111]

A more complex question is whether the anticompetitive effects must be caused by the impugned behaviour. In Google Shopping, the Court of Justice described the establishment of a causal link between a firm’s behaviour and the claimed actual or potential anticompetitive effects as one of the ‘essential constituent elements of an infringement of competition law’.^[112] This would appear to preclude a finding of breach of Articles 101 or 102 where the harmful effects on competition are attributable more or less entirely to the impact of the relevant regulation, even if defendant undertaking(s) have engaged in what looks like wholly unmeritorious behaviour from a competition perspective. Accordingly, while there is a high threshold to prove that regulation robs firm behaviour of its autonomous character, there may be greater scope to argue that the pervasive distortive effects of regulation mean the necessary causal link cannot be found. Yet Google Shopping does not require an exacting causation standard, with the Court accepting that it is sufficient merely to demonstrate a ‘correlation’ between suspect practices and the evolution of the market concerned, provided that a causal relationship can be confirmed by ‘additional information’ such as the position of market participants.^[113] The question in regulated markets is therefore whether it is possible to identify a sufficient diminution in effective competition arising from the suspect behaviour, over and above any limitations or distortions arising from the regulatory restrictions.

Conversely, regulation may facilitate new entry which enhances competition overall. This is, for instance, a principal objective of much of the market liberalisation legislation introduced in the EU since the 1990s.^[114] There are a number of common regulatory techniques which are deployed frequently to increase opportunities for competition: such as requiring the removal of existing monopoly rights that protect (typically dominant) incumbent operators from new entry; imposing structural separation on vertically-integrated incumbents to reduce incentives to engage in exclusionary behaviour; and requiring incumbents to share existing infrastructure or even data with potential new rivals, to enable or assist entry. Regulation aimed at market-opening may be supplemented by regulation that controls the behaviour of former monopolists in liberalised sectors, a point developed further below. In such circumstances, antitrust takes account of the altered market dynamics effectuated by the presence of the regulation, so that a defendant cannot, for instance, argue that its behaviour should be assessed as if unencumbered by its regulatory burdens.^[115] On the other hand, the regulatory framework may make it more plausible for a defendant to claim that its superficially restrictive behaviour generates countervailing efficiencies.^[116]

(iii) Regulation and its impact on market behaviour

Regulation may also influence a firm’s market behaviour in ways that have implications for competition assessment: either because it directly controls market behaviour or because it incentivises conduct that generates (anti)competitive effects. Here, several scenarios can be envisaged.

Most constructively, regulation may guide a firm to competition law-compliant behaviour. Where appropriately calibrated and effectively implemented, a regulatory framework may prevent regulated firms from engaging in the sorts of behaviour that would otherwise cause competition problems: by limiting a firm’s capacity to charge prices that are excessively high or have exclusionary effect, for instance, or by preventing use of ‘unfair trading conditions’ in contractual relations. This outcome is most likely to manifest itself obliquely, in an absence of antitrust problems in a sector.^[117] To the extent that context-specific competition assessment takes place prospectively, the general starting point is that the market behaviour of economic actors is shaped and constrained by other regulatory norms that apply within that sector and to that conduct.^[118] This is most salient in merger control, where it is assumed that firms comply with their regulatory obligations, and that their scope for ‘free’ market conduct is curtailed accordingly.

More contentiously, regulation may guide a firm to behaviour that is essentially at odds with competition law, a scenario most likely to arise where the regulatory regime endorses ‘inefficient’ market outcomes in pursuit of other policy goals. In this case we may encounter the possibility flagged in Section 3, namely prosecution under the competition rules of behaviour that is not only fully compliant with, but has also been largely influenced by, a regulatory framework that deliberately prioritises other socially valuable goals.^[119] This is an essentially unsatisfactory outcome, which is considered further below.

The first two scenarios each envisage effective compliance with regulatory obligations, whether this leads to competitive market outcomes or not. But firms do not always respect their regulatory obligations or make good faith efforts to do so. Within the competition jurisprudence, it is possible to identify several scenarios whereby an antitrust defendant’s liability is essentially contingent upon its interaction with another regulatory framework. It is these cases, in which regulatory norms function as the ‘hook’ for additional antitrust liability, which raise the most interesting but also the most difficult questions regarding the impact of regulatory norms on competition law.

Most straightforwardly, a regulated firm may simply disregard the constraints that are placed on its behaviour. In Meta Platforms, a digital firm’s data-gathering policy was deemed to amount to an abuse of its market power because the policy failed to respect the relevant data protection standard. A similar approach has been adopted in the Commission’s enforcement activity in liberalised utilities sectors, where failure to abide by regulatory duties to share infrastructure is also construed as an abuse of dominance by the incumbent operator.^[120] Here, the regulatory violation – which may or may not have been prosecuted by the sector regulator in parallel – is usually obvious; the more interesting question is how and why the behaviour might additionally become the focus of an antitrust investigation.

Beyond the problem of mere non-compliance, the competition case-law discloses a variety of instances where liability hinges on a firm’s disruptive interaction with a regulatory framework. In essence, these cases involve so-called ‘regulatory gaming’, namely ‘[b]ehaviour that abuses a neutral or procompetitive regulatory structure and wields it as a tool to accomplish [anticompetitive] results’.^[121] What distinguishes the non-compliance cases from the regulatory gaming ones is that, generally speaking, in the latter the firm’s behaviour is unproblematic (or at least unprosecutable) under the relevant regulatory regime. Instead, antitrust liability derives from the defendant’s efforts to avoid the regulatory constraints while remaining within the strict letter of the law (what we term ‘bad faith compliance’), or to exploit opportunities created by the existing regulation to anticompetitive ends (‘abusive compliance’).

Many examples of bad faith compliance relate to the highly regulated pharmaceutical sector. In AstraZeneca,^[122] the first EU competition case that directly embraced regulatory gaming as a theory of harm, the defendant took advantage of loopholes in the rules on pharmaceutical marketing authorisation and weaknesses in the administration of national patent offices to exclude would-be generic competitors. In Teva,^[123] the defendant strategically lodged and withdrew divisional patent applications in a manner permitted by the regulatory framework but which was clearly intended to frustrate new entry. Most remarkably, in Aspen Pharma the defendant successfully implemented an excessive pricing campaign in a context where every Member State had domestic price regulation, by exploiting design flaws which allowed it to manipulate the regulatory price-setting processes.^[124] For our purposes, what is most notable about these cases is how they make the leap from attributing antitrust liability for non-compliance with the express rules of a regulatory framework to non-compliance with the broader spirit or underlying policy objectives of those rules.

In abusive compliance cases, the regulatory context is primarily of relevance insofar as it provides an explanation for strategic behaviour that otherwise lacks an anticompetitive rationale. This is exemplified by AB InBev,^[125] where the defendant changed its packaging for beer sold in France and the Netherlands to prevent cross-border resales in Belgium, where it charged higher wholesale prices. This change – to no longer include information in both Dutch and French on beer sold in countries other than Belgium – only made sense as an exclusionary strategy in view of Belgian legislation on food labelling, which meant that larger retailers only purchased products labelled in both languages. In Roche & Novartis,^[126] the defendants engaged in collusive lobbying efforts to persuade the Commission to change its advice on the off-label prescribing of a certain drug, which enhanced the profits of both cartelists. Similarly, in Car Emissions,^[127] car manufacturers agreed to design new vehicles to the minimum regulatory standard for nitrogen oxide cleaning, an arrangement that the Commission prosecuted as a secret cartel. What is notable is that the cartelists explicitly agreed to adhere to their existing regulatory obligations, albeit knowing that these were likely to become more demanding in future; in doing so, however, they effectively agreed not to compete on cleaning technology as a dimension of competition, which the Commission construed as a limitation both of consumer choice and future innovation. Pharmaceutical pay-to-delay agreements, whereby drug companies pay off potential new entrants in the guise of patent litigation settlements, can also be seen as abusive compliance.^[128]

The real world is complicated, of course, and some regulations may be equivocal in terms of their overall effects on competition, so that a neat dichotomy between regulation that limits competition versus that which enhances competition often cannot be drawn. The GDPR provides an example of such mixed effects. On the one hand, by imposing limits on what market actors can do with personal data, it constrains the market power of Big Tech companies, which have disproportionately plentiful access to such commercially valuable information.^[129] On the other, GDPR compliance is an expensive business, and so the very existence of these rules, which apply to all data controllers regardless of size, may constitute a barrier to entry.^[130] The GDPR says little about its intended effects on competition beyond a vague reference to levelling the playing field,^[131] yet it has a significant impact on many digital business models, as recognised in Meta Platforms.^[132]

C. Understanding the role of regulatory norms

Regulation accordingly feeds into competition assessment in numerous (occasionally contradictory) dimensions, from reducing the scope to identify autonomous market behaviour to enhancing the risk that firm conduct generates harmful effects. Where competition law is applied in regulated markets, the competition rules do not generally endorse or apply the regulatory norms as such. But by accepting a market context where these norms reflect the accepted ‘rules of the game’ for market participants, competition law essentially reinforces the more diverse public interest values that are reflected within other regulatory regimes. Occasionally, this means deferring entirely to alternative values by allowing the regulatory regime to displace the application of competition law or to provide at least a good defence. More usually, these norms are absorbed within the contextual analysis, and the case-law increasingly recognises the legitimacy of aligning the thrust of competition analysis with pre-existing regulatory obligations (provided, of course, that these are not fundamentally at odds). There is obvious potential for tension, where for instance competition law turns its focus on market power that has been created by regulation as, say, a reward for innovation or to ensure the universal delivery of public services. But even here, competition analysis typically finds a way to accommodate potentially conflicting values, by distinguishing between the lawful scope of intellectual property rights and efforts to extend those rights beyond what is envisaged by IP law,^[133] for instance, or by recognising the fundamentally different position of liberalised and non-liberalised markets.^[134] Even if the goal of competition law is merely to achieve and protect undistorted competition, the space given to other regulatory norms within the contextual approach allows for the accommodation of complementary (and sometimes conflicting) values. These cases thus illustrate the tendency towards a ‘sponge’^[135]-like or ‘polycentric’^[136] quality within competition law, to the extent that a diversity of values can be accommodated – if not necessarily pursued directly – within contemporary enforcement.

The above discussion nonetheless identified two broad areas where this narrative of an essentially passive-but-tolerant inter-relationship between competition law and other regulatory norms does not provide an effective explanation. The first scenario arises where a (typically national) regulatory framework favours inefficient market outcomes in pursuit of non-economic objectives, albeit leaving sufficient scope for independent conduct to attract antitrust attention. The unsatisfactory position in the current law has been noted: in line with the orthodox approach, unless the narrow State action defence applies, distortions generated by the regulatory framework can be reflected only by discounting the penalty that might be imposed.^[137] This position can be critiqued from two dimensions. It creates unfairness for, or at least imposes a heavy burden on, defendants, who seem to have a positive responsibility to seek to align their regulatory obligations with the perceived demands of competition law.^[138] But even leaving the question of fairness aside, when viewed through the lens of contextual analysis this approach gives little weight to the obviously significant impact of the regulatory framework on competition dynamics in the market concerned. Instead, there is a strong argument that the (typically, non-economic) considerations factored into the regulatory framework should inform, though not necessarily determine, the competition enforcer’s understanding of what ‘competition on the merits’ looks like in a sector. While this may be insufficient to provide a good defence where there is further evidence that the defendant’s conduct is objectively non-meritorious in competition law terms, it would at least provide some recognition of the public interests that motivate the regulatory framework. To the extent that the latter are pretextual or at least insufficient to justify the restrictions on competition entailed, the more appropriate competition policy response is to attack the regulation directly.^[139]

A second category of cases that require more in-depth consideration are those involving non-compliance and regulatory gaming, which pose a conceptual dilemma from a competition policy perspective. Through such behaviour, a firm may engage in conduct that is harmful to effective market competition, which seems to fall within the legitimate scope of the competition rules. Yet insofar as the firm’s ‘unmeritorious’ behaviour revolves principally around its interactions with another regulatory framework, one might ask what makes this an appropriate subject for antitrust intervention, as opposed to the more obviously and directly applicable regulatory regime. In cases of non-compliance, where the behaviour violates the regulatory norm, the question is how to justify competition enforcement as opposed to (or in addition to) direct regulatory enforcement. In cases of bad faith compliance, the question is why competition law might take it upon itself to redesign or fill in perceived gaps within a regulatory framework rather than deferring to the regulation itself. Cases of abusive compliance are less intimately entwined with the obligations imposed by a regulatory framework, making these easier to explain as examples of ‘regulation as context’, discussed above. Yet such cases almost unavoidably involve value judgements regarding the nature of the defendant’s interaction with the regulatory regime, and so raise a similar question to other categories: how can we explain the role played by regulatory norms within the competition law assessment in such cases? This is the focus of the next section of this article.

5. Understanding the role of regulatory norms in competition assessment: explaining the non-compliance and regulatory gaming case-law

The crux of concern is that these cases involve, in essence, the ‘outsourcing’ of liability standards within competition law to other regulatory fields. A competition lawyer schooled in the ‘two barriers’ approach to the competition law/regulation inter-relationship would not question the legitimate existence of regulatory norms ‘other than competition law policies’.^[140] Yet it is less obvious why competition law should effectively defer to standards of behaviour developed in other regulatory policy areas – a problem that becomes more acute in cases of regulatory gaming, which involve deference to vaguer understandings of what a regulatory regime ought to entail if working well. The objection is not that the policy goals reflected in the regulatory framework lack legitimacy or value. Instead, the question is why competition law internalises these goals by applying regulatory norms as the determinative standard of competition liability in such cases. Put another way, why should protection of the goals, values and policy choices that are reflected in a specific regulatory framework fall within the legitimate scope (and enforcement purview) of competition law in addition?

Before exploring this question, let us recap several core principles of the ‘orthodox’ jurisprudence. Formally, antitrust compliance is unrelated to regulatory compliance within EU law: a firm might breach the competition rules while respecting its regulatory obligations, and vice versa. This links to the absence of a fault requirement, meaning that good faith regulatory compliance does not immunise a firm from competition liability in respect of the same behaviour. Competition enforcers are limited to the pursuit of ‘competition issues’; a somewhat woolly limitation, but one which confirms that the competition rules are not a panacea for all public policy ills. Accordingly, to explain the non-compliance and regulatory gaming cases, we must consider why as a matter of competition law such behaviours come within the legitimate scope of the competition rules. To do so, we explore the case-law from three distinct perspectives – through the lenses of effectiveness, competition on the merits, and regulatory non-compliance and gaming as strategic action – asking whether each provides a plausible explanation for addressing the behaviour under competition law.

A. Effectiveness of competition law as a regulatory enforcement mechanism

It is important to recognise why both non-compliance and regulatory gaming cases typically arise in practice: namely, the effectiveness of competition law as a regulatory enforcement mechanism, unconstrained by substantive or procedural limitation that may hinder enforcement of the regulatory framework directly. This is apparent from the background to many cases considered above. In Telecom Polska, the sector regulator had repeatedly enforced national telecommunications law against the defendant, but the penalties it imposed were insignificant and failed to deter non-compliance. In Slovak Telekom, the defendant had evaded its regulatory obligation to grant access to infrastructure for five years but received no pushback from the (captured or under-resourced) regulator. In Meta Platforms, the problem was a mismatch of jurisdiction: while the data privacy concerns were felt most acutely by German consumers and policymakers, power to enforce the GDPR lay with the Irish data protection authority, which dragged its feet. The deficiency of national patent systems is a key undercurrent in many competition cases in pharmaceutical markets.^[141]

Viewed pragmatically, it is easy to understand why competition law is used to reinforce regulatory norms: in effect, competition enforcement might get the job done more effectively than the regulation itself. Competition law brings much to the table as a mechanism to enforce regulatory rules. In particular, the broad sweep of its prohibitions means that competition law is sufficiently wide and flexible to catch such behaviour, including problematic conduct that might lie outside the formal parameters of the regulation itself. EU law also imposes demanding minimum standards to protect the independence, resourcing and sanctioning powers of competition authorities,^[142] which sectoral regulators often lack. The resistance within EU law to accept antitrust preclusion even in the presence of significant regulation supports the proposition that competition law can legitimately seek to resolve market problems that arise from malfunctioning regulation, providing a ‘complementary legal response[]’ in order to form ‘a coherent whole’ of market supervision.^[143]

Yet the bare fact that competition law can enforce regulatory norms more successfully provides an inadequate explanation for these cases. On the one hand, using competition law as a sticking-plaster on ineffective regulation gives little weight to the substance, and substantive limits, of the underlying regulatory framework. Cases like bpost and Meta Platforms demonstrate that, if competition law is applied to ‘regulated’ problems, this requires a degree of intentionality and coherence and must be done in a manner that recognises the ‘respective powers and competences’^[144] of each regime. Deploying competition law on the pragmatic basis that it can better achieve the goals of a regulatory framework fails to acknowledge the central role of the underpinning regulation in such cases, both in terms of how ‘the social problem involved’^[145] and the appropriate enforcement response should be framed.

On the other hand, justifying these cases on an effectiveness basis offers little insight into how such regulatory failures can also become, in the language of Orlen, ‘competition issues’.^[146] Regulation, as discussed, often incorporates a diversity of policy goals that extend beyond the protection of undistorted competition. Regulatory standards of acceptable behaviour typically seek to advance values outside the scope of competition law as such. To take an notable example, the GDPR states that ‘[t]he processing of personal data should be designed to serve mankind’, and explains that the rules seek to balance a variety of fundamental rights including privacy and family life, freedom of thought, freedom of expression, freedom to conduct a business, fair trial rights, and cultural, religious and linguistic diversity.^[147] While it is difficult to argue against the importance of these varied considerations, it is clear that a data protection standard – such as that applied to impose antitrust liability in Meta Platforms – reflects a wider ranges of values than the mere protection of undistorted competition. If we want to bring the former within the legitimate scope of the latter, we need a more nuanced explanation as to when and why diverse regulatory concerns are also ‘competition issues’.

B. Regulatory compliance as a reflection of ‘competition on the merits’

Instead, we return to the concept of ‘competition on the merits’: the idea that competition law does not seek to constrain firms from competing, even aggressively, but merely limits market activity to the acceptable parameters of ‘normal competition’. The concept is prominent in recent jurisprudence and in the Commission’s Draft Guidelines on Article 102.^[148] The latter follow the approach in SEN,^[149] by decoupling the notion of ‘conduct departing from competition on the merits’ from ‘capability to produce exclusionary effects’, while indicating that both elements must be satisfied to establish exclusionary abuse of dominance.^[150] The Draft Guidelines nonetheless suggests that capability can be presumed for a variety of practices,^[151] making competition on the merits the key determinant of liability in many circumstances.

The proposition that competition law protects only ‘competition on the merits’ might justify the direct application of regulatory norms to the extent that the standard of ‘normal’ competition implies that market actors, inter alia, follow their regulatory obligations. Regulatory compliance is often costly for market actors, whether because the regulation curtails a firm’s scope to engage in profitable activity^[152] or requires it to incur costs it would prefer to avoid,^[153] or simply because of direct compliance costs.^[154] While escaping or minimising regulatory obligations can give individual undertakings a competitive advantage, doing so undermines the public interest values that the regulation advances and protects. Viewed in this manner, non-compliance or gaming of regulatory norms is inherently at odds with what we understand meritorious, and thus permissible, market competition to look like. Such an approach aligns with the ‘level playing field’ notion that permeates EU law, which is premised on the idea that market actors should be subject to common rules and restrictions to ensure fair competition.^[155] Since the regulatory framework structures the ‘fair’ competitive game in the level playing field of the marketplace, it is reasonable to require adherence under competition law even if some regulatory rules pursue values that are non-economic in nature.

At its simplest, the existing regulatory framework might be taken to reflect the lawful scope of a firm’s freedom in the marketplace. By incorporating regulatory standards into antitrust assessment, we do not ‘enforce’ these norms as such. Yet because a defendant can have no legitimate expectation to act contrary to its regulatory obligations, it cannot object where this behaviour is curtailed or punished through competition enforcement. Whereas in Slovak Telekom and Baltic Rail the Court denied that a regulatory violation might comprise an abuse in itself, it accepted that the presence of regulation may permit a less demanding legal standard for antitrust intervention.^[156] It justified this essentially on the basis that pre-existing regulation already circumscribed what the defendants could reasonably expect to be allowed to do in the marketplace. Competition law need not maintain the fiction that, say, a firm can validly refuse to grant access to its infrastructure unless an antitrust duty to deal arises exceptionally under Article 102, if sector-specific regulation has already imposed an unambiguous mandatory sharing obligation. A similar logic is discernible in the Commission’s approach to sustainability agreements in its Horizontal Cooperation Guidelines.^[157] Where ‘appropriate regulation’ exists to adequately address market failures, the Guidelines are sceptical of the acceptability of private coordination in tandem,^[158] ‘given that the legislator has already decided that each undertaking must individually comply with the obligation in question’.^[159] Yet private action may be more necessary to address ‘residual market failures that are not or not fully addressed by public policies and regulation’.^[160] In the latter case, undertakings have plausible claims for greater freedom of (private) action, since regulation has not yet delimited the legitimate scope of market activity. This reflects, in the language of Meta Platforms, ‘the reality of economic development’ in regulated markets.^[161]

Yet treating a regulatory framework as merely indicative of the limits to a firm’s economic freedom provides an incomplete explanation of the non-compliance and gaming case-law. First, it is less convincing in respect of behaviour that violates the spirit of a regulatory norm rather than its express provisions. It is one thing to assert that a defendant can have no legitimate expectation, for example, to refuse to comply with a mandatory sharing obligation imposed by sector-specific regulation; it is quite another to make this claim regarding behaviour that is, say, objectively ‘misleading’ but not strictly speaking unlawful.^[162] Arguably, the mere existence of collusion or dominance limits what firms can legitimately expect to do in markets, even absent abusive behaviour;^[163] and applying this logic, we might extend the proscriptive potential of regulation to a wider subset of behaviours at variance with its policy objectives more broadly.^[164] Yet this creates a legal certainty problem: how can firms know in advance what sorts of behaviours are likely to conflict with the broader ‘regulatory orbit’, if this is not clearly delimited by regulation? Second, this approach provides only a partial account of what happens in non-compliance and gaming cases: it tells us why defendants cannot object to competition enforcement covering the same ground, but not why such behaviour can legitimately be construed as a ‘competition issue’ as well as a regulatory breach.

A more muscular version of this argument is what we call the ‘dominance-plus’ approach. From this perspective, because non-compliance or regulatory gaming is inherently non-meritorious, any conduct of this sort is fair game under competition law provided that a threshold criterion for antitrust jurisdiction (most naturally a dominant position under Article 102) is also satisfied. While competition law cannot compel firms directly to comply with their regulatory obligations, it can punish non-compliance or gaming insofar as such conduct falls within the discrete prohibitions in the competition rules. It is the existence of significant market power (or the fact of coordination, if we extend the logic to a collusion-plus approach under Article 101^[165]) that brings the matter within the scope of competition law; the regulatory misconduct then provides the plus factor that crystallises the competition infringement. This was, effectively, the position of the Bundeskartellamt in Meta Platforms.^[166]

This approach has a degree of intuitive appeal. It is hard to argue that a firm acts meritoriously if it disregards or takes advantage of its regulatory obligations, though it is more debatable as to whether this is ‘normal’ market behaviour. Although the regulatory regime may reflect values beyond the task of protecting undistorted competition, the enforcement of such values in non-compliance and gaming cases can be explained as an example of competition law’s ‘sponge’-like quality, a term developed by Ezrachi to describe the capacity to absorb and reflect policy influences from other sources such as market regulation.^[167] Undeniably, this is an understanding of competition law’s role that reflects a more overtly ‘political’ vision of antitrust as a tool directly to regulate economic power, a perspective which typically gives less weight to more ‘legal’ questions of the permitted parameters of a dominant firm’s market freedom.^[168] Yet it is not obviously precluded by the Article 102 jurisprudence, which recognises the ‘special responsibility’ of dominant firms to protect effective competition,^[169] and which does not require misuse of the market power that underlies a dominant position in order to establish ‘abuse’.^[170]

Yet even accepting that dominant firms have a special responsibility to avoid distorting competition, objections can be raised against construing regulatory non-compliance or gaming of norms as ‘abuses’ in an antitrust sense. The case-law establishes that, even if ‘a regulatory obligation can be relevant for the assessment of abusive conduct’, the presence of regulation ‘cannot relieve the Commission of the requirement of establishing that there is abuse’ on the facts.^[171] While Meta Platforms saw the most ambitious use of regulatory norms to determine liability, the Court still stopped short of treating regulatory non-compliance as conclusive proof of antitrust abuse, and it emphasised the differing ‘objectives and tasks’ pursued by competition law compared with other regulatory fields.^[172] In bpost, where the Court acknowledged that competition law might ‘legitimately’ tackle behaviour that is also subject to regulatory controls, it clearly envisaged that parallel proceedings would address diverse regulatory objectives.^[173] It is doubtful whether repackaging a regulatory violation as an antitrust one merely due to the presence of dominance (or indeed coordination) respects this limitation. The dominance-plus approach also sits uncomfortably with the principle that good regulatory compliance is generally irrelevant to the objective question of whether conduct attracts antitrust liability, provided that the regulatory regime allows sufficient scope for autonomous behaviour that can come within the reach of the competition rules.^[174]

This concern – that defendants might somehow be damned if they do and damned if they don’t respect regulatory norms – highlights an additional objection, namely the risk of disproportionate punishments for what are, at their core, regulatory violations. bpost does not prohibit the parallel enforcement of regulatory norms, although it includes proportionality considerations in the determination of whether an exception to ne bis in idem is permissible in individual cases. But even where proceedings are limited to competition enforcement, the consequences for defendants are often much more disadvantageous than if prosecution was pursued under the regulation directly.^[175] Even if one has little sympathy for firms that disregard their regulatory obligations, reinterpreting a regulatory violation as a competition abuse can raise the stakes considerably, including the possibility of private damages claims alongside public enforcement.^[176] There is also the question of how to impose fair and sensible limits on the scope of liability from a dominance-plus perspective: might any regulatory breach – whether of tax law, zoning rules, labour protections etc – become an antitrust abuse merely due to the existence of dominance (or collusion)? Concerns about disproportionate liability are even more acute in regulatory gaming cases, where the defendant has not in fact violated the relevant regulatory norm, so that antitrust enforcement not only reinforces but extends regulatory liability.

Yet within the emerging Article 102 jurisprudence the existence of conduct departing competition on the merits is not (typically) the sole determinant of whether an abuse arises; additionally, at least in principle, the conduct must have the capability to produce exclusionary effects.^[177] An effective limiting principle for non-compliance and regulatory gaming cases could conceivably be derived by adding a robust requirement to demonstrate anticompetitive effects arising from the regulatory breach/gaming, alongside a ‘but for’ causation standard directly linking the behaviour to its harmful effects. Demonstrating actual (or at least potential)^[178] and appreciable anticompetitive effects would squarely locate the problem within the sphere of ‘competition issues’ that fall within the legitimate scope of competition law. Requiring a ‘but for’ causation standard, additionally, would enable us to explain why this regulatory abuse should also generate liability under competition law. Such an approach has much to recommend it in principle.

It is at this juncture, however, that we run up against the constraints of the existing case-law. Although the recent jurisprudence makes greater space for effects-based considerations, in practice it still falls short of both proposed elements in most instances. While the case-law is not a model of clarity,^[179] it is only with respect to the ‘by effect’ limb of Article 101(1) that we find anything close to such a robust effects requirement.^[180] Under the ‘by object’ limb,^[181] it is essentially unnecessary to examine effects where an arrangement discloses intrinsic harm to competition ‘by its very nature’.^[182] Under Article 102, which is the basis for most of the cases considered here, behaviour needs only to be capable of restricting competition,^[183] a standard that can be met despite evidence that the conduct ‘has not produced actual anti-competitive effects’.^[184] To the extent that the case-law imposes a causation requirement, it foresees merely a need to demonstrate a ‘correlation’ between the defendant’s conduct and the claimed (likelihood or capacity to cause) harm.^[185] There are sound competition policy reasons to defend a more prophylactic approach, in order to prohibit overtly restrictive arrangements,^[186] or where the presence of a dominant firm implies a structural absence of competition.^[187] Yet the continuing distance between the rhetoric and reality of the effects-based approach means that a simple dominance-plus perspective may provide an over-inclusive account of when regulatory non-compliance and gaming of norms should also constitute competition issues.

C. (Non) compliance as a positive strategic action

Another way of thinking about these cases is to view non-compliance or regulatory gaming as a positive strategic action for competition law purposes. Here, the focus is the deliberate choice of defendants to neglect or ‘game’ their regulatory obligations as part of an anticompetitive course of conduct: the antitrust problem stems not merely from the fact the regulatory norm has been disregarded or disrupted, but also from the motivation for this behaviour. In effect, the regulatory framework is ‘weaponised’ to anticompetitive ends, and it is this dimension – rather than the regulatory element as such—that is pursued under competition law. The key aspect is thus the strategic behaviour of the defendant undertaking in terms of how it interacts with, takes advantage of, or disregards its regulatory obligations.

To provide an effective explanation of the non-compliance and regulatory gaming case, the question is whether (and in what circumstances) ‘strategic’ behaviour of this sort can breach the competition rules. Recent case-law provides support in several dimensions for an approach to liability that hinges on evidence of the anticompetitive intention of defendants. The transformative Intel judgment emphasised the relevance of evidence of an anticompetitive ‘strategy’ to find abuse of dominance.^[188] The Court in Google Shopping endorsed a focus on the ‘positive acts’ undertaken by the defendant,^[189] which served to distinguish its (abusive) actions from other (permissible) forms of self-preferencing. It also held that, when engaging in context-specific assessment, it is ‘clearly legitimate for the Commission to assess subjective matters, namely the motives underlying the business strategy in question’.^[190] The more contextual approach to ‘by object’ determination under Article 101 similarly extends consideration of the inherent ‘nature’ of coordination to the ‘aims which that conduct seeks to achieve from a competition standpoint’.^[191]

Yet intention-based requirements are controversial in competition law. Two broad objections can be advanced against conceptualising the non-compliance and gaming cases as instances of ‘strategic’ behaviour: one substantive and one practical. In substantive terms, it has been argued that in the presence of market power or coordination, certain forms of conduct should be objectively unacceptable due to their inherent capacity to distort competition, regardless of what a defendant seeks to achieve by its actions. This perspective is reflected in the jurisprudence, which rejects a determinative role for intention evidence within the framework of EU competition analysis. The perceived ‘objective aims’ of coordination are distinguished from the ‘subjective intention’ of coordinating parties, with the latter ‘not decisive for the purposes of the application of Article 101(1)’.^[192] Under Article 102, while ‘proof of an anti-competitive intention’ may constitute ‘an indication of the nature and objectives pursued by the strategy of that undertaking and, on that basis, may be taken into account’, the recent case-law reiterates that ‘demonstration of the existence of such an intention is neither necessary nor sufficient, in itself, to establish the existence of an abuse’.^[193] In Google Shopping, the Court described ‘the existence of any anticompetitive intent’ as ‘only one of a number of facts which may be taken into account’,^[194] and emphasised that there is ‘no obligation to establish the existence of such intent … in order to render Article 102 applicable’.^[195]

The substantive objection is not trivial, and imposing an across-the-board obligation under Articles 101 or 102 to demonstrate subjective anticompetitive intention in addition to establishing behaviour that is objectively harmful to competition would clearly be inappropriate. Yet as demonstrated by the case-law on predatory pricing, intention can play a valuable, and determinative, role in establishing antitrust liability in certain circumstances. Under the Akzo test,^[196] anticompetitive intention serves to confirm the unacceptability in antitrust terms of behaviour that is objectively ambiguous from a competition policy perspective.^[197] Intention evidence can usefully perform a similar function in regulated conduct cases, marking the dividing line between behaviour that is purely a problem within the context of a regulatory regime and behaviour with a dual character that also merits antitrust intervention (ie that reveals a ‘competition issue’). Here, again, we come up against a similar difficulty to that encountered when discussing the use of an effects-based limiting principle under the dominance-plus approach, namely the limits of the existing case-law. Yet it could be argued that recognising another intention-based exception where this possibility is already established in the jurisprudence is less radical than requiring a proper effects-and-causation analysis. In any event, these elements are not mutually exclusive.

The more procedural objection to making liability hinge on evidence of an anticompetitive strategy relates to the difficulties of demonstrating a ‘guilty mind’ where defendants are typically large corporate entities. On the one hand, there is the problem of determining what counts as the intention of the undertaking as a whole; on the other, there is the inherent tension, outlined above, of construing evidence of a desire to enhance an undertaking’s own market share as evidence of an objectively anticompetitive intention.^[198]

Yet, practically speaking, this has not prevented the accumulation of significant evidence of a defendant’s deliberate efforts to evade or defraud its regulatory obligations in many of the non-compliance and gaming cases discussed earlier.^[199] Indeed, the frustrating aspect of the Court’s insistence on the formal irrelevance of intention is that it is hard to reconcile with what occurred in these cases. For instance, whereas the General Court in AstraZeneca maintained that ‘proof of the deliberate nature of the conduct and of the bad faith of the undertaking in a dominant position is not required’,^[200] the higher Court of Justice stressed that such a hypothetical scenario was ‘radically different from … the present case’.^[201] The Teva decision, similarly, is imbued with the language of (bad) intention: the Commission found that the defendant ‘aimed at’, ‘consciously’, ‘intentionally’ and ‘clearly pursued’ its anticompetitive divisional patent applications.^[202] The decision formally maintained that there is ‘no requirement’ to establish anticompetitive intention,^[203] however, and cited Google Shopping to the effect that ‘the existence of an intention to compete on the merits, even if it were established, could not prove the absence of abuse’.^[204] Yet it is essentially impossible to envisage a plausible ‘good faith’ version of regulatory non-compliance or gaming as a competition law theory of harm. We have thus another example of the problem that so concerned Advocate General Wahl in his pivotal Opinion in Intel, with the Court saying one thing but doing quite another in practice.^[205] The policy question is whether it makes sense for EU competition law, in essence, to keep its options open by maintaining the fantasy that motivation is not a key element of the theory of harm in such cases; a question that pits legal principle against a desire to preserve the practical effectiveness of competition law to the greatest extent possible.

A focus on strategic anticompetitive intention in cases where the theory of harm is tied up with the question of regulatory compliance has a further aspect to commend itself: it could arguably provide a more satisfactory explanation for the problematic cases where a defendant is held to breach competition law despite its compliance with regulatory obligations that strongly influence its supposedly anticompetitive conduct. We identified two objections to this jurisprudence: the possible unfairness to defendants faced with the responsibility to correct policy choices made by a regulator, and the fact that it gives insufficient weight to how regulation affects ‘competition on the merits’ in the sector concerned. Yet these objections are less compelling if would-be ‘good faith’ compliance is not all that it seems. Where there is evidence that a defendant is acting in a strategically anticompetitive manner in its interaction with the regulatory framework, it is manifestly less unfair to pursue such behaviour under competition law. Evidence of anticompetitive intention can also provide an effective rejoinder to the claim that the regulatory regime demarcates the boundaries of ‘meritorious’ or permissible competition within a market. Of course, such an approach again departs from the orthodox ‘objectivity’ of competition assessment and may be criticised on this basis. Yet as a compromise between the two poles of deferring to and simply ignoring the coercive impact of regulation on private market behaviour, focusing on what the regulated defendant itself sought to achieve through its interaction with the regulatory framework arguably provides a reasonable limiting factor.

6. Conclusion

This article began with a simple (perhaps simplistic) dichotomy between competition law, which promotes undistorted competition, and other market regulation, which has a greater or lesser capacity to distort competition in service of alternative policy goals. Yet Ezrachi has criticised the ‘pretence of purity’ that competition law adopts,^[206] and even if antitrust liability professes to be formally unrelated to compliance with other regulatory norms, the presence of regulation can have a profound effect on the task of competition assessment in practice.

Regulation and the diverse public interest values that it represents are a constant feature of modern markets. Context-specific competition enforcement necessarily absorbs and reflects these values to some degree, to the extent that it works within the ‘distortions’ of competition that follow from regulation. In the absence of deliberate efforts by policymakers to make competition impossible, the competition rules rarely cede jurisdiction to other regulatory norms, and whereas competition law may tolerate the presence of regulation that generates market power for certain regulated actors, it tends to be unforgiving of autonomous behaviour that misuses such market power. Yet it is rare for competition assessment to disregard the regulatory context; where such cases do arise, this article has argued for a more nuanced understanding of what ‘normal’ competition looks like in such circumstances.

Yet regulation is not always just ‘part of the scenery’ in competition cases. Efforts to escape or frustrate regulation can constitute standalone antitrust infringements if carried out by an entity falling within the scope of competition law, whether due to dominance or acting in concert. This article has argued that the apparent outsourcing of antitrust liability standards within the non-compliance and regulatory gaming case-law provides an effective vehicle by which to interrogate and better appreciate the inter-relationship between competition law and other regulation. One way to understand these cases is to see the regulatory framework as setting the limits of fair competition in a marketplace, establishing the ‘rules of the game’ that all market actors can reasonably be expected to respect. While simple non-compliance or gaming of regulatory norms is arguably not a competition issue as such, it more plausibly becomes one where such conduct is the direct cause of appreciable anticompetitive effects. Another way to see these cases is as involving the ‘weaponisation’ of regulatory norms to anticompetitive ends as part of a deliberate strategy by regulated firms. From this perspective, the key antitrust consideration is what the defendant intends to achieve through its interaction with the regulatory framework. These perspectives are not mutually exclusive, moreover, and each finds some reflection in the existing case-law on non-compliance and regulatory gaming. Yet in many instances, the true explanation for punishing this conduct through competition enforcement is more pragmatic than principled: using competition law allows us to sidestep limitations or gaps within the regulatory framework itself. These cases thus involve an almost complete convergence between the goals of competition law and other market regulatory norms, and – for better or worse – take the legitimate scope of competition law to its outer limits.

1. * Law School, London School of Economics. Email: n.m.dunne@lse.ac.uk. Thank you to LSE colleagues who gave helpful comments on an earlier draft of this presentation at a staff seminar, in particular Tarun Khaitan and Emmanuel Voyiakis; to two anonymous referees who similarly provided very useful feedback; and to Current Legal Problems editor Deni Mantzari for her advice, support and expert shepherding of this piece through the review process.

   See, eg, Cases C-322/81 Michelin EU:C:1983:313, para 10, C‑42/21 P Baltic Rail EU:C:2023:12, para 86 and C‑252/21 Meta Platforms EU:C:2023:537, para 46. Also recital 11 of Regulation (EU) 2022/1925 on contestable and fair markets in the digital sector (Digital Markets Act) [2022] OJ L265/1. ↑

2. As explained in Section 2, the term ‘regulation’ can be defined as broadly as ‘all forms of economic or social influence’ or as narrowly as ‘a specific set of commands’ imposed by the State: see text accompanying nn 26 and 27. ↑

3. As the General Court recognised in Case T-168 GSK EU:T:2006:265, para 104. ↑

4. See, for example, the OECD’s Competition Assessment Toolkit (2019), which is designed to assist governments in identifying and reducing barriers to competition and other market distortions which may stem from State regulation. ↑

5. Article 7 TFEU states that ‘[t]he Union shall ensure consistency between its policies and activities, taking all of its objectives into account and in accordance with the principle of conferral of powers’. There is, however, limited practice applying the provision: N Nic Shuibhne, ‘Deconstructing and Reconstructing Article 7 TFEU’ in F Ippolito, ME Bartoloni and M Condinanzi (eds), The EU and the Proliferation of Integration Principles under the Lisbon Treaty (Routledge 2018). ↑

6. See eg P Ibáñez Colomo, The New EU Competition Law (Hart Publishing 2023) ch 3. ↑

7. See eg E Fox and E Healey, ‘When the State Harms Competition – The Role for Competition Law’ (2014) 79 Antitrust law Journal 769. ↑

8. See T Khaitan and S Steel, ‘Theorising Areas of Law: A Taxonomy of Special Jurisprudence’ (2022) 28 Legal Theory 325, 329–34. ↑

9. This debate has several dimensions: a normative component, seeking to determine the optimal goals (eg E Deutscher, ‘The Competition–Democracy Nexus Unpacked – Competition Law, Republican Liberty, and Democracy’ (2022) 41 YEL 197); an explicatory aspect, seeking to describe how the competition rules absorb and reflect different policy objectives (eg A Ezrachi, ‘Sponge’ (2017) 5 JAE 49); and an empirical aspect, seeking to identify the goals actually pursued in enforcement practice (see, eg, M Iacovides and K Stylianou, ‘The New Goals of EU Competition Law: Sustainability, Labour Rights, and Privacy’ (2024) 3 European Law Open 587). ↑

10. See n 1. ↑

11. O Odudu, ‘The Meaning of Undertaking within Article 81 EC’ (2005) 7 Cambridge Yearbook of European Legal Studies 211, 212. ↑

12. See Opinion of Advocate General Maduro in Case C-205/03 P FENIN EU:C:2005:666, para 26, on the distinction between ‘the power of the State’, which is subject to ‘democratic control’, and the actions of ‘economic operators’, which fall within the purview of competition law. ↑

13. Article 3(3) TEU. ↑

14. Council Regulation 139/2004 of 20 January 2004 on the control of concentrations between undertakings [2004] OJ L24/1. ↑

15. See eg discussion in Case C-307/18 Generics (UK) EU:C:2020:52, paras 87, 90 and 111 (referring to Article 101) and para 152 (referring to Article 102). ↑

16. Case C-85/86 Hoffmann La Roche EU:C:1979:36, para 91; also use of the term in C‑252/21, Meta Platforms EU:C:2023:537, para 47. ↑

17. European Commission, Guidelines on the application of Article 102 of the Treaty on the Functioning of the European Union to abusive exclusionary conduct by dominant undertakings (hereafter ‘Draft Article 102 Guidelines’), August 2024, paras 47–57. Discussing the role the concept plays within contemporary case-law, see P Ibáñez Colomo, ‘Competition on the Merits’ (2024) 61 CMLRev 387. ↑

18. So, for example, the need to ‘meet competition’ is generally not considered an acceptable defence to prima facie restrictive behaviour, see eg Case T-671/19 Qualcomm EU:T:2024:626, paras 598–99. ↑

19. The principal exception is the prohibition on hard-core cartel behaviour, where contextual analysis is ‘limited to what is strictly necessary’ to understand the anticompetitive nature of the conduct: Case C‑373/14 P Toshiba EU:C:2016:26, para 29. ↑

20. See eg Cases C-228/18 Budapest Bank EU:C:2020:265 and C-413/14 P Intel EU:C:2017:632. ↑

21. Provided for formally by the Article 101(3) TFEU exception, with the possibility of ‘objective justification’ read into the Article 102 TFEU jurisprudence, see eg Case C‑209/10 Post Danmark EU:C:2012:172. ↑

22. For instance, specific competition rules exist for agriculture pursuant to Articles 42 and 43 TFEU. ↑

23. See discussion of the distinction between ex ante and ex post enforcement in Case C-449/21 Towercast EU:C:2023:207. ↑

24. Case C-376/20 P CK Telecoms EU:C:2023:561, paras 84–86. ↑

25. While there is scope to agree to behavioural remedies under the EUMR, this approach is disfavoured, at least in principle: Commission notice on remedies acceptable under Council Regulation 139/2004 [2008] OJ C267/1. ↑

26. R Baldwin, M Cave and M Lodge, Understanding Regulation: Theory, Strategy and Practice (2nd edn, OUP 2011) 3. ↑

27. Regulation need not adopt a classic top-down ‘command and control’ form to come within this understanding, but pure self-regulation by undertakings is treated simply as economic activity within the scope of the competition rules in EU law, see eg the professional rules in Case C-309/00 Wouters EU:C:2002:98 and the industry reorganisation efforts in Case C-209/07 BIDS EU:C:2008:643. For discussion of ‘decentred’ approaches to regulation, see J Black, ‘Decentring Regulation: Understanding the Role of Regulation and Self-Regulation in a “Post-Regulatory” World’ (2001) 54 Current Legal Problems 103. ↑

28. Indeed, viewed from a regulatory theory perspective, competition law can be seen as a variety of regulation that falls within the broad category of ‘market-harnessing controls’ (Baldwin and others (n 26) 114–16). The competition jurisprudence, however, distinguishes the competition rules from ‘other’ regulation that may impact market dynamics, and that is the approach adopted within this piece. ↑

29. Digital Markets Act, Recital 11 (see also cases in n 1). ↑

30. For example, Directive 2019/944 on common rules for the internal market for electricity (OJ L158/125), recital (2), included as its aims: ‘to deliver real choice for all Union final customers, [] new business opportunities, competitive prices, efficient investment signals and higher standards of service, and to contribute to security of supply and sustainability.’ ↑

31. Case C-280/08 P Deutsche Telekom EU:C:2010:603. ↑

32. Opinion in Case Deutsche Telekom EU:C:2010:212, para 21. ↑

33. Case C-280/08 P Deutsche Telekom EU:C:2010:603, paras 80–84; reaffirmed in Case C-220/24 Regia Autonomă Aeroportul Internaţional ‘Avram Iancu’ Cluj v Consiliul Concurenţei EU:C:2025:124, para 27. ↑

34. Indeed, in Case C-382/22 P Cathay Pacific Airways v Commission EU:C:2026:129, para 125, the Court of Justice stated that the State action defence ‘has been only partially accepted’ in its jurisprudence. ↑

35. C-280/08 P Deutsche Telekom EU:C:2010:603, para 81. ↑

36. For a recent example where the State action defence actually succeeded on the facts, see Case T-136/19 Bulgarian Energy Holding EU:T:2023:669. ↑

37. Verizon Communications v Law Offices of Curtis V. Trinko, LLP, 540 US 398 (2003). The complexities of the position within US law are discussed in H Shelanski, ‘The Case for Rebalancing Antitrust and Regulation’ (2011) 109 Michigan Law Review 683. ↑

38. Case T-398/07 Spain v Commission EU:T:2012:173, para 55. ↑

39. Opinion in Case Deutsche Telekom EU:C:2010:212, para 25. ↑

40. In Case C-117/20 bpost EU:C:2022:202, the Court recognised an ‘objective of general interest’ in ensuring the effective application of competition law and other market regulation in parallel, ‘since they are pursuing [] distinct legitimate objectives’ (para 50, emphasis added). ↑

41. See, eg, Cases C-453/99 Courage and Crehan EU:C:2001:465, para 26 and C-74/14 Eturas EU:C:2016:42, paras 35–37. ↑

42. Case C-280/08 P Deutsche Telekom EU:C:2010:603. ↑

43. Case C-209/07 BIDS EU:C:2008:643. ↑

44. As was the case in support of the Siemens/Alstom merger, which was nonetheless prohibited by the Commission (M.8677, 6 February 2019). ↑

45. See Ursula von der Leyen, European Commission Political Guidelines 2024–29: ‘I believe we need a new approach to competition policy, … more supportive of companies scaling up in global markets’. See also Barbara Moens, ‘EU to Relax Merger Rules in Bid to Create ‘European Champions’ Financial Times (16 April 2026). ↑

46. Cases T‑321/05 AstraZeneca EU:T:2010:266, para 356 and T‑814/17 Baltic Rail EU:T:2020:545, para 189. ↑

47. Case C-209/07 BIDS EU:C:2008:643, para 21. ↑

48. Case C‑307/18 Generics (UK) and Others EU:C:2020:52. ↑

49. The Commission may take account of evidence of how the merged entity intends to act on the market going forward, but it places greater emphasis on whether that behaviour is anticipated to cause anticompetitive effects in fact: see eg Case M.11382—AGCO/TRIMBLE/JV (decision of 25.03.2024), for an example where the Commission concluded that distortive post-merger behaviour would not significantly impede effective competition in the circumstances. ↑

50. As the US Supreme Court said in Bell Atlantic v Twombly 550 U.S. 544 (2007), ‘resisting competition is routine market conduct’. ↑

51. This is a core tenet of competition policy, with roots that can be traced back to Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations (1776). For an argument that Smith’s influence should extend beyond this ‘libertarian’ vision, see S Makris, ‘A Smithian Political Economy Approach for the Competition Law of the 21st Century’ (2025) 88 Modern Law Review 712. ↑

52. Article 23, Regulation 1/2003. ↑

53. Commission Guidelines on the method of setting fines imposed pursuant to Article 23(2)(a) of Regulation No 1/2003 [2006] OJ C210/2. ↑

54. C‑457/10 P AstraZeneca EU:C:2012:770, para 132; reaffirmed in Case C-220/24 Aeroportul Cluj EU:C:2025:124, para 28. ↑

55. Cases C-295/12 P Telefónica EU:C:2014:2062, para 133 and C-220/24 Aeroportul Cluj EU:C:2025:124, para 28. ↑

56. Case C-280/08 P Deutsche Telekom EU:C:2010:603, para 90, applying the principle developed in Case C‑344/98 Masterfoods EU:C:2000:689, para 48. ↑

57. On the basis that only the Commission is empowered to find that there has been no breach of Article 102, whereas national regulators can only find breach or discontinue proceedings: Case C-375/09 Tele2 Polska EU:C:2011:270. ↑

58. Case C-280/08 P Deutsche Telekom EU:C:2010:603, para 89. ↑

59. Case C‑457/10 P AstraZeneca EU:C:2012:770, para 132. ↑

60. Case C-238/05 Asnef-Equifax EU:C:2006:734, para 63. ↑

61. Guidelines on the assessment of non-horizontal mergers under the Council Regulation on the control of concentrations between undertakings [2008] OJ C265/6, para 46. ↑

62. M.8124—Microsoft/LinkedIn, 6 December 2016. ↑

63. European Commission, ‘Mergers: Commission Clears Acquisition of Fitbit by Google, Subject to Conditions’ (Press Release, 17 December 2020). ↑

64. Case C-255/22 P Orlen EU:C:2024:790, para 96. ↑

65. ibid, para 96. ↑

66. ibid, para 97. ↑

67. ibid, para 96. ↑

68. ibid, para 95 (emphasis added). ↑

69. To use the language of AG Rantos in his Opinion in Case C-255/22 P Orlen EU:C:2024:466, para 58. ↑

70. See n 9. ↑

71. Cases C-204/00 P etc Aalborg Portland EU:C:2004:6, para 338. ↑

72. Case C-117/20 bpost EU:C:2022:202, paras 28–37. ↑

73. ibid, para 41. ↑

74. ibid, para 49. ↑

75. ibid, para 51. ↑

76. Case C-333/21 Superleague EU:C:2023:1011, para 130 (discussing specifically Article 102 TFEU). ↑

77. ibid, para 166 (discussing specifically Article 101 TFEU). ↑

78. See text accompanying n 33. ↑

79. Case C-280/08 P Deutsche Telekom EU:C:2010:603, para 224. ↑

80. ibid. ↑

81. To quote the defendant’s argument in Case C-280/08 P Deutsche Telekom EU:C:2010:603, para 205. ↑

82. Both the defendant and its rivals used revenues from call services to cross-subsidise their loss-making activity in the provision of line access services. The Commission, however, applied the margin squeeze test only to the defendant’s wholesale and retail-level line access activities. This approach was approved by the Court of Justice, which held that the Commission was entitled to proceed as if tariff rebalancing had occurred as required by EU telecommunications law, even though the German regulator had failed to implement tariff rebalancing for social policy reasons (to protect access to telephone lines for poorer consumers). ↑

83. Cases C‑165/19 P Slovak Telekom EU:C:2021:239, para 57 and C‑42/21 P Baltic Rail EU:C:2023:12, para 88. ↑

84. See Cases C-220/24 DB Station & Service EU:C:2022:832, para 82, T-136/19 BEH EU:T:2023:669, paras 784 and 961, and C-220/24 Aeroportul Cluj EU:C:2025:124, para 31. ↑

85. The Bundeskartellamt applied only domestic competition law, a choice that has been criticised as potentially inconsistent with its obligations under Regulation 1/2003. By the time the case made it to the Court of Justice, the Court spoke in terms of the application of EU competition law. ↑

86. Regulation 2016/679 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data [2016] OJ L119/1. ↑

87. Case C‑252/21 Meta Platforms EU:C:2023:537, paras 44–46. ↑

88. ibid, para 47 (emphasis added). ↑

89. The Court stated that, ‘in the context of the examination of an abuse of a dominant position by an undertaking on a particular market, it may be necessary for the competition authority … also to examine whether that undertaking’s conduct complies with rules other than those relating to competition law, such as’ – but by implication not limited to – ‘the rules on the protection of personal data laid down by the GDPR’: ibid, para 48 (emphasis added). ↑

90. ibid, para 50. ↑

91. ibid, para 51. ↑

92. See text accompanying n 75. ↑

93. Case C‑252/21 Meta Platforms EU:C:2023:537, para 55. ↑

94. As in eg Cases C-205/03 P FENIN EU:C:2006:453 and T-155/04 SELEX EU:T:2006:387, respectively. ↑

95. Case C-359/95 P Ladbroke Racing, para 33. ↑

96. Case C-333/21 Superleague EU:C:2023:1011, para 183. ↑

97. Case C-309/99 Wouters EU:C:2002:98, para 9. ↑

98. See n 94 above. ↑

99. Contrast Case C‑333/21 Superleague EU:C:2023:1011, paras 183–188, where the Court of Justice rejected the application of the Wouters exception to an organisation governed solely by private law (albeit not on that basis expressly). ↑

100. See nn 34 and 35 above. ↑

101. Draft Article 102 Guidelines, paras 167–71. ↑

102. Case C-233/23 Android Auto EU:C:2025:110, para 75. See also Case C‑42/21 P Baltic Rail EU:C:2023:12, which impliedly accepted that health and safety regulatory requirements could provide an objective justification for a de facto refusal to grant access to infrastructure, although the claimed necessity was deemed to be pretextual on the facts. ↑

103. Draft Article 102 Guidelines, para 168. ↑

104. ibid. ↑

105. As in eg Cases C‑42/21 P Baltic Rail EU:C:2023:12 and C‑457/10 P AstraZeneca EU:C:2012:770, respectively. ↑

106. Discussing the impact of privatisation specifically, see Case C‑245/24 Lukoil Bulgaria EU:C:2025:987. ↑

107. As in the E.ON/GDF pipeline cartel (Case COMP/39.401—E.ON/GDF, decision of 8 July 2009). ↑

108. As in Case C-331/21 EDP—Energias de Portugal EU:C:2023:812; in EDP, this fact was treated as evidence of the anticompetitive object of the arrangement contrary to Article 101(1) (see para 102). ↑

109. As acknowledged in Case C‑252/21 Meta Platforms EU:C:2023:537, para 47. ↑

110. See, eg, Cases C-413/14 P Intel EU:C:2017:632 and C-333/21 Superleague EU:C:2023:1011. ↑

111. Case C‑252/21 Meta Platforms EU:C:2023:537, para 47. ↑

112. Case C‑48/22 P Google Shopping EU:C:2024:726, para 224. ↑

113. ibid, para 225 (the Court expressly rejected Google’s argument that the Commission should have conducted a counterfactual analysis to find causation, see para 228). ↑

114. Including in the fields of airports, telecommunications, energy and rail transport. ↑

115. Case C‑165/19 P Slovak Telekom EU:C:2021:239. ↑

116. See, eg, Cases C‑501/06 P etc, GSK EU:C:2009:610, paras 102–04. ↑

117. For a recent study of such missing case-law, see M Heim, ‘The Curious Case of the European Commission’s Missing Antitrust Jurisprudence: Lessons from Abandoned Article 102 Investigations’ (2025) 16 JECLAP 285. ↑

118. Case M.8124—Microsoft/LinkedIn, 6 December 2016, para 177. ↑

119. See n 55. ↑

120. As in Cases 39.525—Telekomunikacja Polska (22 June 2011), C‑165/19 P Slovak Telekom EU:C:2021:239 and C‑42/21 P Baltic Rail EU:C:2023:12. ↑

121. S Dogan and M Lemley, ‘Antitrust Law and Regulatory Gaming’ (2009) 87 Texas Law Review 685. ↑

122. Case COMP/A.37.507/F3—AstraZeneca, 15 June 2005; upheld on appeal in Cases T‑321/05 AstraZeneca EU:T:2010:266 and C-457/10 P AstraZeneca EU:C:2012:770. ↑

123. Case AT.40588—Teva, 31 October 2024. ↑

124. Case AT.40394—Aspen Pharma, 10 February 2021. ↑

125. Case AT.40134—AB InBev Beer Trade Restrictions (29 June 2016). ↑

126. Case C-179/16 F. Hoffmann-La Roche and Others EU:C:2018:25. ↑

127. Case AT.40178—Car Emissions, 8 July 2021. ↑

128. Including the Commission decisions in Cases AT.39226 Lundbeck and AT.39612 Servier and the preliminary ruling in Case C-307/18 Generics (UK) EU:C:2020:52. ↑

129. See eg M Moore & D Tambini (eds), Regulating Big Tech. Policy Response to Digital Dominance (OUP 2022). ↑

130. See eg M Gal and O Aviv, ‘The Competitive Effects of the GDPR’ (2020) 16 JCLE 349 and CB Frey & G Presidente, ‘Privacy Regulation and Firm Performance: Estimating the GDPR Effect Globally’ (2024) 62 Economic Inquiry 1074. ↑

131. Recital 9, GDPR, which states that uneven implementation of the previous EU data protection regime could ‘distort competition’. ↑

132. See n 91. ↑

133. As in Cases C‑457/10 P AstraZeneca EU:C:2012:770 and C-591/16 P Lundbeck EU:C:2021:243, for example. ↑

134. As in the E.ON/GDF gas pipeline cartel, for example (see in particular, the judgment of the General Court on appeal, Case T-370/09 GDF Suez v Commission EU:T:2012:333. ↑

135. Ezrachi (n 9). ↑

136. I Lianos, ‘Polycentric Competition Law’ (2018) 71 Current Legal Problems 161. ↑

137. For a more detailed critique of the adequacy of this approach, see N Dunne, Competition Law and Economics Regulation (CUP 2015) ch 4. ↑

138. In Deutsche Telekom, for instance, the Court dismissed the significant impact of the national regulatory framework on the defendant’s behaviour on the rather blithe ground that it could have sought to petition the regulator to adopt more cost-reflective price regulation (in spite of evidence that the regulator had deliberately chosen to prioritise social solidarity over efficiency in its regulatory policymaking). For further critique of this case from a fairness perspective, see Dunne (n 137) 224–27. ↑

139. This was a significant theme within Deutsche Telekom: in such circumstances, the Court took a generous view of the Commission’s discretion to initiate competition law proceedings against the undertaking or infringement proceedings against the Member State maintaining the anticompetitive regulation. ↑

140. See n 69. ↑

141. As the General Court remarked in Lundbeck, a pay-to-delay case, ‘it is in the public interest to eliminate any obstacle to economic activity which may arise where a patent was granted in error’: Case T-472/13 Lundbeck EU:T:2016:449, paras 119, 390 and 487. This is also an issue in the background in Case C‑457/10 P AstraZeneca EU:C:2012:770 and Case AT.40588—Teva, 31 October 2024 amongst other cases. ↑

142. As provided by Article 4 (independence), Article 5 (resources) and Articles 10 and 13–16 (remedies and fines) of Directive (EU) 2019/1 to empower the competition authorities of the Member States to be more effective enforcers and to ensure the proper functioning of the internal market (ECN+ Directive) [2019] OJ L11/3. ↑

143. See n 74. ↑

144. Case C‑252/21 Meta Platforms, para 54. ↑

145. Case C-117/20 bpost EU:C:2022:202, para 49. ↑

146. See n 67. ↑

147. GDPR, recital 4. ↑

148. See fn. 15. ↑

149. Case C-377/20 Servizio Elettrico Nazionale EU:C:2022:379, para 61. ↑

150. Draft Article 102 Guidelines, para 45. ↑

151. ibid, para 60. ↑

152. For example, data protection rules limit the extent to which firms may profitably combine data-sets containing personal information, while environmental protection rules limit the extent to which firms can use cheaper but more polluting technologies. ↑

153. For example, EU employment law prevents firms from dismissing pregnant employees and requires the provision of paid leave. ↑

154. On regulatory compliance costs generally, see OECD, OECD Regulatory Compliance Cost Assessment Guidance (OECD Publishing 2014). ↑

155. For discussion of the level-playing-field notion generally, see eg M Gillis, ‘Let’s Play?: An Examination of the “Level Playing Field” in EU Free Trade Agreements’ (2021) 55 Journal of World Trade Law 715. ↑

156. Case C‑165/19 P Slovak Telekom EU:C:2021:239, paras58–60, and C‑42/21 P Baltic Rail EU:C:2023:12, para 89. See also Case T-136/19 BEH EU:T:2023:669, para 873. ↑

157. Commission Guidelines on the applicability of Article 101 to horizontal co-operation agreements [2023] OJ C259. ↑

158. ibid, para 520. ↑

159. ibid, para 564. ↑

160. ibid, para 520. ↑

161. C‑252/21 Meta Platforms EU:C:2023:537, para 51. ↑

162. As in Cases C‑457/10 P AstraZeneca EU:C:2012:770 and AT.40588—Teva, 31 October 2024. ↑

163. Although coordination or dominance does not attract antitrust liability itself, the jurisprudence emphasises that such firms have heightened obligations: coordinating firms must ensure that each still ‘determine[s] independently the policy which he intends to adopt on the … market’ (Case C-40/73 Suiker Unie EU:C:1975:174, para 173) while dominant firms have a distinctive ‘special responsibility’ to avoid distorting competition (Case C-322/81 Michelin EU:C:1983:313, para 10). ↑

164. The argument would be that such firms are expected to comply both with their regulatory obligations and to refrain from practices that might reasonably be considered at odds with the broad thrust of a regulatory scheme. An analogy might be drawn to the concept of potential competition, which is established where there are ‘real and concrete possibilities’ of new entry (see eg Case C‑307/18 Generics (UK) EU:C:2020:52, para 36). This standard can be met even if there is considerable ambiguity as to whether new entrants can lawfully compete in a market given existing regulatory barriers to entry. In pay-to-delay cases, for instance, competition law does not tolerate agreements to exclude potential competitors even if there is a reasonable chance that patent law might facilitate this result: for instance, in Case T‑472/13 Lundbeck EU:T:2016:449, paras 121–31, the court found potential competition where there was only a 50–60 per cent chance that the would-be competitor could lawfully enter the market concerned. ↑

165. Coordination falling within Article 101 could also act as a jurisdictional trigger for the application of competition law; a ‘collusion-plus’ approach would then treat coordinated non-compliance or gaming efforts as a restriction of competition. An example of this arose in the Car Emissions cartel, see n 127. ↑

166. The German competition authority also drew on national case-law on consumer protection to support its approach to the concept of market abuse. ↑

167. Ezrachi (n 9). ↑

168. See eg A Ayal, ‘The Market for Bigness: Economic Power and Competition Agencies’ Duty to Curtail It’ (2013) 1 JAE 221; L Khan, ‘The Ideological Roots of America’s Market Power Problem’ (2018) 127 Yale LJF 960; and A Gerbrandy and P Phoa, ‘The Power of Big Tech Corporations as Modern Bigness and a Vocabulary for Shaping Competition Law as Counter-power’ in M Bennett, H Brouwer and R Claassen (eds), Wealth and Power: Philosophical Perspectives (Taylor & Francis 2022). ↑

169. See n 163. ↑

170. Case C-85/86 Hoffmann La Roche EU:C:1979:36, para 91. ↑

171. Cases C‑165/19 P Slovak Telekom EU:C:2021:239, para 57 and C‑42/21 P Baltic Rail EU:C:2023:12, para 88. ↑

172. Case C‑252/21 Meta Platforms EU:C:2023:537, para 44. ↑

173. ‘[P]ublic authorities can legitimately choose complementary legal responses to certain conduct that is harmful to society through different procedures forming a coherent whole so as to address different aspects of the social problem involved’: Case C-117/20 bpost EU:C:2022:202, para 49. ↑

174. Exemplified by Case C-280/08 P Deutsche Telekom EU:C:2010:603. ↑

175. Under the ECN+ Directives, national competition authorities must have the ability to impose behavioural and structural remedies on defendants (Article 10), plus fines of up to at least 10 per cent of the annual worldwide turnover of the undertakings concerned (Article 15) for competition law violations. These provisions mirror the powers of the Commission under Regulation 1/2003. ↑

176. See Directive 2014/104/EU on certain rules governing actions for damages under national law for infringements of the competition law provisions of the Member States and of the European Union [2014] OJ L349/1. ↑

177. See eg Case C-413/14 P Intel EU:C:2017:632, para 138. ↑

178. See eg the discussion in Case C-307/18 Generics (UK) EU:C:2020:52. ↑

179. Surveying the case-law, see P Ibanez Colomo, ‘Anticompetitive Effects in EU Competition Law’ (2020) 17 JCLE 209. ↑

180. This requires it to be demonstrated that coordination ‘has as its actual or potential effect the prevention, restriction or distortion of competition, which must be appreciable’ (Superleague, para 169). Thus it is sufficient that the conduct merely limits potential (as opposed to existing) competition, although this does require it to be established that ‘real and concrete possibilities’ for new market entry exist (Case C‑307/18 Generics (UK) EU:C:2020:52, para 36). ↑

181. Relied upon, inter alia, in Roche & Novartis (n 126), Car Emissions (n 127) and various ‘pay-to-delay’ cases. ↑

182. Case C-333/21 Superleague EU:C:2023:1011, para 165. ↑

183. Case C-377/20 Servizio Elettrico Nazionale EU:C:2022:379, para 50. ↑

184. ibid, para 54. ↑

185. See n 113. ↑

186. See eg Case C-333/21 Superleague EU:C:2023:1011, para 163. ↑

187. See eg discussion in Case C-377/20 Servizio Elettrico Nazionale EU:C:2022:379, paras 53–54. ↑

188. Case C-413/14 P Intel EU:C:2017:632, para 139; followed in Case C-680/20 Unilever EU:C:2023:33. ↑

189. Case C‑48/22 P Google Shopping EU:C:2024:726, para 240. ↑

190. ibid, para 254. ↑

191. Case C-333/21 Superleague EU:C:2023:1011, para 167. ↑

192. ibid. ↑

193. Case C-680/20 Unilever EU:C:2023:33, para 45 (emphasis added). ↑

194. Case C‑48/22 P Google Shopping EU:C:2024:726, para 255. ↑

195. ibid, para 256. ↑

196. Named for Case C-62/86 Akzo v Commission EU:C:1991:286. ↑

197. The Akzo test is an off-shoot of the Areeda-Turner test, which was developed to provide an objective means to determine when low prices – typically considered desirable from a competition policy perspective – should nonetheless be found abusive by virtue of their exclusionary effects (see PE Areeda and DF Turner, ‘Predatory Pricing and Related Practices under Section 2 of the Sherman Act’ (1975) 88 Harvard Law Review 697). The Areeda-Turner test posits that prices below average variable cost (AVC, a proxy for margin cost) should be deemed conclusively abusive on the basis that no rationale firm would continue to produce if it cannot obtain prices at or above AVC, so that its choice to do so can be taken to disclose sufficient evidence of anticompetitive intention. The Akzo case extended the Areeda-Turner test to prices above AVC but below average total cost where there is evidence that such prices are ‘determined as part of a plan for eliminating a competitor’ (C-62/86, para 72), on the basis that such prices have the capacity to exclude as-efficient competitors, while the actual intention evidence serves to counter any potentially innocent explanations for the pricing behaviour. ↑

198. See n 51 and accompanying text. ↑

199. Including AstraZeneca, Teva, Aspen Pharma, Telekom Polska, Lundbeck and the quite fantastical case of Baltic Rail, where the defendant opted to destroy 19km of its own rail infrastructure rather than share it with a freight rail competitor under the EU rail liberalisation framework. ↑

200. See n 46. ↑

201. Case C‑457/10 P AstraZeneca EU:C:2012:770, para 99. ↑

202. Case AT.40588—Teva, 31 October 2024. ↑

203. ibid, para 1033. ↑

204. ibid, para 1033, citing Case T-612/17 Google Shopping EU:T:2021:763, para 257. ↑

205. Opinion in Case C-413/14 P Intel v Commission EU:C:2016:788, para 66. ↑

206. Ezrachi (n 9) 50. ↑

1. Introduction

What roles do regulatory norms play in EU competition law assessment? The question matters both for the consistency of the supervisory framework for economic activity in the EU, and because it helps us to better understand the legitimate concern of the competition rules, a key theme in contemporary competition policy debates. Competition law (also known as antitrust) aims to promote and protect effective competitive behaviour by economic actors – the jurisprudence uses the term ‘undistorted’^[1] competition to describe the ultimate policy objective. Regulation, a broader and more ambiguous concept,^[2] often applies to and constrains the same market conduct, though typically in pursuit of alternative or additional policy objectives. Regulation can enhance opportunities for effective competition or limit the scope for abusive market behaviour; but it can also ‘distort’^[3] competition in ways that raise concerns from a competition policy perspective, whether by diminishing overall competitive dynamics or by facilitating or incentivising anticompetitive conduct.^[4] EU law has long endorsed the concurrent application of the competition rules to regulated firm behaviour, an approach that maximises the effectiveness of competition policy. Yet the jurisprudence struggles with a problem of internal coherence, seeking to protect the distinctiveness and primacy of the competition rules while also finding space for the increasing centrality of contextual analysis in contemporary competition enforcement. Moreover, applying competition law to behaviour that has already been affected by other regulatory norms creates its own challenges, including issues of legitimacy, consistency and proportionality. These questions are the focus of this piece.

In principle, the competition rules apply only to the autonomous behaviour of firms, and such conduct – whether taking the form of agreements, dominant firm practices or mergers – is prohibited only where it is demonstrated that anticompetitive effects – whether actual, potential or presumed – are likely to result. Yet where competition law is applied to regulated behaviour, there is a higher likelihood that suspect conduct has been influenced or even directed by the regulatory obligations and that any resulting market distortions can be attributed, wholly or partly, to the impact of the regulation. In such contexts, competition law assessment must consider not only whether and how the defendant’s behaviour has caused anticompetitive harm, but also the extent to which that behaviour and its effects were determined by the underlying regulatory framework. The greater the overlap between the scope of the regulatory obligations and the competition law theory of harm, the more complex this task becomes. Since competition law and regulation must co-exist as legitimate and socially valuable mechanisms of market control, there are arguments, both practical and legal,^[5] for ensuring that these separate regimes are interpreted and applied in a consistent, ideally mutually reinforcing, manner. But because this question typically arises where competition law is applied in markets already subject to ex ante regulation, there can be tension between preserving the effectiveness of competition law and its underlying goal of protecting undistorted competition, while simultaneously respecting other public interest values pursued by ‘distortive’ regulation.

A great deal has already been written about the relationship between competition law and other regulation, including much of my own work. This article seeks to make an original contribution by focusing on how substantive regulatory norms are accommodated within competition law analysis and what this tells us about competition law’s place within the wider market regulatory sphere. We leave aside the distinct questions of when competition law itself exhibits certain ‘regulatory’ characteristics,^[6] and of the appropriate competition policy response to anticompetitive State regulation.^[7] In doing so, the article adopts an essentially nonnormative, conceptual approach, which aims to work effectively within the confines of the existing jurisprudence rather than making more aspirational claims for wholesale change.^[8] Our analysis is not directly concerned with the goals of competition law, although there are plenty of articles of this sort about.^[9] But when considering how competition law interacts with other regulatory fields that pursue a more diverse range of policy objectives, it is impossible to avoid the question of what competition law itself seeks to achieve. Above, we described the goal of competition law broadly as the protection of ‘undistorted’ competition, an approach that, although not without its critics, finds support in recent jurisprudence and legislation.^[10] The article thus seeks to provide a more granular exploration of what we mean by undistorted competition, in a context where markets can never really be free of distortions, whether stemming from private or public sources. By considering the extent to which other regulatory norms can provide a conduit by which to reflect a wider range of public interest values within competition enforcement, while also recognising the legal and practical limits on this approach, we seek to obtain a better understanding of what Odudu termed ‘the legitimate scope’^[11] of competition law.

The article is structured as follows. Section II introduces and distinguishes competition law from other regulatory instruments of market supervision. Section III sets out and analyses the ‘orthodox’ understanding of the inter-relationship, which prioritises the independence and effectiveness of competition law as a mechanism of market control. Section IV discusses the more ‘contextual’ contemporary approach, considering the ways in which regulation may provide a relevant element of context within competition law analysis. Section V steps back, to consider the normatively-oriented question of how the ‘outsourcing’ of liability standards in competition law to other regulatory fields can be explained. It examines a variety of perspectives: the effectiveness of competition law as a regulatory enforcement mechanism; regulatory norms as a constraint on the acceptable parameters of competition on the merits; and regulatory (non-)compliance as a positive anticompetitive action. Section VI brings these disparate strands of analysis together and concludes.

2. Distinguishing competition law from ‘regulation’ as instruments of market supervision

A necessary starting point is to distinguish competition law from other forms of market regulation. Competition law comprises a set of legal rules that aim to ensure open, undistorted and fair competition in the internal market, by controlling the acquisition and exercise of market power (as distinct, inter alia, from public power).^[12] Within the framework of the EU treaties, the competition rules operate as important flanking provisions to secure the objective of ‘a highly competitive social market economy’.^[13] Yet most competition cases do not belabour this point. Instead, enforcement tends to be a relatively technical exercise aimed at identifying and proscribing various agreements (under Article 101 of the Treaty on the Functioning of the European Union, or TFEU), practices of dominant undertakings (under Article 102 TFEU) and mergers (under the EU Merger Control Regulation, or EUMR)^[14] that are deemed restrictive of competition in some legally relevant sense. A key theme within recent case-law is the notion of ‘competition on the merits’:^[15] although competition law is aimed at anticompetitive firm behaviour, the rules do not seek to prevent firms, even where they are dominant or acting in concert, from competing ‘on the merits’ (what is also termed ‘normal competition’^[16]). What amounts to normal or meritorious competition in an individual case is a question of competition law. It is thus determined by applying the legal tests and principles developed in the competition jurisprudence,^[17] instead of deferring to the behavioural or regulatory norms within a sector.^[18]

Contemporary EU competition law has several salient aspects for our purposes. First, competition assessment is, in almost every instance,^[19] a highly context-specific exercise: conduct is held to restrict competition within its market circumstances,^[20] while also taking account of any countervailing pro-competitive efficiencies that might be generated.^[21] Second, and somewhat by contrast, competition law is largely sector-indifferent in its application. With a few exceptions,^[22] the competition rules apply across all sectors, supervising vastly different industry practices under the same legal provisions. Finally, the competition rules governing anticompetitive agreements and dominant firm practices are largely ex post in application, serving to punish infringing behaviour after-the-fact.^[23] The EUMR, conversely, is prospective in analysis,^[24] but the object of the exercise is not (generally)^[25] to regulate the behaviour of newly-merged firms going forward, but rather to catch and proscribe potentially harmful concentrations before these occur. In this article, we focus on the impact of pre-existing regulation on the application of Articles 101 and 102 TFEU to market behaviour, while considering insights from merger control where appropriate.

The term regulation is less precise in its meaning. It can be defined as broadly as ‘all forms of economic or social influence’ that may stem from public or private sources, or as a narrowly as ‘a specific set of commands’ imposed by the State and pertaining to closely identified activities.^[26] For the purposes of this article, we adopt a relatively loose definition, referring to State-imposed (or at least State-endorsed)^[27] norms that, directly or indirectly, control or constrain market activity by economic actors, broadly understood. This encompasses both sector-specific regulation, which applies to market participants that are active only within particular sectors (such as telecommunications, energy, transport or digital platform services), and to regulatory regimes of more general application (such as data protection, health and safety, or environmental law).

Applying this definition, it can readily be seen that most markets today are subject to multiple layers of regulation that, to borrow the language of the competition jurisprudence, may ‘distort’ competitive outcomes to a greater or lesser extent. In some instances, it is obvious how overlap with competition law arises: for example, regulation may grant an economic actor special or exclusive rights that give it a position of market dominance, or regulation may mandate pro-competitive behaviour (such as imposing access requirements) or proscribe direct abuses of market power (through, for example, pricing controls). In other instances, the potential for interaction with competition law arises more obliquely, for example where regulation constrains a firm’s ability to maximise its profits or provides opportunities for exploitative or exclusionary behaviour.

Given the breadth of our understanding of regulation, and by implication the variety of regulatory models that it encompasses, it is impossible to neatly distinguish competition law from other forms of market regulation that may apply to the same conduct.^[28] Two distinctions of relevance might, however, be advanced. First, although the precise ‘goals’ of EU competition law are debated, broadly speaking the rules aim at a straightforward objective, namely ‘the protection of undistorted competition’.^[29] While a regulatory regime may seek to enhance effective competition, this need not be its only policy objective, and most of the examples of regulation that overlap with competition law within the case-law have pursued other regulatory goals additionally or alternatively,^[30] some of which may sit uncomfortably with the efficiency-focused perspective of the competition rules. Second, one of the necessary trade-offs that comes from the wide applicability of competition law is that its prohibitions are sketched in broad and abstract terms. This creates flexibility and scope for progressive development of the law to address emerging market problems, but such advantages arise at the expense of ex ante precision and certainty. Some regulatory regimes operate in a similar manner, yet regulation can be more prescriptive, identifying with greater exactness the specific regulated actors or the content of mandatory norms. Specification can assist enforcement, enabling the supervision of complex market problems. Yet where regulated entities are eager to resist regulation, a precisely defined and limited set of rules may facilitate strategic behaviour to escape or exploit regulatory controls, a problem considered in Section 5.

The broad scope of the competition rules coupled with the pervasiveness of regulation in modern markets leads to overlapping jurisdiction and potential conflict. The jurisprudence demarcating the inter-relationship of competition law and other regulation within the EU legal system is complex. One strand of case-law emphasises the independence and effectiveness of competition law as a mechanism of market control, reflecting an ‘orthodox’ understanding of the inter-relationship that prioritises but also effectively siloes competition law. Another strand embraces the more contextual approach that has come to the fore in the general competition case-law, recognising the extent to which regulation may feed into and even determine the substance of competition law assessment. While these jurisprudential currents are not inherently at odds, they reflect notably different visions of this inter-relationship. There are also points of inconsistency and even conflict within the case-law, which remain underexplored in the existing jurisprudence. In what follows, we set out first the apparently resolute ‘orthodox’ position (Section 3) and then the more nuanced contextual approach (Section 4). Doing so identifies several cases in which it becomes more difficult to reconcile the orthodox and contextual accounts, which are explored in Section 5.

3. The interaction of competition law and regulation: the orthodox view

Within EU law, the orthodox position is that the competition rules are essentially distinct from and unaffected by other regulatory norms that may apply to the same market behaviour. This approach is exemplified by Deutsche Telekom, in which the defendant was held to have breached Article 102 in circumstances where it had complied with its sector-specific regulatory obligations, where the sector regulator had either set or endorsed the pricing practices that comprised the abuse, and where the national regulatory framework pursued a legitimate public policy objective (social solidarity) by endorsing a pricing structure that the Commission subsequently held to be inefficient.^[31] An analogy of ‘two barriers’ to lawful market participation was suggested by the Advocate General in Deutsche Telekom: economic actors must comply with the overarching obligations imposed by the competition rules, while simultaneously discharging any parallel regulatory obligations.^[32] Typically, moreover, these legal requirements are enforced by different regulators, who each decide independently whether the norms falling within their jurisdictions have been breached. This orthodoxy has several important implications.

A. No ouster of competition law in the presence of regulation

First, the existence of regulation within a market does not generally oust the applicability of competition law to the same behaviour, even if the regulatory regime has a significant impact on the conduct or underlying level of competitiveness within the market.^[33] The principal exception, the ‘State action’ doctrine, is narrowly construed,^[34] applying only where anticompetitive conduct is either required by regulation or where the regulatory framework removes all scope for competitive (and, by implication, anticompetitive) behaviour.^[35] The logic behind this defence is that competition law regulates anticompetitive behaviour by firms and not merely uncompetitive market structures; but its narrowness in practice suggests a deep reluctance to decouple the abstract notion of competition from the concrete actions of competitors.^[36]

This can be contrasted with the approach in US antitrust law, which more readily embraces a rule of antitrust preclusion where ‘a regulatory structure designed to deter and remedy anticompetitive harm’ can be identified.^[37] There are at least three reasons which explain the distinctive position to be found in EU law. First, the competition rules are primary EU law, meaning that there is a formal obstacle within the EU’s hierarchical legal structure to allowing ‘special’ regulatory rules found in secondary or national legislation to take precedence over the ‘general’ competition rules.^[38] Second, even if regulation takes account, inter alia, of the goal of effective competition, the assumption is that it almost always pursues ‘objectives which differ from those of [EU] competition policy’.^[39] The policy objectives of one set of rules cannot be subsumed into the other; with the result that, to the extent we wish to pursue both sets of goals simultaneously, both sets of rules must remain in operation.^[40] Finally, EU competition law incorporates a remarkably strong principle of effectiveness, which has aided the progressive development and expansion of these rules in many areas.^[41] Limiting any rule of antitrust preclusion to only the most exceptional circumstances where regulation removes all scope for (anti)competitive conduct serves to protect the effectiveness of competition law by ensuring that these rules continue to apply in the largest number of circumstances.

This means, for instance, that a dominant firm can be held to have engaged in an abusive margin squeeze (which entails manipulating the relationship between wholesale and retail prices for a product) where both price levels have been set or approved by a sector-specific regulator.^[42] Similarly, industry efforts to reorganise a sector suffering from an overcapacity crisis can be treated effectively as a cartel, even if undertaken at the behest of and in accordance with a plan devised by national government.^[43] And under the EUMR, strong governmental support has not conventionally provided a compelling counterweight where the Commission concludes that a proposed concentration is likely to significantly impede effective competition,^[44] although this position may prove more difficult to sustain in future.^[45]

The fact that, in EU competition law, defendants can breach the rules even where their conduct or the anticompetitive effects that follow are heavily influenced by State-imposed regulation is made possible by the lack of a fault or ‘bad’ intention requirement. Thus, under Article 102, ‘proof of the deliberate nature of the conduct and of the bad faith of the undertaking in a dominant position is not required for the purposes of identifying an abuse’.^[46] Under Article 101, it is ‘irrelevant’ to the assessment of whether a restriction by object exists that the coordinating parties acted in pursuit of ‘legitimate’ (ie not deliberately anticompetitive) objectives.^[47] The ‘by effects’ limb of that provision, additionally, considers only the impact in fact of an agreement on the market, rather than the reasons motivating its adoption.^[48] Intention also plays a very limited role in determining the essentially factual question of how a merger is likely to affect competition if it proceeds.^[49]

The absence of a fault requirement can be explained by the fact that the competition rules are aimed at behaviour that harms competition; yet the neoclassical economics that underpin the conventional competition law framework assumes that all firms ultimately intend to ‘harm’ competition in a literal sense by increasing their market share and profits at the expensive of competitors.^[50] Indeed, such self-interested behaviour has long been considered an essentially virtuous aim to the extent that it drives the competitive dynamics necessary to maximise efficiency.^[51] The upshot is that competition law takes an explicitly ‘objective’ approach, at least in principle, to the assessment of whether behaviour amounts to normal, meritorious – and so permissible—competition. To the extent that an absence of fault or bad intention is relevant within EU competition enforcement, it goes primarily to the penalty imposed: breaches must be committed intentionally or negligently to attract fines under Regulation 1/2003,^[52] while the influence of regulation on (otherwise autonomous) firm behaviour can be reflected in a discount on the overall fine.^[53]

B. Competition law and regulatory compliance are unrelated

Second, the question of whether behaviour violates competition law is formally ‘unrelated to its compliance or non-compliance with other legal rules’,^[54] regardless of how strong the link between the regulatory framework and the allegedly abusive behaviour. This means, on the one hand, that behaviour that conforms to a relevant regulatory standard might still breach competition law, if it amounts to an anticompetitive agreement or abuse of dominance.^[55] This position has been defended within EU law on the bases, inter alia, that decisions of (national) regulators should not bind the Commission,^[56] even if the regulator purports to apply and find compliance with the competition rules in parallel;^[57] and that the absence of a fault requirement, discussed above, means that good faith regulatory compliance is insufficient to avoid competition liability.^[58] The general principle that regulatory compliance is ‘unrelated to’ competition law liability would seem to infer that regulatory non-compliance is also largely irrelevant from an antitrust perspective, though Meta Platforms (discussed in Section 4.A) nuances this assumption.

Accordingly, the case-law recognises that ‘in the majority of cases, abuses of dominant positions consist of behaviour which is otherwise lawful under branches of law other than competition law’.^[59] The jurisprudence is more complicated under Article 101, insofar as the sanction of nullity in Article 101(2) means that agreements that violate the provision are void as a matter of contract law too. But even in this context, the Court considers that compliance with other regulatory norms is, generally, ‘not, as such, a matter for competition law’,^[60] and so does not determine the question of whether an arrangement has the object or effect of restricting competition. Under the EUMR, when assessing whether a merged entity is likely to engage in harmful behaviour, the Commission takes account of any pre-existing regulation that might effectively inhibit the merged entity from doing so.^[61] It has also sought, in certain instances, to repackage regulatory compliance as a dimension of quality, to the extent that users place economic value on the underlying public interest protected (such as the level of data privacy offered by a product).^[62] Yet the Commission is clear that, if the objection to a concentration is a fear the merged entity will fail to comply with other regulatory rules in future, ‘[s]uch concerns are not within the remit of merger control and there are regulatory tools better placed to address them’.^[63]

C. Constrained powers of Commission when acting as competition regulator

This links to a third implication of the orthodox separation of competition law and other regulation within EU law, namely the limits of the Commission’s powers when ‘acting as competition regulator’.^[64] In the recent case of Orlen, the Court of Justice emphasised that, when exercising its competition powers, the Commission is restricted to imposing remedies or penalties that ‘may respond to the competition concerns [it has identified] on the market concerned’.^[65] Thus the Commission cannot legitimately use competition enforcement to advance other regulatory goals or interests,^[66] unless these can be absorbed into the ‘competition issues’ at stake.^[67] The Court acknowledged that the Commission’s competition enforcement activity ‘could not lead to a result which is contrary to’ other regulatory norms within EU law,^[68] and in particular, it cannot make binding remedies that would conflict with other (EU-level, at least) regulation. But that it quite different from saying that the Commission might positively impose ‘obligations … motivated by policies other than competition law policies’^[69] through the avenue of competition enforcement; this is something, the Court was clear, the Commission cannot do.

The attitude of the Court in Orlen potentially poses a not-insignificant significant hurdle to realising more ambitious applications of the competition rules,^[70] by establishing that competition law enforcement can pursue only ‘competition issues’. Of course, there is a risk of circularity, insofar as what we consider to legitimately constitute a ‘competition issue’ may vary depending upon how broadly we treat the legitimate objectives of competition law. But the judgment clarifies a point of importance to this article: competition enforcement is not an instrumental tool by which to advance a grab bag of ‘other’ regulatory objectives but must confine itself to responding to competition-related concerns.

D. Ne bis in idem and parallel enforcement of competition law and regulation

Finally, this orthodox understanding of the relationship between competition law and other regulatory norms – as discrete sources of legal obligations for market actors, applied by distinct regulators pursuing different public interest goals – creates potential for overlapping jurisdiction and enforcement. Compliance with regulatory obligations does not, as explained, certify that the regulated behaviour also complies with competition law. Yet overlapping jurisdiction does not always generate conflict, and it is entirely possible that the same behaviour might be found to violate both the competition law and the regulatory standards.

For many years, EU competition law recognised a distinctive principle of double jeopardy or ne bis in idem, which precluded the parallel application of multiple sanctions only if it was demonstrated that the regulatory regime pursued the same legal interest as the competition rules, in addition to unity of facts and offender.^[71] This unique approach reinforced the orthodox view of competition law and regulation as distinct and unrelated tools for market supervision. More recently, the Grand Chamber in bpost has realigned competition law with other areas of EU law by shifting to an idem factum standard requiring only the same facts and offender.^[72] The Court nonetheless recognised that duplicate proceedings and penalties might be permitted where they are a necessary and proportionate response to achieve public interest objectives or to protect the rights and freedoms of others.^[73] Of particular relevance is the Court’s strong assertion, when considering the proportionality of parallel proceedings, that ‘public authorities can legitimately choose complementary legal responses to certain conduct that is harmful to society through different procedures forming a coherent whole so as to address different aspects of the social problem involved, provided that the accumulated legal responses do not represent an excessive burden for the individual concerned’.^[74] Yet parallel jurisdiction cannot be exercised wholly unilaterally if it is to comprise a proportionate response. Instead, the Court in bpost stressed the need for sufficient ‘coordination’ between the competition regulator and other prosecuting authority to ensure a truly ‘complementary’ response.^[75]

4. The interaction of competition law and regulation: the contextual view

One shortcoming of the ‘two barriers’ understanding of the interaction of competition law and other regulation is that it sets up a dichotomy between these instruments of market supervision that does not reflect the reality of how most markets work today. In its competition case-law, the Court of Justice increasingly emphasises that the determination of whether conduct amounts to a breach of the rules requires a detailed assessment of context, meaning that decision-makers must take account of ‘all of the relevant factual circumstances’^[76] and the ‘economic and legal context’^[77] in which allegedly anticompetitive conduct occurs. The underlying regulatory landscape, insofar as it may shape competition dynamics and/or influence the defendant’s own behaviour, is an obvious contextual element that should be considered. Although EU law rejects the proposition that the presence of specific regulation should oust the more general competition rules as a matter of law, the former may play a significant role in determining whether competition law is breached on the facts.

A. Relevance of regulatory norms in principle: from Deutsche Telekom to Meta Platforms

Indeed, even as the Court in Deutsche Telekom sidelined the State action defence and discounted the significance of good faith regulatory compliance in competition assessment,^[78] it recognised the relevance of regulatory context insofar as it ‘contributes to the determination of the competitive conditions under which an undertaking … carries on its business in the relevant markets’.^[79] The Court elaborated three areas where regulatory context might be relevant: ‘defining the relevant markets, assessing the abusive nature of such conduct or setting the amount of the fines’.^[80] In reality, the invocation of regulatory context in Deutsche Telekom was somewhat disingenuous, enabling the Court to endorse a cost–price calculation mechanism that was ‘at odds with the realities of the market’^[81] for the defendant and its rivals.^[82] Despite this, in both Slovak Telekom and Baltic Rail, the Court drew upon this statement from Deutsche Telekom to conclude that the regulatory context justified application of a less exacting legal standard under Article 102, reiterating that ‘a regulatory obligation can be relevant for the assessment of abusive conduct’.^[83] This language is echoed in later cases.^[84] It is the 2023 judgment of the Grand Chamber in Meta Platforms, however, that provides the most explicit support for a potentially pivotal role for regulatory norms within competition law assessment.

The Meta Platforms case began with an infringement decision of the German competition authority against the owner of the Facebook social network, which aimed at the latter’s user-data-gathering policies. In essence, the decision construed Meta’s policies for use of Facebook services as an abuse of market power,^[85] using as a proxy for unfairness in competition law terms the fact that these policies were non-compliant with relevant data protection standards as reflected in the EU’s General Data Protection Regulation (GDPR).^[86] The Court of Justice, called upon to advise the fairly unconvinced national court, began by reiterating the orthodoxy of the distinctiveness of competition law from other regulation, including the different legal interests protected and enforcement by different regulators.^[87] Yet it quickly turned to acknowledge the applicability of regulatory standards and the role of compliance within the nominally distinct task of competition assessment.

The Court went much further than in previous cases, where the regulatory context had been described as potentially ‘relevant’ to this determination; in Meta Platforms, the Court held that ‘compliance or non-compliance … may, depending on the circumstances, be a vital clue’ as to the permissibility of behaviour under competition law.^[88] Notably, although the questions referred pertained only to data protection, the Court’s response was pitched in broader terms.^[89] Though the Court did not elaborate at length on how regulatory compliance might offer a ‘vital clue’ to competition liability, it noted the ‘great importance’ of having access to user data for many digital economy business models,^[90] so that it might undermine the effectiveness of competition enforcement to ‘disregard the reality of this economic development’.^[91] Echoing bpost,^[92] the Court then considered how national authorities should collaborate to minimise ‘the risk of divergences’,^[93] an approach that it grounded in the general EU law duty of sincere cooperation in Article 4(3) TEU. The judgment thus offers a more nuanced account of the inter-relationship between competition law and other regulation, by acknowledging that the latter may influence or even determine the application of the former in practice, depending upon its effects within a relevant market.

B. Relevance of regulatory norms in practice

The impact of the regulatory context on competition law assessment can be grouped into three broad categories: where regulation effectively excludes competition law liability; where it changes the structural competition dynamics within a market; and where it influences the behaviour of economic actors in a manner that prevents – or conversely provokes – competition law infringements.

(i) Regulation as exemption of competition law liability

Most radically, though also most exceptionally, regulation might remove the scope for applying the competition rules to market behaviour or at least preclude a finding of liability. The competition rules apply to ‘undertakings’, a concept that has been interpreted to exclude certain (typically highly regulated) market actors that pursue objectives related to social solidarity or that exercise quasi-public powers.^[94] The State action defence, noted above, provides an effective means to avoid Articles 101 and 102, but only where the applicable regulatory regime removes all capacity for ‘autonomous conduct’, competitive or otherwise, by regulated undertakings.^[95] Similarly, under Article 106(2) undertakings that are ‘entrusted’ by Member States with the operation of services of general economic interest can be exempted from application of competition law, but only to the extent that being subject to such obligations would ‘obstruct the performance, in law or in fact’ of this task. These three disparate exceptions share a common theme: although the would-be defendant is engaged in what is, strictly speaking, ‘economic’ activity on the market, the nature or impact of the underlying regulatory framework removes its actions from the purview of the competition rules as a matter of law, regardless of their effects on competition in fact.

Alternatively, the operation of the regulatory framework may provide a good defence that enables a defendant to rebut the claim that its conduct is harmful from a competition perspective. The Article 101 jurisprudence recognises the ‘Wouters exception’, which enables certain prima facie restrictive arrangements to escape prohibition ‘by effect’ on the basis that the coordination pursues ‘legitimate objectives in the public interest which are not per se anticompetitive’, and subject to indispensability and proportionality requirements.^[96] Formally, the exception does not hinge upon the impact of regulation on the activities under scrutiny; that is, the concept of ‘legitimate objectives in the public interest’ does not require that defendants are empowered to achieve those objectives under public regulation. But it is notable that, on the facts of Wouters, the organisation concerned (the Dutch bar association) was expressly tasked under domestic legislation to adopt the rules under scrutiny, ‘in the interests of the proper practice of the [legal] profession’.^[97] In this sense, a parallel can be drawn to the exceptions to the ‘undertaking’ concept that have similarly developed in the jurisprudence.^[98] In each instance, although the activity concerned is ‘economic’ in nature the predominant interests at stake are not, and typically (indeed, perhaps necessarily in practice) this will be reflected in the underlying regulatory framework.^[99]

Under Article 102, it is difficult to successfully argue that regulatory obligations deprive a dominant undertaking’s conduct of its ‘autonomous’ nature so as to preclude antitrust scrutiny.^[100] But there may be greater scope, where behaviour is driven by such considerations, for advancing a claim of ‘objective necessity’, which can serve as an objective justification, or defence, to a prima facie finding of abuse.^[101] This potential is seen most clearly in the case-law on refusal to deal, where the jurisprudence explicitly recognises that the ‘applicable regulatory framework’ may impose constraints on dominant firm behaviour in a manner that essentially legitimates what would otherwise be an abusive refusal to grant access to the dominant firm’s property.^[102] But objective necessity claims are narrowly construed in practice, and defendants have the burden of establishing that their behaviour is both necessary to achieve the public interest aim and that the actual or potential exclusionary effects resulting from the conduct are proportionate to the allegedly necessary aim.^[103] The objective necessity defence moreover does not apply where a dominant defendant claims, ‘rightly or wrongly’, that its behaviour was necessary to prevent other market actors from breaching their regulatory obligations,^[104] the idea being that private actors should leave the task of enforcement to the professionals, ie the applicable public regulator.

(ii) Regulation as a structural feature of market competition

Most of the time the underlying regulatory framework does not provide anything so dramatic as a wholesale dispensation from competition scrutiny. It is more usual for regulation to feed into antitrust assessment either by heightening or dampening competition dynamics (ie the possibilities for effective competition within a market) or by altering (whether by constraining or directing) the behaviour of economic actors in ways that make anticompetitive conduct more or less likely.

Regulation, first, can function structurally as a barrier to entry or expansion by competitors, typically enhancing the power of incumbent market actors and making anticompetitive effects more likely or plausible. The case-law recognises a wide variety of such circumstances, including statutory monopolies and intellectual property rights.^[105] Consideration of barriers to entry is a key component of the initial assessment of dominance under Article 102, and can also be highly informative of whether a merger is likely to create or strengthen a dominant position, so as to violate the ‘significant impediment to effective competition’ standard under the EUMR. Such structural effects may endure even after regulation: so, for instance, entities that previously benefited from special or exclusive rights under regulation may continue to experience de facto benefits after such legislation is removed,^[106] firms may maintain anticompetitive arrangements after the regulatory justification for doing so has disappeared,^[107] or firms may even seek subsequently to reimplement regulatory barriers through private arrangements after these have been dismantled by market liberalisation.^[108]

Regulatory barriers to entry may also be relevant to the assessment of likely anticompetitive effects that follow from behaviour.^[109] It was noted above that contemporary competition assessment is highly attuned to the specific context in which conduct occurs; increasingly, the case-law also emphasises that competition law aims at behaviour that has at least potential anticompetitive effects.^[110] The regulatory framework may therefore feed into the evaluation of whether coordination violates Article 101(1) or that single firm conduct constitutes an abuse of existing market power, insofar as the regulatory context may influence ‘the consequences of a certain practice in the market or for consumers’.^[111]

A more complex question is whether the anticompetitive effects must be caused by the impugned behaviour. In Google Shopping, the Court of Justice described the establishment of a causal link between a firm’s behaviour and the claimed actual or potential anticompetitive effects as one of the ‘essential constituent elements of an infringement of competition law’.^[112] This would appear to preclude a finding of breach of Articles 101 or 102 where the harmful effects on competition are attributable more or less entirely to the impact of the relevant regulation, even if defendant undertaking(s) have engaged in what looks like wholly unmeritorious behaviour from a competition perspective. Accordingly, while there is a high threshold to prove that regulation robs firm behaviour of its autonomous character, there may be greater scope to argue that the pervasive distortive effects of regulation mean the necessary causal link cannot be found. Yet Google Shopping does not require an exacting causation standard, with the Court accepting that it is sufficient merely to demonstrate a ‘correlation’ between suspect practices and the evolution of the market concerned, provided that a causal relationship can be confirmed by ‘additional information’ such as the position of market participants.^[113] The question in regulated markets is therefore whether it is possible to identify a sufficient diminution in effective competition arising from the suspect behaviour, over and above any limitations or distortions arising from the regulatory restrictions.

Conversely, regulation may facilitate new entry which enhances competition overall. This is, for instance, a principal objective of much of the market liberalisation legislation introduced in the EU since the 1990s.^[114] There are a number of common regulatory techniques which are deployed frequently to increase opportunities for competition: such as requiring the removal of existing monopoly rights that protect (typically dominant) incumbent operators from new entry; imposing structural separation on vertically-integrated incumbents to reduce incentives to engage in exclusionary behaviour; and requiring incumbents to share existing infrastructure or even data with potential new rivals, to enable or assist entry. Regulation aimed at market-opening may be supplemented by regulation that controls the behaviour of former monopolists in liberalised sectors, a point developed further below. In such circumstances, antitrust takes account of the altered market dynamics effectuated by the presence of the regulation, so that a defendant cannot, for instance, argue that its behaviour should be assessed as if unencumbered by its regulatory burdens.^[115] On the other hand, the regulatory framework may make it more plausible for a defendant to claim that its superficially restrictive behaviour generates countervailing efficiencies.^[116]

(iii) Regulation and its impact on market behaviour

Regulation may also influence a firm’s market behaviour in ways that have implications for competition assessment: either because it directly controls market behaviour or because it incentivises conduct that generates (anti)competitive effects. Here, several scenarios can be envisaged.

Most constructively, regulation may guide a firm to competition law-compliant behaviour. Where appropriately calibrated and effectively implemented, a regulatory framework may prevent regulated firms from engaging in the sorts of behaviour that would otherwise cause competition problems: by limiting a firm’s capacity to charge prices that are excessively high or have exclusionary effect, for instance, or by preventing use of ‘unfair trading conditions’ in contractual relations. This outcome is most likely to manifest itself obliquely, in an absence of antitrust problems in a sector.^[117] To the extent that context-specific competition assessment takes place prospectively, the general starting point is that the market behaviour of economic actors is shaped and constrained by other regulatory norms that apply within that sector and to that conduct.^[118] This is most salient in merger control, where it is assumed that firms comply with their regulatory obligations, and that their scope for ‘free’ market conduct is curtailed accordingly.

More contentiously, regulation may guide a firm to behaviour that is essentially at odds with competition law, a scenario most likely to arise where the regulatory regime endorses ‘inefficient’ market outcomes in pursuit of other policy goals. In this case we may encounter the possibility flagged in Section 3, namely prosecution under the competition rules of behaviour that is not only fully compliant with, but has also been largely influenced by, a regulatory framework that deliberately prioritises other socially valuable goals.^[119] This is an essentially unsatisfactory outcome, which is considered further below.

The first two scenarios each envisage effective compliance with regulatory obligations, whether this leads to competitive market outcomes or not. But firms do not always respect their regulatory obligations or make good faith efforts to do so. Within the competition jurisprudence, it is possible to identify several scenarios whereby an antitrust defendant’s liability is essentially contingent upon its interaction with another regulatory framework. It is these cases, in which regulatory norms function as the ‘hook’ for additional antitrust liability, which raise the most interesting but also the most difficult questions regarding the impact of regulatory norms on competition law.

Most straightforwardly, a regulated firm may simply disregard the constraints that are placed on its behaviour. In Meta Platforms, a digital firm’s data-gathering policy was deemed to amount to an abuse of its market power because the policy failed to respect the relevant data protection standard. A similar approach has been adopted in the Commission’s enforcement activity in liberalised utilities sectors, where failure to abide by regulatory duties to share infrastructure is also construed as an abuse of dominance by the incumbent operator.^[120] Here, the regulatory violation – which may or may not have been prosecuted by the sector regulator in parallel – is usually obvious; the more interesting question is how and why the behaviour might additionally become the focus of an antitrust investigation.

Beyond the problem of mere non-compliance, the competition case-law discloses a variety of instances where liability hinges on a firm’s disruptive interaction with a regulatory framework. In essence, these cases involve so-called ‘regulatory gaming’, namely ‘[b]ehaviour that abuses a neutral or procompetitive regulatory structure and wields it as a tool to accomplish [anticompetitive] results’.^[121] What distinguishes the non-compliance cases from the regulatory gaming ones is that, generally speaking, in the latter the firm’s behaviour is unproblematic (or at least unprosecutable) under the relevant regulatory regime. Instead, antitrust liability derives from the defendant’s efforts to avoid the regulatory constraints while remaining within the strict letter of the law (what we term ‘bad faith compliance’), or to exploit opportunities created by the existing regulation to anticompetitive ends (‘abusive compliance’).

Many examples of bad faith compliance relate to the highly regulated pharmaceutical sector. In AstraZeneca,^[122] the first EU competition case that directly embraced regulatory gaming as a theory of harm, the defendant took advantage of loopholes in the rules on pharmaceutical marketing authorisation and weaknesses in the administration of national patent offices to exclude would-be generic competitors. In Teva,^[123] the defendant strategically lodged and withdrew divisional patent applications in a manner permitted by the regulatory framework but which was clearly intended to frustrate new entry. Most remarkably, in Aspen Pharma the defendant successfully implemented an excessive pricing campaign in a context where every Member State had domestic price regulation, by exploiting design flaws which allowed it to manipulate the regulatory price-setting processes.^[124] For our purposes, what is most notable about these cases is how they make the leap from attributing antitrust liability for non-compliance with the express rules of a regulatory framework to non-compliance with the broader spirit or underlying policy objectives of those rules.

In abusive compliance cases, the regulatory context is primarily of relevance insofar as it provides an explanation for strategic behaviour that otherwise lacks an anticompetitive rationale. This is exemplified by AB InBev,^[125] where the defendant changed its packaging for beer sold in France and the Netherlands to prevent cross-border resales in Belgium, where it charged higher wholesale prices. This change – to no longer include information in both Dutch and French on beer sold in countries other than Belgium – only made sense as an exclusionary strategy in view of Belgian legislation on food labelling, which meant that larger retailers only purchased products labelled in both languages. In Roche & Novartis,^[126] the defendants engaged in collusive lobbying efforts to persuade the Commission to change its advice on the off-label prescribing of a certain drug, which enhanced the profits of both cartelists. Similarly, in Car Emissions,^[127] car manufacturers agreed to design new vehicles to the minimum regulatory standard for nitrogen oxide cleaning, an arrangement that the Commission prosecuted as a secret cartel. What is notable is that the cartelists explicitly agreed to adhere to their existing regulatory obligations, albeit knowing that these were likely to become more demanding in future; in doing so, however, they effectively agreed not to compete on cleaning technology as a dimension of competition, which the Commission construed as a limitation both of consumer choice and future innovation. Pharmaceutical pay-to-delay agreements, whereby drug companies pay off potential new entrants in the guise of patent litigation settlements, can also be seen as abusive compliance.^[128]

The real world is complicated, of course, and some regulations may be equivocal in terms of their overall effects on competition, so that a neat dichotomy between regulation that limits competition versus that which enhances competition often cannot be drawn. The GDPR provides an example of such mixed effects. On the one hand, by imposing limits on what market actors can do with personal data, it constrains the market power of Big Tech companies, which have disproportionately plentiful access to such commercially valuable information.^[129] On the other, GDPR compliance is an expensive business, and so the very existence of these rules, which apply to all data controllers regardless of size, may constitute a barrier to entry.^[130] The GDPR says little about its intended effects on competition beyond a vague reference to levelling the playing field,^[131] yet it has a significant impact on many digital business models, as recognised in Meta Platforms.^[132]

C. Understanding the role of regulatory norms

Regulation accordingly feeds into competition assessment in numerous (occasionally contradictory) dimensions, from reducing the scope to identify autonomous market behaviour to enhancing the risk that firm conduct generates harmful effects. Where competition law is applied in regulated markets, the competition rules do not generally endorse or apply the regulatory norms as such. But by accepting a market context where these norms reflect the accepted ‘rules of the game’ for market participants, competition law essentially reinforces the more diverse public interest values that are reflected within other regulatory regimes. Occasionally, this means deferring entirely to alternative values by allowing the regulatory regime to displace the application of competition law or to provide at least a good defence. More usually, these norms are absorbed within the contextual analysis, and the case-law increasingly recognises the legitimacy of aligning the thrust of competition analysis with pre-existing regulatory obligations (provided, of course, that these are not fundamentally at odds). There is obvious potential for tension, where for instance competition law turns its focus on market power that has been created by regulation as, say, a reward for innovation or to ensure the universal delivery of public services. But even here, competition analysis typically finds a way to accommodate potentially conflicting values, by distinguishing between the lawful scope of intellectual property rights and efforts to extend those rights beyond what is envisaged by IP law,^[133] for instance, or by recognising the fundamentally different position of liberalised and non-liberalised markets.^[134] Even if the goal of competition law is merely to achieve and protect undistorted competition, the space given to other regulatory norms within the contextual approach allows for the accommodation of complementary (and sometimes conflicting) values. These cases thus illustrate the tendency towards a ‘sponge’^[135]-like or ‘polycentric’^[136] quality within competition law, to the extent that a diversity of values can be accommodated – if not necessarily pursued directly – within contemporary enforcement.

The above discussion nonetheless identified two broad areas where this narrative of an essentially passive-but-tolerant inter-relationship between competition law and other regulatory norms does not provide an effective explanation. The first scenario arises where a (typically national) regulatory framework favours inefficient market outcomes in pursuit of non-economic objectives, albeit leaving sufficient scope for independent conduct to attract antitrust attention. The unsatisfactory position in the current law has been noted: in line with the orthodox approach, unless the narrow State action defence applies, distortions generated by the regulatory framework can be reflected only by discounting the penalty that might be imposed.^[137] This position can be critiqued from two dimensions. It creates unfairness for, or at least imposes a heavy burden on, defendants, who seem to have a positive responsibility to seek to align their regulatory obligations with the perceived demands of competition law.^[138] But even leaving the question of fairness aside, when viewed through the lens of contextual analysis this approach gives little weight to the obviously significant impact of the regulatory framework on competition dynamics in the market concerned. Instead, there is a strong argument that the (typically, non-economic) considerations factored into the regulatory framework should inform, though not necessarily determine, the competition enforcer’s understanding of what ‘competition on the merits’ looks like in a sector. While this may be insufficient to provide a good defence where there is further evidence that the defendant’s conduct is objectively non-meritorious in competition law terms, it would at least provide some recognition of the public interests that motivate the regulatory framework. To the extent that the latter are pretextual or at least insufficient to justify the restrictions on competition entailed, the more appropriate competition policy response is to attack the regulation directly.^[139]

A second category of cases that require more in-depth consideration are those involving non-compliance and regulatory gaming, which pose a conceptual dilemma from a competition policy perspective. Through such behaviour, a firm may engage in conduct that is harmful to effective market competition, which seems to fall within the legitimate scope of the competition rules. Yet insofar as the firm’s ‘unmeritorious’ behaviour revolves principally around its interactions with another regulatory framework, one might ask what makes this an appropriate subject for antitrust intervention, as opposed to the more obviously and directly applicable regulatory regime. In cases of non-compliance, where the behaviour violates the regulatory norm, the question is how to justify competition enforcement as opposed to (or in addition to) direct regulatory enforcement. In cases of bad faith compliance, the question is why competition law might take it upon itself to redesign or fill in perceived gaps within a regulatory framework rather than deferring to the regulation itself. Cases of abusive compliance are less intimately entwined with the obligations imposed by a regulatory framework, making these easier to explain as examples of ‘regulation as context’, discussed above. Yet such cases almost unavoidably involve value judgements regarding the nature of the defendant’s interaction with the regulatory regime, and so raise a similar question to other categories: how can we explain the role played by regulatory norms within the competition law assessment in such cases? This is the focus of the next section of this article.

5. Understanding the role of regulatory norms in competition assessment: explaining the non-compliance and regulatory gaming case-law

The crux of concern is that these cases involve, in essence, the ‘outsourcing’ of liability standards within competition law to other regulatory fields. A competition lawyer schooled in the ‘two barriers’ approach to the competition law/regulation inter-relationship would not question the legitimate existence of regulatory norms ‘other than competition law policies’.^[140] Yet it is less obvious why competition law should effectively defer to standards of behaviour developed in other regulatory policy areas – a problem that becomes more acute in cases of regulatory gaming, which involve deference to vaguer understandings of what a regulatory regime ought to entail if working well. The objection is not that the policy goals reflected in the regulatory framework lack legitimacy or value. Instead, the question is why competition law internalises these goals by applying regulatory norms as the determinative standard of competition liability in such cases. Put another way, why should protection of the goals, values and policy choices that are reflected in a specific regulatory framework fall within the legitimate scope (and enforcement purview) of competition law in addition?

Before exploring this question, let us recap several core principles of the ‘orthodox’ jurisprudence. Formally, antitrust compliance is unrelated to regulatory compliance within EU law: a firm might breach the competition rules while respecting its regulatory obligations, and vice versa. This links to the absence of a fault requirement, meaning that good faith regulatory compliance does not immunise a firm from competition liability in respect of the same behaviour. Competition enforcers are limited to the pursuit of ‘competition issues’; a somewhat woolly limitation, but one which confirms that the competition rules are not a panacea for all public policy ills. Accordingly, to explain the non-compliance and regulatory gaming cases, we must consider why as a matter of competition law such behaviours come within the legitimate scope of the competition rules. To do so, we explore the case-law from three distinct perspectives – through the lenses of effectiveness, competition on the merits, and regulatory non-compliance and gaming as strategic action – asking whether each provides a plausible explanation for addressing the behaviour under competition law.

A. Effectiveness of competition law as a regulatory enforcement mechanism

It is important to recognise why both non-compliance and regulatory gaming cases typically arise in practice: namely, the effectiveness of competition law as a regulatory enforcement mechanism, unconstrained by substantive or procedural limitation that may hinder enforcement of the regulatory framework directly. This is apparent from the background to many cases considered above. In Telecom Polska, the sector regulator had repeatedly enforced national telecommunications law against the defendant, but the penalties it imposed were insignificant and failed to deter non-compliance. In Slovak Telekom, the defendant had evaded its regulatory obligation to grant access to infrastructure for five years but received no pushback from the (captured or under-resourced) regulator. In Meta Platforms, the problem was a mismatch of jurisdiction: while the data privacy concerns were felt most acutely by German consumers and policymakers, power to enforce the GDPR lay with the Irish data protection authority, which dragged its feet. The deficiency of national patent systems is a key undercurrent in many competition cases in pharmaceutical markets.^[141]

Viewed pragmatically, it is easy to understand why competition law is used to reinforce regulatory norms: in effect, competition enforcement might get the job done more effectively than the regulation itself. Competition law brings much to the table as a mechanism to enforce regulatory rules. In particular, the broad sweep of its prohibitions means that competition law is sufficiently wide and flexible to catch such behaviour, including problematic conduct that might lie outside the formal parameters of the regulation itself. EU law also imposes demanding minimum standards to protect the independence, resourcing and sanctioning powers of competition authorities,^[142] which sectoral regulators often lack. The resistance within EU law to accept antitrust preclusion even in the presence of significant regulation supports the proposition that competition law can legitimately seek to resolve market problems that arise from malfunctioning regulation, providing a ‘complementary legal response’ in order to form ‘a coherent whole’ of market supervision.^[143]

Yet the bare fact that competition law can enforce regulatory norms more successfully provides an inadequate explanation for these cases. On the one hand, using competition law as a sticking-plaster on ineffective regulation gives little weight to the substance, and substantive limits, of the underlying regulatory framework. Cases like bpost and Meta Platforms demonstrate that, if competition law is applied to ‘regulated’ problems, this requires a degree of intentionality and coherence and must be done in a manner that recognises the ‘respective powers and competences’^[144] of each regime. Deploying competition law on the pragmatic basis that it can better achieve the goals of a regulatory framework fails to acknowledge the central role of the underpinning regulation in such cases, both in terms of how ‘the social problem involved’^[145] and the appropriate enforcement response should be framed.

On the other hand, justifying these cases on an effectiveness basis offers little insight into how such regulatory failures can also become, in the language of Orlen, ‘competition issues’.^[146] Regulation, as discussed, often incorporates a diversity of policy goals that extend beyond the protection of undistorted competition. Regulatory standards of acceptable behaviour typically seek to advance values outside the scope of competition law as such. To take an notable example, the GDPR states that ‘[t]he processing of personal data should be designed to serve mankind’, and explains that the rules seek to balance a variety of fundamental rights including privacy and family life, freedom of thought, freedom of expression, freedom to conduct a business, fair trial rights, and cultural, religious and linguistic diversity.^[147] While it is difficult to argue against the importance of these varied considerations, it is clear that a data protection standard – such as that applied to impose antitrust liability in Meta Platforms – reflects a wider ranges of values than the mere protection of undistorted competition. If we want to bring the former within the legitimate scope of the latter, we need a more nuanced explanation as to when and why diverse regulatory concerns are also ‘competition issues’.

B. Regulatory compliance as a reflection of ‘competition on the merits’

Instead, we return to the concept of ‘competition on the merits’: the idea that competition law does not seek to constrain firms from competing, even aggressively, but merely limits market activity to the acceptable parameters of ‘normal competition’. The concept is prominent in recent jurisprudence and in the Commission’s Draft Guidelines on Article 102.^[148] The latter follow the approach in SEN,^[149] by decoupling the notion of ‘conduct departing from competition on the merits’ from ‘capability to produce exclusionary effects’, while indicating that both elements must be satisfied to establish exclusionary abuse of dominance.^[150] The Draft Guidelines nonetheless suggests that capability can be presumed for a variety of practices,^[151] making competition on the merits the key determinant of liability in many circumstances.

The proposition that competition law protects only ‘competition on the merits’ might justify the direct application of regulatory norms to the extent that the standard of ‘normal’ competition implies that market actors, inter alia, follow their regulatory obligations. Regulatory compliance is often costly for market actors, whether because the regulation curtails a firm’s scope to engage in profitable activity^[152] or requires it to incur costs it would prefer to avoid,^[153] or simply because of direct compliance costs.^[154] While escaping or minimising regulatory obligations can give individual undertakings a competitive advantage, doing so undermines the public interest values that the regulation advances and protects. Viewed in this manner, non-compliance or gaming of regulatory norms is inherently at odds with what we understand meritorious, and thus permissible, market competition to look like. Such an approach aligns with the ‘level playing field’ notion that permeates EU law, which is premised on the idea that market actors should be subject to common rules and restrictions to ensure fair competition.^[155] Since the regulatory framework structures the ‘fair’ competitive game in the level playing field of the marketplace, it is reasonable to require adherence under competition law even if some regulatory rules pursue values that are non-economic in nature.

At its simplest, the existing regulatory framework might be taken to reflect the lawful scope of a firm’s freedom in the marketplace. By incorporating regulatory standards into antitrust assessment, we do not ‘enforce’ these norms as such. Yet because a defendant can have no legitimate expectation to act contrary to its regulatory obligations, it cannot object where this behaviour is curtailed or punished through competition enforcement. Whereas in Slovak Telekom and Baltic Rail the Court denied that a regulatory violation might comprise an abuse in itself, it accepted that the presence of regulation may permit a less demanding legal standard for antitrust intervention.^[156] It justified this essentially on the basis that pre-existing regulation already circumscribed what the defendants could reasonably expect to be allowed to do in the marketplace. Competition law need not maintain the fiction that, say, a firm can validly refuse to grant access to its infrastructure unless an antitrust duty to deal arises exceptionally under Article 102, if sector-specific regulation has already imposed an unambiguous mandatory sharing obligation. A similar logic is discernible in the Commission’s approach to sustainability agreements in its Horizontal Cooperation Guidelines.^[157] Where ‘appropriate regulation’ exists to adequately address market failures, the Guidelines are sceptical of the acceptability of private coordination in tandem,^[158] ‘given that the legislator has already decided that each undertaking must individually comply with the obligation in question’.^[159] Yet private action may be more necessary to address ‘residual market failures that are not or not fully addressed by public policies and regulation’.^[160] In the latter case, undertakings have plausible claims for greater freedom of (private) action, since regulation has not yet delimited the legitimate scope of market activity. This reflects, in the language of Meta Platforms, ‘the reality of economic development’ in regulated markets.^[161]

Yet treating a regulatory framework as merely indicative of the limits to a firm’s economic freedom provides an incomplete explanation of the non-compliance and gaming case-law. First, it is less convincing in respect of behaviour that violates the spirit of a regulatory norm rather than its express provisions. It is one thing to assert that a defendant can have no legitimate expectation, for example, to refuse to comply with a mandatory sharing obligation imposed by sector-specific regulation; it is quite another to make this claim regarding behaviour that is, say, objectively ‘misleading’ but not strictly speaking unlawful.^[162] Arguably, the mere existence of collusion or dominance limits what firms can legitimately expect to do in markets, even absent abusive behaviour;^[163] and applying this logic, we might extend the proscriptive potential of regulation to a wider subset of behaviours at variance with its policy objectives more broadly.^[164] Yet this creates a legal certainty problem: how can firms know in advance what sorts of behaviours are likely to conflict with the broader ‘regulatory orbit’, if this is not clearly delimited by regulation? Second, this approach provides only a partial account of what happens in non-compliance and gaming cases: it tells us why defendants cannot object to competition enforcement covering the same ground, but not why such behaviour can legitimately be construed as a ‘competition issue’ as well as a regulatory breach.

A more muscular version of this argument is what we call the ‘dominance-plus’ approach. From this perspective, because non-compliance or regulatory gaming is inherently non-meritorious, any conduct of this sort is fair game under competition law provided that a threshold criterion for antitrust jurisdiction (most naturally a dominant position under Article 102) is also satisfied. While competition law cannot compel firms directly to comply with their regulatory obligations, it can punish non-compliance or gaming insofar as such conduct falls within the discrete prohibitions in the competition rules. It is the existence of significant market power (or the fact of coordination, if we extend the logic to a collusion-plus approach under Article 101^[165]) that brings the matter within the scope of competition law; the regulatory misconduct then provides the plus factor that crystallises the competition infringement. This was, effectively, the position of the Bundeskartellamt in Meta Platforms.^[166]

This approach has a degree of intuitive appeal. It is hard to argue that a firm acts meritoriously if it disregards or takes advantage of its regulatory obligations, though it is more debatable as to whether this is ‘normal’ market behaviour. Although the regulatory regime may reflect values beyond the task of protecting undistorted competition, the enforcement of such values in non-compliance and gaming cases can be explained as an example of competition law’s ‘sponge’-like quality, a term developed by Ezrachi to describe the capacity to absorb and reflect policy influences from other sources such as market regulation.^[167] Undeniably, this is an understanding of competition law’s role that reflects a more overtly ‘political’ vision of antitrust as a tool directly to regulate economic power, a perspective which typically gives less weight to more ‘legal’ questions of the permitted parameters of a dominant firm’s market freedom.^[168] Yet it is not obviously precluded by the Article 102 jurisprudence, which recognises the ‘special responsibility’ of dominant firms to protect effective competition,^[169] and which does not require misuse of the market power that underlies a dominant position in order to establish ‘abuse’.^[170]

Yet even accepting that dominant firms have a special responsibility to avoid distorting competition, objections can be raised against construing regulatory non-compliance or gaming of norms as ‘abuses’ in an antitrust sense. The case-law establishes that, even if ‘a regulatory obligation can be relevant for the assessment of abusive conduct’, the presence of regulation ‘cannot relieve the Commission of the requirement of establishing that there is abuse’ on the facts.^[171] While Meta Platforms saw the most ambitious use of regulatory norms to determine liability, the Court still stopped short of treating regulatory non-compliance as conclusive proof of antitrust abuse, and it emphasised the differing ‘objectives and tasks’ pursued by competition law compared with other regulatory fields.^[172] In bpost, where the Court acknowledged that competition law might ‘legitimately’ tackle behaviour that is also subject to regulatory controls, it clearly envisaged that parallel proceedings would address diverse regulatory objectives.^[173] It is doubtful whether repackaging a regulatory violation as an antitrust one merely due to the presence of dominance (or indeed coordination) respects this limitation. The dominance-plus approach also sits uncomfortably with the principle that good regulatory compliance is generally irrelevant to the objective question of whether conduct attracts antitrust liability, provided that the regulatory regime allows sufficient scope for autonomous behaviour that can come within the reach of the competition rules.^[174]

This concern – that defendants might somehow be damned if they do and damned if they don’t respect regulatory norms – highlights an additional objection, namely the risk of disproportionate punishments for what are, at their core, regulatory violations. bpost does not prohibit the parallel enforcement of regulatory norms, although it includes proportionality considerations in the determination of whether an exception to ne bis in idem is permissible in individual cases. But even where proceedings are limited to competition enforcement, the consequences for defendants are often much more disadvantageous than if prosecution was pursued under the regulation directly.^[175] Even if one has little sympathy for firms that disregard their regulatory obligations, reinterpreting a regulatory violation as a competition abuse can raise the stakes considerably, including the possibility of private damages claims alongside public enforcement.^[176] There is also the question of how to impose fair and sensible limits on the scope of liability from a dominance-plus perspective: might any regulatory breach – whether of tax law, zoning rules, labour protections etc – become an antitrust abuse merely due to the existence of dominance (or collusion)? Concerns about disproportionate liability are even more acute in regulatory gaming cases, where the defendant has not in fact violated the relevant regulatory norm, so that antitrust enforcement not only reinforces but extends regulatory liability.

Yet within the emerging Article 102 jurisprudence the existence of conduct departing competition on the merits is not (typically) the sole determinant of whether an abuse arises; additionally, at least in principle, the conduct must have the capability to produce exclusionary effects.^[177] An effective limiting principle for non-compliance and regulatory gaming cases could conceivably be derived by adding a robust requirement to demonstrate anticompetitive effects arising from the regulatory breach/gaming, alongside a ‘but for’ causation standard directly linking the behaviour to its harmful effects. Demonstrating actual (or at least potential)^[178] and appreciable anticompetitive effects would squarely locate the problem within the sphere of ‘competition issues’ that fall within the legitimate scope of competition law. Requiring a ‘but for’ causation standard, additionally, would enable us to explain why this regulatory abuse should also generate liability under competition law. Such an approach has much to recommend it in principle.

It is at this juncture, however, that we run up against the constraints of the existing case-law. Although the recent jurisprudence makes greater space for effects-based considerations, in practice it still falls short of both proposed elements in most instances. While the case-law is not a model of clarity,^[179] it is only with respect to the ‘by effect’ limb of Article 101(1) that we find anything close to such a robust effects requirement.^[180] Under the ‘by object’ limb,^[181] it is essentially unnecessary to examine effects where an arrangement discloses intrinsic harm to competition ‘by its very nature’.^[182] Under Article 102, which is the basis for most of the cases considered here, behaviour needs only to be capable of restricting competition,^[183] a standard that can be met despite evidence that the conduct ‘has not produced actual anti-competitive effects’.^[184] To the extent that the case-law imposes a causation requirement, it foresees merely a need to demonstrate a ‘correlation’ between the defendant’s conduct and the claimed (likelihood or capacity to cause) harm.^[185] There are sound competition policy reasons to defend a more prophylactic approach, in order to prohibit overtly restrictive arrangements,^[186] or where the presence of a dominant firm implies a structural absence of competition.^[187] Yet the continuing distance between the rhetoric and reality of the effects-based approach means that a simple dominance-plus perspective may provide an over-inclusive account of when regulatory non-compliance and gaming of norms should also constitute competition issues.

C. (Non) compliance as a positive strategic action

Another way of thinking about these cases is to view non-compliance or regulatory gaming as a positive strategic action for competition law purposes. Here, the focus is the deliberate choice of defendants to neglect or ‘game’ their regulatory obligations as part of an anticompetitive course of conduct: the antitrust problem stems not merely from the fact the regulatory norm has been disregarded or disrupted, but also from the motivation for this behaviour. In effect, the regulatory framework is ‘weaponised’ to anticompetitive ends, and it is this dimension – rather than the regulatory element as such—that is pursued under competition law. The key aspect is thus the strategic behaviour of the defendant undertaking in terms of how it interacts with, takes advantage of, or disregards its regulatory obligations.

To provide an effective explanation of the non-compliance and regulatory gaming case, the question is whether (and in what circumstances) ‘strategic’ behaviour of this sort can breach the competition rules. Recent case-law provides support in several dimensions for an approach to liability that hinges on evidence of the anticompetitive intention of defendants. The transformative Intel judgment emphasised the relevance of evidence of an anticompetitive ‘strategy’ to find abuse of dominance.^[188] The Court in Google Shopping endorsed a focus on the ‘positive acts’ undertaken by the defendant,^[189] which served to distinguish its (abusive) actions from other (permissible) forms of self-preferencing. It also held that, when engaging in context-specific assessment, it is ‘clearly legitimate for the Commission to assess subjective matters, namely the motives underlying the business strategy in question’.^[190] The more contextual approach to ‘by object’ determination under Article 101 similarly extends consideration of the inherent ‘nature’ of coordination to the ‘aims which that conduct seeks to achieve from a competition standpoint’.^[191]

Yet intention-based requirements are controversial in competition law. Two broad objections can be advanced against conceptualising the non-compliance and gaming cases as instances of ‘strategic’ behaviour: one substantive and one practical. In substantive terms, it has been argued that in the presence of market power or coordination, certain forms of conduct should be objectively unacceptable due to their inherent capacity to distort competition, regardless of what a defendant seeks to achieve by its actions. This perspective is reflected in the jurisprudence, which rejects a determinative role for intention evidence within the framework of EU competition analysis. The perceived ‘objective aims’ of coordination are distinguished from the ‘subjective intention’ of coordinating parties, with the latter ‘not decisive for the purposes of the application of Article 101(1)’.^[192] Under Article 102, while ‘proof of an anti-competitive intention’ may constitute ‘an indication of the nature and objectives pursued by the strategy of that undertaking and, on that basis, may be taken into account’, the recent case-law reiterates that ‘demonstration of the existence of such an intention is neither necessary nor sufficient, in itself, to establish the existence of an abuse’.^[193] In Google Shopping, the Court described ‘the existence of any anticompetitive intent’ as ‘only one of a number of facts which may be taken into account’,^[194] and emphasised that there is ‘no obligation to establish the existence of such intent … in order to render Article 102 applicable’.^[195]

The substantive objection is not trivial, and imposing an across-the-board obligation under Articles 101 or 102 to demonstrate subjective anticompetitive intention in addition to establishing behaviour that is objectively harmful to competition would clearly be inappropriate. Yet as demonstrated by the case-law on predatory pricing, intention can play a valuable, and determinative, role in establishing antitrust liability in certain circumstances. Under the Akzo test,^[196] anticompetitive intention serves to confirm the unacceptability in antitrust terms of behaviour that is objectively ambiguous from a competition policy perspective.^[197] Intention evidence can usefully perform a similar function in regulated conduct cases, marking the dividing line between behaviour that is purely a problem within the context of a regulatory regime and behaviour with a dual character that also merits antitrust intervention (ie that reveals a ‘competition issue’). Here, again, we come up against a similar difficulty to that encountered when discussing the use of an effects-based limiting principle under the dominance-plus approach, namely the limits of the existing case-law. Yet it could be argued that recognising another intention-based exception where this possibility is already established in the jurisprudence is less radical than requiring a proper effects-and-causation analysis. In any event, these elements are not mutually exclusive.

The more procedural objection to making liability hinge on evidence of an anticompetitive strategy relates to the difficulties of demonstrating a ‘guilty mind’ where defendants are typically large corporate entities. On the one hand, there is the problem of determining what counts as the intention of the undertaking as a whole; on the other, there is the inherent tension, outlined above, of construing evidence of a desire to enhance an undertaking’s own market share as evidence of an objectively anticompetitive intention.^[198]

Yet, practically speaking, this has not prevented the accumulation of significant evidence of a defendant’s deliberate efforts to evade or defraud its regulatory obligations in many of the non-compliance and gaming cases discussed earlier.^[199] Indeed, the frustrating aspect of the Court’s insistence on the formal irrelevance of intention is that it is hard to reconcile with what occurred in these cases. For instance, whereas the General Court in AstraZeneca maintained that ‘proof of the deliberate nature of the conduct and of the bad faith of the undertaking in a dominant position is not required’,^[200] the higher Court of Justice stressed that such a hypothetical scenario was ‘radically different from … the present case’.^[201] The Teva decision, similarly, is imbued with the language of (bad) intention: the Commission found that the defendant ‘aimed at’, ‘consciously’, ‘intentionally’ and ‘clearly pursued’ its anticompetitive divisional patent applications.^[202] The decision formally maintained that there is ‘no requirement’ to establish anticompetitive intention,^[203] however, and cited Google Shopping to the effect that ‘the existence of an intention to compete on the merits, even if it were established, could not prove the absence of abuse’.^[204] Yet it is essentially impossible to envisage a plausible ‘good faith’ version of regulatory non-compliance or gaming as a competition law theory of harm. We have thus another example of the problem that so concerned Advocate General Wahl in his pivotal Opinion in Intel, with the Court saying one thing but doing quite another in practice.^[205] The policy question is whether it makes sense for EU competition law, in essence, to keep its options open by maintaining the fantasy that motivation is not a key element of the theory of harm in such cases; a question that pits legal principle against a desire to preserve the practical effectiveness of competition law to the greatest extent possible.

A focus on strategic anticompetitive intention in cases where the theory of harm is tied up with the question of regulatory compliance has a further aspect to commend itself: it could arguably provide a more satisfactory explanation for the problematic cases where a defendant is held to breach competition law despite its compliance with regulatory obligations that strongly influence its supposedly anticompetitive conduct. We identified two objections to this jurisprudence: the possible unfairness to defendants faced with the responsibility to correct policy choices made by a regulator, and the fact that it gives insufficient weight to how regulation affects ‘competition on the merits’ in the sector concerned. Yet these objections are less compelling if would-be ‘good faith’ compliance is not all that it seems. Where there is evidence that a defendant is acting in a strategically anticompetitive manner in its interaction with the regulatory framework, it is manifestly less unfair to pursue such behaviour under competition law. Evidence of anticompetitive intention can also provide an effective rejoinder to the claim that the regulatory regime demarcates the boundaries of ‘meritorious’ or permissible competition within a market. Of course, such an approach again departs from the orthodox ‘objectivity’ of competition assessment and may be criticised on this basis. Yet as a compromise between the two poles of deferring to and simply ignoring the coercive impact of regulation on private market behaviour, focusing on what the regulated defendant itself sought to achieve through its interaction with the regulatory framework arguably provides a reasonable limiting factor.

6. Conclusion

This article began with a simple (perhaps simplistic) dichotomy between competition law, which promotes undistorted competition, and other market regulation, which has a greater or lesser capacity to distort competition in service of alternative policy goals. Yet Ezrachi has criticised the ‘pretence of purity’ that competition law adopts,^[206] and even if antitrust liability professes to be formally unrelated to compliance with other regulatory norms, the presence of regulation can have a profound effect on the task of competition assessment in practice.

Regulation and the diverse public interest values that it represents are a constant feature of modern markets. Context-specific competition enforcement necessarily absorbs and reflects these values to some degree, to the extent that it works within the ‘distortions’ of competition that follow from regulation. In the absence of deliberate efforts by policymakers to make competition impossible, the competition rules rarely cede jurisdiction to other regulatory norms, and whereas competition law may tolerate the presence of regulation that generates market power for certain regulated actors, it tends to be unforgiving of autonomous behaviour that misuses such market power. Yet it is rare for competition assessment to disregard the regulatory context; where such cases do arise, this article has argued for a more nuanced understanding of what ‘normal’ competition looks like in such circumstances.

Yet regulation is not always just ‘part of the scenery’ in competition cases. Efforts to escape or frustrate regulation can constitute standalone antitrust infringements if carried out by an entity falling within the scope of competition law, whether due to dominance or acting in concert. This article has argued that the apparent outsourcing of antitrust liability standards within the non-compliance and regulatory gaming case-law provides an effective vehicle by which to interrogate and better appreciate the inter-relationship between competition law and other regulation. One way to understand these cases is to see the regulatory framework as setting the limits of fair competition in a marketplace, establishing the ‘rules of the game’ that all market actors can reasonably be expected to respect. While simple non-compliance or gaming of regulatory norms is arguably not a competition issue as such, it more plausibly becomes one where such conduct is the direct cause of appreciable anticompetitive effects. Another way to see these cases is as involving the ‘weaponisation’ of regulatory norms to anticompetitive ends as part of a deliberate strategy by regulated firms. From this perspective, the key antitrust consideration is what the defendant intends to achieve through its interaction with the regulatory framework. These perspectives are not mutually exclusive, moreover, and each finds some reflection in the existing case-law on non-compliance and regulatory gaming. Yet in many instances, the true explanation for punishing this conduct through competition enforcement is more pragmatic than principled: using competition law allows us to sidestep limitations or gaps within the regulatory framework itself. These cases thus involve an almost complete convergence between the goals of competition law and other market regulatory norms, and – for better or worse – take the legitimate scope of competition law to its outer limits.

1. * Law School, London School of Economics. Email: n.m.dunne@lse.ac.uk. Thank you to LSE colleagues who gave helpful comments on an earlier draft of this presentation at a staff seminar, in particular Tarun Khaitan and Emmanuel Voyiakis; to two anonymous referees who similarly provided very useful feedback; and to Current Legal Problems editor Deni Mantzari for her advice, support and expert shepherding of this piece through the review process.

   See, eg, Cases C-322/81 Michelin EU:C:1983:313, para 10, C‑42/21 P Baltic Rail EU:C:2023:12, para 86 and C‑252/21 Meta Platforms EU:C:2023:537, para 46. Also recital 11 of Regulation (EU) 2022/1925 on contestable and fair markets in the digital sector (Digital Markets Act) [2022] OJ L265/1. ↑

2. As explained in Section 2, the term ‘regulation’ can be defined as broadly as ‘all forms of economic or social influence’ or as narrowly as ‘a specific set of commands’ imposed by the State: see text accompanying nn 26 and 27. ↑

3. As the General Court recognised in Case T-168 GSK EU:T:2006:265, para 104. ↑

4. See, for example, the OECD’s Competition Assessment Toolkit (2019), which is designed to assist governments in identifying and reducing barriers to competition and other market distortions which may stem from State regulation. ↑

5. Article 7 TFEU states that ‘[t]he Union shall ensure consistency between its policies and activities, taking all of its objectives into account and in accordance with the principle of conferral of powers’. There is, however, limited practice applying the provision: N Nic Shuibhne, ‘Deconstructing and Reconstructing Article 7 TFEU’ in F Ippolito, ME Bartoloni and M Condinanzi (eds), The EU and the Proliferation of Integration Principles under the Lisbon Treaty (Routledge 2018). ↑

6. See eg P Ibáñez Colomo, The New EU Competition Law (Hart Publishing 2023) ch 3. ↑

7. See eg E Fox and E Healey, ‘When the State Harms Competition – The Role for Competition Law’ (2014) 79 Antitrust law Journal 769. ↑

8. See T Khaitan and S Steel, ‘Theorising Areas of Law: A Taxonomy of Special Jurisprudence’ (2022) 28 Legal Theory 325, 329–34. ↑

9. This debate has several dimensions: a normative component, seeking to determine the optimal goals (eg E Deutscher, ‘The Competition–Democracy Nexus Unpacked – Competition Law, Republican Liberty, and Democracy’ (2022) 41 YEL 197); an explicatory aspect, seeking to describe how the competition rules absorb and reflect different policy objectives (eg A Ezrachi, ‘Sponge’ (2017) 5 JAE 49); and an empirical aspect, seeking to identify the goals actually pursued in enforcement practice (see, eg, M Iacovides and K Stylianou, ‘The New Goals of EU Competition Law: Sustainability, Labour Rights, and Privacy’ (2024) 3 European Law Open 587). ↑

10. See n 1. ↑

11. O Odudu, ‘The Meaning of Undertaking within Article 81 EC’ (2005) 7 Cambridge Yearbook of European Legal Studies 211, 212. ↑

12. See Opinion of Advocate General Maduro in Case C-205/03 P FENIN EU:C:2005:666, para 26, on the distinction between ‘the power of the State’, which is subject to ‘democratic control’, and the actions of ‘economic operators’, which fall within the purview of competition law. ↑

13. Article 3(3) TEU. ↑

14. Council Regulation 139/2004 of 20 January 2004 on the control of concentrations between undertakings [2004] OJ L24/1. ↑

15. See eg discussion in Case C-307/18 Generics (UK) EU:C:2020:52, paras 87, 90 and 111 (referring to Article 101) and para 152 (referring to Article 102). ↑

16. Case C-85/86 Hoffmann La Roche EU:C:1979:36, para 91; also use of the term in C‑252/21, Meta Platforms EU:C:2023:537, para 47. ↑

17. European Commission, Guidelines on the application of Article 102 of the Treaty on the Functioning of the European Union to abusive exclusionary conduct by dominant undertakings (hereafter ‘Draft Article 102 Guidelines’), August 2024, paras 47–57. Discussing the role the concept plays within contemporary case-law, see P Ibáñez Colomo, ‘Competition on the Merits’ (2024) 61 CMLRev 387. ↑

18. So, for example, the need to ‘meet competition’ is generally not considered an acceptable defence to prima facie restrictive behaviour, see eg Case T-671/19 Qualcomm EU:T:2024:626, paras 598–99. ↑

19. The principal exception is the prohibition on hard-core cartel behaviour, where contextual analysis is ‘limited to what is strictly necessary’ to understand the anticompetitive nature of the conduct: Case C‑373/14 P Toshiba EU:C:2016:26, para 29. ↑

20. See eg Cases C-228/18 Budapest Bank EU:C:2020:265 and C-413/14 P Intel EU:C:2017:632. ↑

21. Provided for formally by the Article 101(3) TFEU exception, with the possibility of ‘objective justification’ read into the Article 102 TFEU jurisprudence, see eg Case C‑209/10 Post Danmark EU:C:2012:172. ↑

22. For instance, specific competition rules exist for agriculture pursuant to Articles 42 and 43 TFEU. ↑

23. See discussion of the distinction between ex ante and ex post enforcement in Case C-449/21 Towercast EU:C:2023:207. ↑

24. Case C-376/20 P CK Telecoms EU:C:2023:561, paras 84–86. ↑

25. While there is scope to agree to behavioural remedies under the EUMR, this approach is disfavoured, at least in principle: Commission notice on remedies acceptable under Council Regulation 139/2004 [2008] OJ C267/1. ↑

26. R Baldwin, M Cave and M Lodge, Understanding Regulation: Theory, Strategy and Practice (2nd edn, OUP 2011) 3. ↑

27. Regulation need not adopt a classic top-down ‘command and control’ form to come within this understanding, but pure self-regulation by undertakings is treated simply as economic activity within the scope of the competition rules in EU law, see eg the professional rules in Case C-309/00 Wouters EU:C:2002:98 and the industry reorganisation efforts in Case C-209/07 BIDS EU:C:2008:643. For discussion of ‘decentred’ approaches to regulation, see J Black, ‘Decentring Regulation: Understanding the Role of Regulation and Self-Regulation in a “Post-Regulatory” World’ (2001) 54 Current Legal Problems 103. ↑

28. Indeed, viewed from a regulatory theory perspective, competition law can be seen as a variety of regulation that falls within the broad category of ‘market-harnessing controls’ (Baldwin and others (n 26) 114–16). The competition jurisprudence, however, distinguishes the competition rules from ‘other’ regulation that may impact market dynamics, and that is the approach adopted within this piece. ↑

29. Digital Markets Act, Recital 11 (see also cases in n 1). ↑

30. For example, Directive 2019/944 on common rules for the internal market for electricity (OJ L158/125), recital (2), included as its aims: ‘to deliver real choice for all Union final customers, [] new business opportunities, competitive prices, efficient investment signals and higher standards of service, and to contribute to security of supply and sustainability.’ ↑

31. Case C-280/08 P Deutsche Telekom EU:C:2010:603. ↑

32. Opinion in Case Deutsche Telekom EU:C:2010:212, para 21. ↑

33. Case C-280/08 P Deutsche Telekom EU:C:2010:603, paras 80–84; reaffirmed in Case C-220/24 Regia Autonomă Aeroportul Internaţional ‘Avram Iancu’ Cluj v Consiliul Concurenţei EU:C:2025:124, para 27. ↑

34. Indeed, in Case C-382/22 P Cathay Pacific Airways v Commission EU:C:2026:129, para 125, the Court of Justice stated that the State action defence ‘has been only partially accepted’ in its jurisprudence. ↑

35. C-280/08 P Deutsche Telekom EU:C:2010:603, para 81. ↑

36. For a recent example where the State action defence actually succeeded on the facts, see Case T-136/19 Bulgarian Energy Holding EU:T:2023:669. ↑

37. Verizon Communications v Law Offices of Curtis V. Trinko, LLP, 540 US 398 (2003). The complexities of the position within US law are discussed in H Shelanski, ‘The Case for Rebalancing Antitrust and Regulation’ (2011) 109 Michigan Law Review 683. ↑

38. Case T-398/07 Spain v Commission EU:T:2012:173, para 55. ↑

39. Opinion in Case Deutsche Telekom EU:C:2010:212, para 25. ↑

40. In Case C-117/20 bpost EU:C:2022:202, the Court recognised an ‘objective of general interest’ in ensuring the effective application of competition law and other market regulation in parallel, ‘since they are pursuing [] distinct legitimate objectives’ (para 50, emphasis added). ↑

41. See, eg, Cases C-453/99 Courage and Crehan EU:C:2001:465, para 26 and C-74/14 Eturas EU:C:2016:42, paras 35–37. ↑

42. Case C-280/08 P Deutsche Telekom EU:C:2010:603. ↑

43. Case C-209/07 BIDS EU:C:2008:643. ↑

44. As was the case in support of the Siemens/Alstom merger, which was nonetheless prohibited by the Commission (M.8677, 6 February 2019). ↑

45. See Ursula von der Leyen, European Commission Political Guidelines 2024–29: ‘I believe we need a new approach to competition policy, … more supportive of companies scaling up in global markets’. See also Barbara Moens, ‘EU to Relax Merger Rules in Bid to Create ‘European Champions’ Financial Times (16 April 2026). ↑

46. Cases T‑321/05 AstraZeneca EU:T:2010:266, para 356 and T‑814/17 Baltic Rail EU:T:2020:545, para 189. ↑

47. Case C-209/07 BIDS EU:C:2008:643, para 21. ↑

48. Case C‑307/18 Generics (UK) and Others EU:C:2020:52. ↑

49. The Commission may take account of evidence of how the merged entity intends to act on the market going forward, but it places greater emphasis on whether that behaviour is anticipated to cause anticompetitive effects in fact: see eg Case M.11382—AGCO/TRIMBLE/JV (decision of 25.03.2024), for an example where the Commission concluded that distortive post-merger behaviour would not significantly impede effective competition in the circumstances. ↑

50. As the US Supreme Court said in Bell Atlantic v Twombly 550 U.S. 544 (2007), ‘resisting competition is routine market conduct’. ↑

51. This is a core tenet of competition policy, with roots that can be traced back to Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations (1776). For an argument that Smith’s influence should extend beyond this ‘libertarian’ vision, see S Makris, ‘A Smithian Political Economy Approach for the Competition Law of the 21st Century’ (2025) 88 Modern Law Review 712. ↑

52. Article 23, Regulation 1/2003. ↑

53. Commission Guidelines on the method of setting fines imposed pursuant to Article 23(2)(a) of Regulation No 1/2003 [2006] OJ C210/2. ↑

54. C‑457/10 P AstraZeneca EU:C:2012:770, para 132; reaffirmed in Case C-220/24 Aeroportul Cluj EU:C:2025:124, para 28. ↑

55. Cases C-295/12 P Telefónica EU:C:2014:2062, para 133 and C-220/24 Aeroportul Cluj EU:C:2025:124, para 28. ↑

56. Case C-280/08 P Deutsche Telekom EU:C:2010:603, para 90, applying the principle developed in Case C‑344/98 Masterfoods EU:C:2000:689, para 48. ↑

57. On the basis that only the Commission is empowered to find that there has been no breach of Article 102, whereas national regulators can only find breach or discontinue proceedings: Case C-375/09 Tele2 Polska EU:C:2011:270. ↑

58. Case C-280/08 P Deutsche Telekom EU:C:2010:603, para 89. ↑

59. Case C‑457/10 P AstraZeneca EU:C:2012:770, para 132. ↑

60. Case C-238/05 Asnef-Equifax EU:C:2006:734, para 63. ↑

61. Guidelines on the assessment of non-horizontal mergers under the Council Regulation on the control of concentrations between undertakings [2008] OJ C265/6, para 46. ↑

62. M.8124—Microsoft/LinkedIn, 6 December 2016. ↑

63. European Commission, ‘Mergers: Commission Clears Acquisition of Fitbit by Google, Subject to Conditions’ (Press Release, 17 December 2020). ↑

64. Case C-255/22 P Orlen EU:C:2024:790, para 96. ↑

65. ibid, para 96. ↑

66. ibid, para 97. ↑

67. ibid, para 96. ↑

68. ibid, para 95 (emphasis added). ↑

69. To use the language of AG Rantos in his Opinion in Case C-255/22 P Orlen EU:C:2024:466, para 58. ↑

70. See n 9. ↑

71. Cases C-204/00 P etc Aalborg Portland EU:C:2004:6, para 338. ↑

72. Case C-117/20 bpost EU:C:2022:202, paras 28–37. ↑

73. ibid, para 41. ↑

74. ibid, para 49. ↑

75. ibid, para 51. ↑

76. Case C-333/21 Superleague EU:C:2023:1011, para 130 (discussing specifically Article 102 TFEU). ↑

77. ibid, para 166 (discussing specifically Article 101 TFEU). ↑

78. See text accompanying n 33. ↑

79. Case C-280/08 P Deutsche Telekom EU:C:2010:603, para 224. ↑

80. ibid. ↑

81. To quote the defendant’s argument in Case C-280/08 P Deutsche Telekom EU:C:2010:603, para 205. ↑

82. Both the defendant and its rivals used revenues from call services to cross-subsidise their loss-making activity in the provision of line access services. The Commission, however, applied the margin squeeze test only to the defendant’s wholesale and retail-level line access activities. This approach was approved by the Court of Justice, which held that the Commission was entitled to proceed as if tariff rebalancing had occurred as required by EU telecommunications law, even though the German regulator had failed to implement tariff rebalancing for social policy reasons (to protect access to telephone lines for poorer consumers). ↑

83. Cases C‑165/19 P Slovak Telekom EU:C:2021:239, para 57 and C‑42/21 P Baltic Rail EU:C:2023:12, para 88. ↑

84. See Cases C-220/24 DB Station & Service EU:C:2022:832, para 82, T-136/19 BEH EU:T:2023:669, paras 784 and 961, and C-220/24 Aeroportul Cluj EU:C:2025:124, para 31. ↑

85. The Bundeskartellamt applied only domestic competition law, a choice that has been criticised as potentially inconsistent with its obligations under Regulation 1/2003. By the time the case made it to the Court of Justice, the Court spoke in terms of the application of EU competition law. ↑

86. Regulation 2016/679 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data [2016] OJ L119/1. ↑

87. Case C‑252/21 Meta Platforms EU:C:2023:537, paras 44–46. ↑

88. ibid, para 47 (emphasis added). ↑

89. The Court stated that, ‘in the context of the examination of an abuse of a dominant position by an undertaking on a particular market, it may be necessary for the competition authority … also to examine whether that undertaking’s conduct complies with rules other than those relating to competition law, such as’ – but by implication not limited to – ‘the rules on the protection of personal data laid down by the GDPR’: ibid, para 48 (emphasis added). ↑

90. ibid, para 50. ↑

91. ibid, para 51. ↑

92. See text accompanying n 75. ↑

93. Case C‑252/21 Meta Platforms EU:C:2023:537, para 55. ↑

94. As in eg Cases C-205/03 P FENIN EU:C:2006:453 and T-155/04 SELEX EU:T:2006:387, respectively. ↑

95. Case C-359/95 P Ladbroke Racing, para 33. ↑

96. Case C-333/21 Superleague EU:C:2023:1011, para 183. ↑

97. Case C-309/99 Wouters EU:C:2002:98, para 9. ↑

98. See n 94 above. ↑

99. Contrast Case C‑333/21 Superleague EU:C:2023:1011, paras 183–188, where the Court of Justice rejected the application of the Wouters exception to an organisation governed solely by private law (albeit not on that basis expressly). ↑

100. See nn 34 and 35 above. ↑

101. Draft Article 102 Guidelines, paras 167–71. ↑

102. Case C-233/23 Android Auto EU:C:2025:110, para 75. See also Case C‑42/21 P Baltic Rail EU:C:2023:12, which impliedly accepted that health and safety regulatory requirements could provide an objective justification for a de facto refusal to grant access to infrastructure, although the claimed necessity was deemed to be pretextual on the facts. ↑

103. Draft Article 102 Guidelines, para 168. ↑

104. ibid. ↑

105. As in eg Cases C‑42/21 P Baltic Rail EU:C:2023:12 and C‑457/10 P AstraZeneca EU:C:2012:770, respectively. ↑

106. Discussing the impact of privatisation specifically, see Case C‑245/24 Lukoil Bulgaria EU:C:2025:987. ↑

107. As in the E.ON/GDF pipeline cartel (Case COMP/39.401—E.ON/GDF, decision of 8 July 2009). ↑

108. As in Case C-331/21 EDP—Energias de Portugal EU:C:2023:812; in EDP, this fact was treated as evidence of the anticompetitive object of the arrangement contrary to Article 101(1) (see para 102). ↑

109. As acknowledged in Case C‑252/21 Meta Platforms EU:C:2023:537, para 47. ↑

110. See, eg, Cases C-413/14 P Intel EU:C:2017:632 and C-333/21 Superleague EU:C:2023:1011. ↑

111. Case C‑252/21 Meta Platforms EU:C:2023:537, para 47. ↑

112. Case C‑48/22 P Google Shopping EU:C:2024:726, para 224. ↑

113. ibid, para 225 (the Court expressly rejected Google’s argument that the Commission should have conducted a counterfactual analysis to find causation, see para 228). ↑

114. Including in the fields of airports, telecommunications, energy and rail transport. ↑

115. Case C‑165/19 P Slovak Telekom EU:C:2021:239. ↑

116. See, eg, Cases C‑501/06 P etc, GSK EU:C:2009:610, paras 102–04. ↑

117. For a recent study of such missing case-law, see M Heim, ‘The Curious Case of the European Commission’s Missing Antitrust Jurisprudence: Lessons from Abandoned Article 102 Investigations’ (2025) 16 JECLAP 285. ↑

118. Case M.8124—Microsoft/LinkedIn, 6 December 2016, para 177. ↑

119. See n 55. ↑

120. As in Cases 39.525—Telekomunikacja Polska (22 June 2011), C‑165/19 P Slovak Telekom EU:C:2021:239 and C‑42/21 P Baltic Rail EU:C:2023:12. ↑

121. S Dogan and M Lemley, ‘Antitrust Law and Regulatory Gaming’ (2009) 87 Texas Law Review 685. ↑

122. Case COMP/A.37.507/F3—AstraZeneca, 15 June 2005; upheld on appeal in Cases T‑321/05 AstraZeneca EU:T:2010:266 and C-457/10 P AstraZeneca EU:C:2012:770. ↑

123. Case AT.40588—Teva, 31 October 2024. ↑

124. Case AT.40394—Aspen Pharma, 10 February 2021. ↑

125. Case AT.40134—AB InBev Beer Trade Restrictions (29 June 2016). ↑

126. Case C-179/16 F. Hoffmann-La Roche and Others EU:C:2018:25. ↑

127. Case AT.40178—Car Emissions, 8 July 2021. ↑

128. Including the Commission decisions in Cases AT.39226 Lundbeck and AT.39612 Servier and the preliminary ruling in Case C-307/18 Generics (UK) EU:C:2020:52. ↑

129. See eg M Moore & D Tambini (eds), Regulating Big Tech. Policy Response to Digital Dominance (OUP 2022). ↑

130. See eg M Gal and O Aviv, ‘The Competitive Effects of the GDPR’ (2020) 16 JCLE 349 and CB Frey & G Presidente, ‘Privacy Regulation and Firm Performance: Estimating the GDPR Effect Globally’ (2024) 62 Economic Inquiry 1074. ↑

131. Recital 9, GDPR, which states that uneven implementation of the previous EU data protection regime could ‘distort competition’. ↑

132. See n 91. ↑

133. As in Cases C‑457/10 P AstraZeneca EU:C:2012:770 and C-591/16 P Lundbeck EU:C:2021:243, for example. ↑

134. As in the E.ON/GDF gas pipeline cartel, for example (see in particular, the judgment of the General Court on appeal, Case T-370/09 GDF Suez v Commission EU:T:2012:333. ↑

135. Ezrachi (n 9). ↑

136. I Lianos, ‘Polycentric Competition Law’ (2018) 71 Current Legal Problems 161. ↑

137. For a more detailed critique of the adequacy of this approach, see N Dunne, Competition Law and Economics Regulation (CUP 2015) ch 4. ↑

138. In Deutsche Telekom, for instance, the Court dismissed the significant impact of the national regulatory framework on the defendant’s behaviour on the rather blithe ground that it could have sought to petition the regulator to adopt more cost-reflective price regulation (in spite of evidence that the regulator had deliberately chosen to prioritise social solidarity over efficiency in its regulatory policymaking). For further critique of this case from a fairness perspective, see Dunne (n 137) 224–27. ↑

139. This was a significant theme within Deutsche Telekom: in such circumstances, the Court took a generous view of the Commission’s discretion to initiate competition law proceedings against the undertaking or infringement proceedings against the Member State maintaining the anticompetitive regulation. ↑

140. See n 69. ↑

141. As the General Court remarked in Lundbeck, a pay-to-delay case, ‘it is in the public interest to eliminate any obstacle to economic activity which may arise where a patent was granted in error’: Case T-472/13 Lundbeck EU:T:2016:449, paras 119, 390 and 487. This is also an issue in the background in Case C‑457/10 P AstraZeneca EU:C:2012:770 and Case AT.40588—Teva, 31 October 2024 amongst other cases. ↑

142. As provided by Article 4 (independence), Article 5 (resources) and Articles 10 and 13–16 (remedies and fines) of Directive (EU) 2019/1 to empower the competition authorities of the Member States to be more effective enforcers and to ensure the proper functioning of the internal market (ECN+ Directive) [2019] OJ L11/3. ↑

143. See n 74. ↑

144. Case C‑252/21 Meta Platforms, para 54. ↑

145. Case C-117/20 bpost EU:C:2022:202, para 49. ↑

146. See n 67. ↑

147. GDPR, recital 4. ↑

148. See fn. 15. ↑

149. Case C-377/20 Servizio Elettrico Nazionale EU:C:2022:379, para 61. ↑

150. Draft Article 102 Guidelines, para 45. ↑

151. ibid, para 60. ↑

152. For example, data protection rules limit the extent to which firms may profitably combine data-sets containing personal information, while environmental protection rules limit the extent to which firms can use cheaper but more polluting technologies. ↑

153. For example, EU employment law prevents firms from dismissing pregnant employees and requires the provision of paid leave. ↑

154. On regulatory compliance costs generally, see OECD, OECD Regulatory Compliance Cost Assessment Guidance (OECD Publishing 2014). ↑

155. For discussion of the level-playing-field notion generally, see eg M Gillis, ‘Let’s Play?: An Examination of the “Level Playing Field” in EU Free Trade Agreements’ (2021) 55 Journal of World Trade Law 715. ↑

156. Case C‑165/19 P Slovak Telekom EU:C:2021:239, paras58–60, and C‑42/21 P Baltic Rail EU:C:2023:12, para 89. See also Case T-136/19 BEH EU:T:2023:669, para 873. ↑

157. Commission Guidelines on the applicability of Article 101 to horizontal co-operation agreements [2023] OJ C259. ↑

158. ibid, para 520. ↑

159. ibid, para 564. ↑

160. ibid, para 520. ↑

161. C‑252/21 Meta Platforms EU:C:2023:537, para 51. ↑

162. As in Cases C‑457/10 P AstraZeneca EU:C:2012:770 and AT.40588—Teva, 31 October 2024. ↑

163. Although coordination or dominance does not attract antitrust liability itself, the jurisprudence emphasises that such firms have heightened obligations: coordinating firms must ensure that each still ‘determine[s] independently the policy which he intends to adopt on the … market’ (Case C-40/73 Suiker Unie EU:C:1975:174, para 173) while dominant firms have a distinctive ‘special responsibility’ to avoid distorting competition (Case C-322/81 Michelin EU:C:1983:313, para 10). ↑

164. The argument would be that such firms are expected to comply both with their regulatory obligations and to refrain from practices that might reasonably be considered at odds with the broad thrust of a regulatory scheme. An analogy might be drawn to the concept of potential competition, which is established where there are ‘real and concrete possibilities’ of new entry (see eg Case C‑307/18 Generics (UK) EU:C:2020:52, para 36). This standard can be met even if there is considerable ambiguity as to whether new entrants can lawfully compete in a market given existing regulatory barriers to entry. In pay-to-delay cases, for instance, competition law does not tolerate agreements to exclude potential competitors even if there is a reasonable chance that patent law might facilitate this result: for instance, in Case T‑472/13 Lundbeck EU:T:2016:449, paras 121–31, the court found potential competition where there was only a 50–60 per cent chance that the would-be competitor could lawfully enter the market concerned. ↑

165. Coordination falling within Article 101 could also act as a jurisdictional trigger for the application of competition law; a ‘collusion-plus’ approach would then treat coordinated non-compliance or gaming efforts as a restriction of competition. An example of this arose in the Car Emissions cartel, see n 127. ↑

166. The German competition authority also drew on national case-law on consumer protection to support its approach to the concept of market abuse. ↑

167. Ezrachi (n 9). ↑

168. See eg A Ayal, ‘The Market for Bigness: Economic Power and Competition Agencies’ Duty to Curtail It’ (2013) 1 JAE 221; L Khan, ‘The Ideological Roots of America’s Market Power Problem’ (2018) 127 Yale LJF 960; and A Gerbrandy and P Phoa, ‘The Power of Big Tech Corporations as Modern Bigness and a Vocabulary for Shaping Competition Law as Counter-power’ in M Bennett, H Brouwer and R Claassen (eds), Wealth and Power: Philosophical Perspectives (Taylor & Francis 2022). ↑

169. See n 163. ↑

170. Case C-85/86 Hoffmann La Roche EU:C:1979:36, para 91. ↑

171. Cases C‑165/19 P Slovak Telekom EU:C:2021:239, para 57 and C‑42/21 P Baltic Rail EU:C:2023:12, para 88. ↑

172. Case C‑252/21 Meta Platforms EU:C:2023:537, para 44. ↑

173. ‘[P]ublic authorities can legitimately choose complementary legal responses to certain conduct that is harmful to society through different procedures forming a coherent whole so as to address different aspects of the social problem involved’: Case C-117/20 bpost EU:C:2022:202, para 49. ↑

174. Exemplified by Case C-280/08 P Deutsche Telekom EU:C:2010:603. ↑

175. Under the ECN+ Directives, national competition authorities must have the ability to impose behavioural and structural remedies on defendants (Article 10), plus fines of up to at least 10 per cent of the annual worldwide turnover of the undertakings concerned (Article 15) for competition law violations. These provisions mirror the powers of the Commission under Regulation 1/2003. ↑

176. See Directive 2014/104/EU on certain rules governing actions for damages under national law for infringements of the competition law provisions of the Member States and of the European Union [2014] OJ L349/1. ↑

177. See eg Case C-413/14 P Intel EU:C:2017:632, para 138. ↑

178. See eg the discussion in Case C-307/18 Generics (UK) EU:C:2020:52. ↑

179. Surveying the case-law, see P Ibanez Colomo, ‘Anticompetitive Effects in EU Competition Law’ (2020) 17 JCLE 209. ↑

180. This requires it to be demonstrated that coordination ‘has as its actual or potential effect the prevention, restriction or distortion of competition, which must be appreciable’ (Superleague, para 169). Thus it is sufficient that the conduct merely limits potential (as opposed to existing) competition, although this does require it to be established that ‘real and concrete possibilities’ for new market entry exist (Case C‑307/18 Generics (UK) EU:C:2020:52, para 36). ↑

181. Relied upon, inter alia, in Roche & Novartis (n 126), Car Emissions (n 127) and various ‘pay-to-delay’ cases. ↑

182. Case C-333/21 Superleague EU:C:2023:1011, para 165. ↑

183. Case C-377/20 Servizio Elettrico Nazionale EU:C:2022:379, para 50. ↑

184. ibid, para 54. ↑

185. See n 113. ↑

186. See eg Case C-333/21 Superleague EU:C:2023:1011, para 163. ↑

187. See eg discussion in Case C-377/20 Servizio Elettrico Nazionale EU:C:2022:379, paras 53–54. ↑

188. Case C-413/14 P Intel EU:C:2017:632, para 139; followed in Case C-680/20 Unilever EU:C:2023:33. ↑

189. Case C‑48/22 P Google Shopping EU:C:2024:726, para 240. ↑

190. ibid, para 254. ↑

191. Case C-333/21 Superleague EU:C:2023:1011, para 167. ↑

192. ibid. ↑

193. Case C-680/20 Unilever EU:C:2023:33, para 45 (emphasis added). ↑

194. Case C‑48/22 P Google Shopping EU:C:2024:726, para 255. ↑

195. ibid, para 256. ↑

196. Named for Case C-62/86 Akzo v Commission EU:C:1991:286. ↑

197. The Akzo test is an off-shoot of the Areeda-Turner test, which was developed to provide an objective means to determine when low prices – typically considered desirable from a competition policy perspective – should nonetheless be found abusive by virtue of their exclusionary effects (see PE Areeda and DF Turner, ‘Predatory Pricing and Related Practices under Section 2 of the Sherman Act’ (1975) 88 Harvard Law Review 697). The Areeda-Turner test posits that prices below average variable cost (AVC, a proxy for margin cost) should be deemed conclusively abusive on the basis that no rationale firm would continue to produce if it cannot obtain prices at or above AVC, so that its choice to do so can be taken to disclose sufficient evidence of anticompetitive intention. The Akzo case extended the Areeda-Turner test to prices above AVC but below average total cost where there is evidence that such prices are ‘determined as part of a plan for eliminating a competitor’ (C-62/86, para 72), on the basis that such prices have the capacity to exclude as-efficient competitors, while the actual intention evidence serves to counter any potentially innocent explanations for the pricing behaviour. ↑

198. See n 51 and accompanying text. ↑

199. Including AstraZeneca, Teva, Aspen Pharma, Telekom Polska, Lundbeck and the quite fantastical case of Baltic Rail, where the defendant opted to destroy 19km of its own rail infrastructure rather than share it with a freight rail competitor under the EU rail liberalisation framework. ↑

200. See n 46. ↑

201. Case C‑457/10 P AstraZeneca EU:C:2012:770, para 99. ↑

202. Case AT.40588—Teva, 31 October 2024. ↑

203. ibid, para 1033. ↑

204. ibid, para 1033, citing Case T-612/17 Google Shopping EU:T:2021:763, para 257. ↑

205. Opinion in Case C-413/14 P Intel v Commission EU:C:2016:788, para 66. ↑

206. Ezrachi (n 9) 50. ↑