EU Digital Competition Law: The Socio-legal Foundations

Abstract

Competition policy in the EU and UK is in the process of a significant reconfiguration. Its key postulates, methodologies, and normative goals are being subject to intense discussion and revision. The emergence of sui generis ‘new competition tools’ in the area of digital markets—EU Digital Markets Act and UK Digital Markets, Competition and Consumers (bill)—epitomises this trend. The purpose of this Article is to attempt to provide legal theoretical foundations for the new subfield of competition law and policy by systematising and conceptualising these trends into the framework of socio-legal scholarship.

I. Introduction

Digital competition law is an emerging and rapidly evolving field. This Article submits that for succeeding, this ambitious reform of recalibrating the normative and methodological foundations of competition policy has to combine harmoniously and in a unique way the elements of traditional competition law and sector specific regulation. The need for such a new field is usually explained by the inability of the established mechanisms of antitrust law (regulating mainly anticompetitive agreements and unilateral abuses of dominant position ex post) to provide for competition in digital markets effective protection but also promotion.¹ The protagonists of the reform submit that the traditional understanding of the role and function of the law regulating economic competition does not capture all the specificities of the digital economy.² A situation, in which any meaningful regulatory interference into market processes is fully dependent on establishing an infringement (preventative mode of competition law) can only remedy digital markets’ imperfections sporadically, in a fragmentary way and often when the harm from the exit of potential competitors is too late to repair. The normative notion of non-interference in digital markets’ is in decline, and digital competition policy is only a part of a much broader trend of regulatory activism in the area of platform society.³ The paramount importance of these markets for the entire economy, and a generational gap between the dynamics of their development on one hand and legislative responses on the other,⁴ requires a more systemic, coordinated and strategic regulatory engagement (proactive mode of competition law).⁵

Remarkably, the new area of digital competition law is emerging simultaneously in various jurisdictions.⁶ This indicates the objective nature of the processes, a noticeable coherence in understanding the specificities of digital platform regulation,⁷ and a global consensus about its unavoidability. Evidently, the shift from the ‘if’ to the ‘how’ raises a host of challenges. Most of them are beyond the scope of this Article. It aims to provide some basic systemic features of the emerging proactive modality of digital competition law, routing it in socio-legal scholarship.

The emerging regulatory reality is attracting a lot of attention in both academic⁸ and policymaking⁹ literature. The field is on the brink of an unprecedented reconfiguration. The reforms, initiated in the UK (Digital Markets, Competition and Consumers Bill (‘DMCCb’)),¹⁰ EU (Digital Markets Act (‘DMA’)),¹¹ US,¹² as well as in other important jurisdictions will have a global spillover effect. While each of the regulatory models has its own specific features, priorities and instruments, they all are united by several paradigmatic elements.

Among the most important features, characterising the new ex ante regimes for regulating competition in digital markets, are the following three:

1. (1) The asymmetric (in the EU) or even bespoke (in the UK) scope of the addressees of the new rules. The most evident elements of asymmetric/bespoke nature of the rules are present in the way how the obligations (DMA)/conduct requirements (DMCCb) are phrased. The very name of Article 6 of the DMA ‘Obligations susceptible of being further specified’ as well as open-ended substance of obligations themselves imply that the enforcers are granted with unprecedented flexibility of selecting different digital activities of gatekeepers and bringing that activity to an end or modifying it significantly. The bespoke nature of the proposed UK regime is even broader. Suffices it to illustrate a provision of Section 20 DMCCb: ‘[The CMA may impose conduct requirements] if they are for the purpose of preventing a designated undertaking from—…’.¹³ In other words, the enforcer is granted with discretionary competence to design obligations ad hoc aiming at achieving a long list of broadly defined goals. These new rules will be applied only to the biggest undertakings, the digital power of which channels competition downstream as well as horizontally. Those undertakings will be designated with the status of ‘gatekeepers’ (in the EU) and undertakings with ‘strategic market status’ (in the UK). Hereinafter a more general term ‘gatekeepers’ will be used unless a specific provision of the UK regime concerns.

2. (2) In addition to the asymmetric scope of the addressees, the rules themselves are characterised by a quite interventionist nature. The gatekeepers will be assigned an extensive list of obligations, none of which is required from any other digital undertaking, and most of which go far beyond the scope of prohibition of unilateral abuse of dominance as present in the traditional ex-post competition law. The nature of these obligations is very demanding even for the omnipotent digital hyperscalers. This regulatory approach is designed to be so blatantly asymmetric in order to treat currently blatantly asymmetric conditions in the digital markets. The general ambition of these rules is to recalibrate competition in the digital economy. According to the protagonists of the reforms, the current state of competition is systemically weak, and the disagreement is mainly about setting up properly the priorities and mechanisms of the rules. It appears for them that the only workable way for restoring systemic imbalances in digital markets and shaping competition is to introduce a disproportionate regulatory model, imposing strict obligations on the systemic market players, which (the obligations) are expected to provide the markets with a new dynamism. Of course, many remain sceptical as to the very capability of such ambitious, interventionist and technically complex rules to bring meaningful changes to the dynamics of digital markets.

3. (3) Finally, in addition to the asymmetric scope of the addressees and a long list of demanding requirements, the new regulatory approach is characterised by its ‘opacity by design’.¹⁴ The obligations are drafted in either all-inclusive or vague language, full of adjectives, inherently open to contextual interpretation: ‘fair’, ‘contestable’, ‘proportionate’, ‘reasonable’, ‘non-discriminatory’, etc. One of the most plausible explanations for this format is that it may allow the rule enforcers a necessary flexibility and discretion making the focus of intervention more targeted, with some practices receiving higher attention than the others. It would also allow the enforcers to manage their limited resources more efficiently, not spending most of them on continuous procedural formalities, laboriously proving the obvious before doing minimally what is absolutely necessary. As experience derived from various previous attempts to regulate digital markets indicates, the biggest undertakings find it easy to comply with the letter of clear, precise and detailed rules.¹⁵ Most of those earlier regulatory incentives do very little in terms of recalibrating the situation in the digital markets.¹⁶ The ‘opacity by design’ format may become precisely the new ‘experimental’ regulatory approach, which could be more successful than its ‘symmetric’ predecessors.

It should be noted with regard to these features that (1) the asymmetry of the addressees of the new rules; (2) the all-inclusive catalogue of their obligations; as well as (3) the open-ended juristic nature of these obligations implies essentially limitless power granted to enforcers. These new discretionary competences however envisage a selective not a comprehensive enforcement and not with frequent applications of the rules. Unlike the ex post competition policy regime, which is underpinned by the rationale ‘the more anticompetitive practices are deterred, the better the markets off’, the new rules imply so-called ‘smart’, ‘selective’ enforcement with the focus on compliance, not fines. The open-ended obligations and high fines aim to facilitate the job of the enforcers when and if they decide to intervene proactively. Unlike the traditional ex post rules, the emerging regime does not aim to be axiomatic. This is also reflected in the very practices of gatekeepers which are becoming prohibited within the DMA/DMCCb regimes. The nature of these prohibitions is not restorative but rather punitive—the practices are prohibited not because they are inherently harmful but as the ultima ratio regulatory toolkit for bringing the new dynamism to competition in digital markets. In this sense, the new obligations are rather emergency mechanisms than a new perception of normality in digital markets. For this reason, it would be counterproductive for the real protagonists of the DMA/DMCCb to incite the holistic and comprehensive enforcement of all obligations by all gatekeepers. Only selective, targeted intervention can explain the new regulatory mechanics. This of course raises a plethora of questions as to the competence of new enforcement units to apply responsibly and in the right way the new powerful mandate.

The main purpose of this Article is to systematise these and the concomitant elements of the new regulatory paradigm for digital markets into a socio-legal proto-theory of digital competition law. This theory has two dimensions: conceptual and practical. The Article begins with the former, offering a constructive critique of the methodological and normative cornerstone of modern competition law: the theory known as Law & Economics. It should be stressed at the very outset that the criticism targets not the approach as such, but only the categorical way in which it has been often perceived in the contemporary theory of competition law. The purpose in this regard is not to explain why (and not even to claim that) Law & Economics is responsible for the current situation in competition law, but to illustrate how the emerging area of digital competition law is founded on a different theoretical background, which is in many regards a counterpoint to the previously dominant ‘more economic approach’.

Unlike many other contemporary critics of Law & Economics, this Article does not see a solution in structuralist theories—the theories opposed to Law & Economics (such as for example Ordoliberalism or Neo-Brandeisian approach), which focus on the structure of the markets and their normative fairness. The new socio-legal approach to the rules shaping competition in digital markets goes beyond the established ‘efficiency vs. structure’ polarity. Both polarities are characterised by different types of economic and legal formalism, are singular and monistic in their methodology and as such do not embrace the new socio-legal dynamics in the field.

The second half of the Article (Parts III and IV) identifies the distinctive features of the new regulatory approach to competition, systematising them into a socio-legal theory of digital competition law. More specifically, this Article conceptualises the key characteristics of the emerging subdiscipline of digital competition law; analyses its socio-legal and economic roots (Part II); explores the preconditions necessitating its emergence (Part III); and describes its key characteristics; elaborates on the eventual trajectories of its coexistence with the main area of competition law (Part IV). The Article addresses these problems through the prism of the regulatory reforms initiated in the EU and the UK.

II. The Decline of Competition Law and Economics

In one of his emblematic articles published by the Harvard Law Review,¹⁷ Richard Posner, a pioneer and a key figure of the neoclassical Law & Economics approach to competition law, has articulated some shortcomings of the self-referential and self-sufficient perception of law qua legal formalism. Such an approach, in Posner's view, manifests the ‘decline of law as an autonomous discipline’. The criticism of legal exclusivity¹⁸ is built upon the established arguments of socio-legal scholarship, with its ‘law is too important to be left to lawyers’ motto.

Schematically, the main problem of legal formalism was identified in an overambitious claim about law's ability to regulate the whole spectrum of societal relationships using an exclusively legal disciplinary epistemic toolkit. Law claims certainty and uniformity but, in reality, it is often inconsistent and unpredictable—particularly in situations where such consistency and predictability are needed most. ‘We now know that if we give a legal problem to two equally distinguished legal thinkers chosen at random we may get completely incompatible solutions; so evidently we cannot rely on legal knowledge alone to provide definitive solutions to legal problems’.¹⁹ This is not an uncommon criticism. A significant part of Law & Economics movement in competition law was built upon the central idea of reconfiguring our perception of law by using universal microeconomic theory, characterised by the attributes of deterministic scientific certainty and objectivity, underpinned by robust empiricism and mathematical methods. Law's inherently interpretive nature, its relative flexibility, and adaptability to specific societal context, has been portrayed as its systemic flaw. Of course, this purified exposition of Law & Economics in antitrust is provided with a degree of stylisation and hyperbolisation. There always were and there will always be different versions and strands of a more economic approach, and even at the peak of its popularity it was never applied comprehensively. The purpose of this style of representation of the movement is not to provide a caricaturised and simplified vision of Law & Economics but to attempt to identify its key systemic failure. This failure would seldom be present explicitly or centrally in the literature and in the most emblematic pro-economics cases—but its implicit presence in the DNA of the more economic approach was evident in policymaking, in enforcement, in academic discussions, and in case law.

Being historically one of the branches of socio-legal scholarship itself,²⁰ and challenging jointly with legal realists the dominance of old legal formalism, Law & Economics has transformed gradually into a new economic formalism.²¹ In the areas such as competition law, the normative absolutism and methodological exclusivity of legal formalism has been replaced by the normative absolutism and methodological exclusivity of Law & Economics.

Under this stylised version of Law & Economics approach, competition law should be applied indiscriminately and uniformly to all types of economic relationship.²² It is acknowledged that while each sector of economic life has its own specificities and indeed requires its own differentiated treatment, these distinctive elements have to be internalised imperceptibly by the universalistic rationale of competition policy. All specificities are commensurated, reduced to the common denominator by the melting pot of the price theoretical calculus.²³ The sectoral nuances are taken into account, internalised as an addendum to the sophisticated modelling, but they are never decisive, and they rarely overshadow the steering role of the general rules and principles of competition law. The well-known normative autarchy of competition law, its solipsistic, inward-oriented reluctance to enter into a dialogue and to reach a compromise with other legitimate societal values, may be explainable by the conceptual dominance of Law & Economics.

From this perspective, Law & Economics can be seen an approach of neoclassical economics, which is characterised by its simultaneous universalism and its reductionism. It is universal in the sense of being capable of converting any type of social relationship into the cost-benefit calculus. It is reductionist, again, in the sense of being capable of converting any type of social relationship into the cost-benefit calculus.

The microeconomic analysis of competition law underpins its normative propositions by excessively and unnecessarily complex mathematics, aiming (and claiming its ability) to calculate and model the whole spectrum of societal needs with advanced econometrics, empiricism, game theory, and adjacent IO techniques.²⁴ The more complex the relationship, the more sophisticated the maths. The higher the stakes, the more sophisticated the maths. The more factors have to be considered, the more sophisticated the maths. Incomprehensible to most policymakers, formulas and calculations are usually supplemented by a short memo, converting the complex findings into a plain prescriptive policy recommendation. Rejecting such a recommendation implies acting against the deterministic neutrality, the evidence-based impartiality of the objective science—one of the biggest ‘sins’ any government, agency, or court would ever dare to commit.

The universality and the reductionism of Law & Economics is complemented by the third distinctive feature: scientific singularity. It assumes the existence of the right solution, which could be calculated and rationally reached if shaped in robust mathematics. This approach transfers the area of economic competition from the domain of social sciences to the domain of natural sciences. The models are attributed with axiomatic causality and absolute determinism. If the right scientific answer for all spectrum of competition-related problems exists, and if the party offers an advanced mathematical explanation and justification of its position, the probability of winning the case becomes almost a part of a logical syllogism. If two parties confront each other by presenting alternative modelling and interpretation of data, the one with more technical expertise usually has a greater chance of succeeding.²⁵ By definition, the technicality will trump all other factors and arguments if all of these factors are from the very beginning represented in their ‘refined’ calculable incarnation.²⁶ Again, this is a stylised description—it does not aim to offer a comprehensive exposition of the approach, referring rather to the unintended implications of its dogmatic perception.

This trend started in the 1970s and reached its peak in the late 2000s. It has been in gradual decline ever since. The decline, however, is slow and inertial, as for decades, the language of neoclassical economics was essentially the only acceptable methodology. It has also expanded its apparatus from the positive description of facts to the normative domain of the goals of competition law.²⁷ Generations of competition lawyers, economists, and policymakers are trained to look at the phenomenon of economic competition through the reductionist lenses of neoclassical Law & Economics. The whole spectrum of societal values and benefits, which economic competition can generate (and also harm), is being reduced to calculable, price-measuring equations. Such an approach to competition policy has been perfectly mastered, and various views opposing the self-isolation of the discipline, hiding behind the facade of advanced mathematical formulas and excessively technical econometric analysis, were perceived and treated as an eccentric and harmless rudiment, a heterodox addendum to the dominant discourse.

It should be pointed out at this junction that Law & Economics was, is, and will always remain to be the main language of competition law and policy (mainly ex post but to a degree also ex ante)—its universalism trumps all described side effects. This Article does not argue for replacing deterministic language of Law & Economics with open-ended language of public interests. It agues that Law & Economics should not be seen as the only vocabulary of the discipline, and that there are public interests which while potentially capable of being transposed into the terminology and bargaining mechanism of Law & Economics can—and in the ex ante modality indeed should—be also represented in the much less formulaic terminology of socio-legal scholarship.

The historical pedigree of competition policy is different.²⁸ It emerged as a societal response to the uncontrolled growth of economic corporation-superpowers and their fusion with (or the ability to control) the political establishment. The main mechanism of competition law was oriented on the ‘form’ and ‘structure’ rather than the ‘result’ and ‘performance’. The origins as well as the very legitimacy mandate of competition law are societal, not economic.²⁹ The idea that there is a need for rules that control, steer, and protect the competitive process is external to the natural cause of economic events, and is not exhausted by the rationale of neoclassical microeconomics.

The original idea of competition policy is one of maintaining the situation in the markets, which would allow meaningful competition between the players at various levels of the supply chain.³⁰ The structure of the markets, the absence of the monopolies was more important than the actual performance at any given moment. Of course, this is another stylisation—some elements of economic efficiency are always present even in the most deontological, value-oriented competition policy theories. The success of well-structured markets was presumed as inevitable, and the fact that some oligopolistic markets occasionally delivered welfare-enhancing outcomes for consumers was not sufficient to tolerate these practices and this composition of the markets.

Whether the industries with competitive dynamics tend to perform better than those cartelised, monopolised, or controlled by the state is an empirical question and, if the answer is positive, to a large extent it is a matter of contingency. As no absolute deterministic causality between the structure of markets and their performance exists, equally no such absolute deterministic causality exists between the markets’ performance and their legitimacy.³¹ Even more so, reducing the societal value of economic competition to its measurable efficiency outcomes diminishes its constitutional status of being an important economic component of liberal democracy. As with all other constitutional, deontological values, economic competition is protected primarily as a matter of principle, not efficiency.³² In this sense, there are clear parallels between the economic incarnation of the phenomenon of competition and its political (elections) and cultural (free speech, marketplace of ideas) dimensions.³³ From this perspective, competition is a foundation of liberalism and democracy, not necessarily a mere recipe for the optimal allocation of resources.³⁴

The decades preceding the conversion of competition law into a subfield of Law & Economics were also by no means unproblematic. Competition policy, indeed, was designed as and oriented towards the structural (and thus formulaic) aspects of the markets rather than based on their performance. However, the theories underpinning the structuralist approaches to measuring and regulating economic competition were myopic as well.³⁵ Myopic in a different way: too often they were susceptible to legal or economic formalisms. These weaknesses were convincingly demonstrated, disproved and remedied by the Law & Economics literature.

III. The Socio-legal Preconditions for the Emergence of Digital Competition Law

Socio-legal scholarship pleads for a more socially accountable approach to law. It is critical of legal formalism and argues that law as a social phenomenon must be open to interpretation by non-legal disciplines. Economics was one of the main driving forces of this approach. In this regard the earlier referred critical anti-formalism argumentation of Posner may well be seen as socio-legal scholarship. It turned out however that legal formalism of competition law has been replaced by the Law & Economics formalism, and thus the current attempt to articulate the socio-legal foundations of digital competition law is inherently critical of the new economic formalism—it is yet not critical of Law & Economics as a technical language necessary for understanding economic conduct and its implications in the markets.

Competition policy is in search of its new raison d’être.³⁶ The decline of Law & Economics impels the discussion about the return to its structuralist roots. Structuralism, however, is inherently susceptible to the systemic shortcomings explicated so persuasively by its antagonist: Law & Economics. Both approaches are inseparable from their side effects, and, despite their inherent weaknesses, both will remain strategically important for the discipline.

The condition of rediscovering its own mission is not the condition of crisis. Competition law is topical as never before. This topicality, however, does not diminish the need for intra-disciplinary discussions about its new modality. There is a growing feeling that the current constellation of the normative goals and methodological tools of competition law may be expanded by a new theory, going beyond the dichotomy of form- vs. effect-based approach.³⁷ It is very likely that the new modality will be offered through the lens of the most vibrant, dynamic and autonomous area of competition law—the area regulating the digital economy. In many respects, it can be seen as a locomotive of the discipline.

The new subdiscipline of digital competition law is based on the mature intellectual, regulatory, and methodological pillars of general competition law, and will never be separated from its conceptual alma mater. This autonomisation process is not a ‘liberation movement’, aiming to be distant from the metropolis wherever possible and becoming its disciplinary counterpoint. On the contrary, it is an organic development, underpinned by the natural evolution taking place in the area of digital economy regulation. Such a development is enriching and enlightening for the main field of competition law.

The emergence of the new subdiscipline is impelled by a number of factors. Some are external to competition law; others constitute its central parts. The impact of these factors is different, but they all cumulatively drive and shape the new vision of digital competition law. Among the main ones are the following five: (1) the revision of the main normative and methodological parameters of competition law; (2) the decline of the universalistic perception of global competition law, refocusing on national context, specificities and interests; (3) rapid shift of the regulatory vision of Internet governance from hands-off non-interventionism to a neo-pragmatism; (4) the global digital race between the biggest Internet corporations, but also between polities; and (5) systemic features of digital markets characterised by the tendency towards mono-/oligopolisation. All five are conceptualised below.

A. The Crisis of the Ethos of Consumer Welfare

Semantically, the idea of consumer welfare appears to be strongly aligned with a proactive regulatory approach: consumers are usually those with asymmetric information, who are more vulnerable and easier to influence, to nudge. The protection of their rights and welfare should therefore go beyond the purely transactional rationale of market processes. Such a perception of consumers is very different to the one applied in competition law. However, its prima facie distributional ethos has enabled an almost unnoticeable transition from pro-intervention to the self-correction paradigm of competition law.

Under the influence of the Chicago School, the idea of consumer welfare is mainly applied as a shield from intervention, not as a sword of intervention. The logic is stunningly simple: markets work if consumer welfare is increasing; if consumer welfare is not harmed, there is a very little reason for intervention. The structure of the markets is not a value in itself as long as it does not serve the interests of consumers. Only those markets which enhance consumer welfare qualify under this perspective to be labelled as competitive markets.³⁸

Competition law has thereby transformed from the guardian of the markets to a servant of consumer welfare. The focus of the investigation has shifted from proof that infringement is anticompetitive to a demonstration of the evidence of why and how such an infringement contributes to consumer welfare.

Due to the reductionist cost-benefit methodology of Law & Economics, consumer welfare is calculated in prices. If, as a consequence of an anticompetitive conduct, prices decrease or do not increase (or in cases of exclusionary abuses do not go too low aiming thereby at making any matching by smaller competitors impossible), it is very easy to conclude that anticompetitive practice merits an exemption or considered as not anticompetitive at all, as it remains anticompetitive only formally (and thus ‘formalistically’). Given the parameters of the factors, influences on consumer welfare are broad and amorphous, and any except the most drastic anticompetitive practices could be found to be in some way welfare-enhancing—particularly if the stakes are high; and if the methodology of calculation is open to further (and further) increase of technicalities; and if the mathematical findings are consensually assigned with the attributes of axiomatic, infallible science and evidence-supported objectivity.

Such a fundamental metamorphosis of the function of competition policy did not stop at the level of justification and exemption of otherwise anticompetitive conduct. It expanded to the very core of competition policy—its goals and the very definition of economic competition. If the only meaningful objective of competition policy is the protection and promotion of consumer welfare, the practices harming consumer welfare were labelled ‘anticompetitive’ and the practices enhancing consumer welfare ‘procompetitive’. The two very different concepts ‘economic competition’ and ‘consumer welfare’ have been placed under a strict subordination of the former to the latter, becoming essentially synonymised if not fused outright.³⁹

This substantially recalibrated vision of the parameters of permissibility of anticompetitive practices has lasted for decades. The influence of the Chicago School has originated in the US,⁴⁰ reaching its peak in the 1980–1990s,⁴¹ and has gradually expanded to the EU's discussions on the goals⁴² and methods⁴³ of competition law and policy.

It is now being criticised and revised from various perspectives. The tonality of the discussions varies from suggesting incremental refinements⁴⁴ to proposing a radical shift.⁴⁵ The new circumstances impel the revival of the structuralist approaches. Many of the new structuralists are united under the ideological roof of the Neo-Brandeisian School whose representatives play an important role in academia and public policy discussions. The pioneers of the Neo-Brandeisian School are gradually—but remarkably—expanding its impact on the area of legislation⁴⁶ and enforcement.⁴⁷ Often, the ideas of new structuralists are underpinned by an ideological hostility to the digital gatekeepers. The arguments epitomise the idea of the curse of bigness,⁴⁸ bringing thereby to the discussion the nuclear option of splitting up Big Tech. Such ideas are gradually shifting from the domain of academic discussions to the domain of executive⁴⁹ and legislative proposals, in both the EU⁵⁰ and the US.⁵¹ Occasionally, the very constitutional foundations of economic competition are being challenged as well.⁵²

At the same time, another omnipotent rhetorical slogan of innovation appears to be taking over the discussion. The innovation criterion, for example, is being considered as a possible steering element of the UK's new pro-competition regime for digital markets.⁵³ There is a risk, however, that many systemic practices of the undertakings with strategic market status may be seen as very innovative (as they usually are). Furthermore, often by restricting competition in some markets, they create competition in others. The final imperative for intervention should therefore be underpinned by reasons, not necessarily deduced from the eventual pro-welfare, pro-innovation, pro-consumer or pro-growth outcomes. Designing the markets implies a more pragmatic vision. This idea appears to be acknowledged in some regulatory circles.⁵⁴

B. The Decline of the Universalistic Perception of Competition Policy

The universalistic parameters of consumer welfare are the most suitable for the spread of competition policy to other jurisdictions. Being originally a very innovative and by far not a self-evident tool for triggering the competitive process, the standardised and in many respects unified conception of competition qua welfare has been rapidly copy-pasted to over 100 jurisdictions. A tool, requiring precise calibration by each polity, adjusting it to the unique constellation of macroeconomic policies, societal needs, ideological preferences, regulatory priorities, and a myriad of other contextual specificities, has been standardised and universalised. The language of Law & Economics has neither a soul nor accents. It is a very practical Esperanto. The mechanisms for calculating welfare-generating practices are genuinely universal. Such a rationale allows predictability, certainty, expediency, and barrierless entry to local markets: the main attributes of globalisation.

The collapse of the USSR, leading to the distortion of all ideological alternatives to the market-oriented economy, has manifested the ‘end of history’. Competition law has become a global phenomenon. This triumphal spirit did not reign long. A range of new geopolitical and geoeconomic challenges emerged, proving that history never ends. The decade of the end of history was underpinned by the consumer-centred, consumeristic vision of competition. These days, the ethos of welfare-focused globalisation is gradually transforming into the ethos of a more fragmented and more pragmatic polycentrism. The revival and an unprecedented growth of the economies with an increased role of central planning and state control questions the self-ascribed claims of Law & Economics to generate the best welfare outcomes and then to share them fairly with consumers by the relevant distributional mechanisms.

The hybrid economies have used the momentum of the uncritical delight, excessive optimism, and self-proclaimed indispensability of Law & Economics, selectively synthesising the elements and the institutional design of competition policy with their own strategic planning mechanisms. The model, which relies on a hypothesis about self-correcting markets, was tacitly confronted with the model that relies on a combination of markets and planning. Not being under an obligation of transparent accountability about the strategic design of the plans and priorities, and using the euphoric complacency of the counterparts, the latter model was able to exploit the advantages offered by the former. This necessitates a revision of the universalist, non-interventionist, globalised, axiomatic, welfare-generating orthodoxy.

Such a situation has triggered numerous legislative responses,⁵⁵ and the revival of the vocabulary of national interests prevailing over the global vision and an acknowledgement that competition policy cannot function without some instrumentalisation⁵⁶ was just a matter of time.⁵⁷ The universalistic model of consumer welfare can only function if all important jurisdictions adhere to the same principles. If some play by different rules, the central requirement of the model—its universality—becomes much harder to meet. If some economies adhere to the principles of consumer-welfare non-interventionism while others strategically use a more differentiated synthesis of market planning, the universality requirement becomes endangered, primarily harming those applying it unreservedly. Different economies begin playing different ballgames. This situation may explain a significant shift in the UK Government's perception of the function of competition law. It is transforming from the universalistic self-referentiality and insulation from any political impact to the idea that ‘[t]o be truly effective […] government's regulatory and economic policy needs to be supported by strong and effective competition law that is enforced by independent regulators but responds to the strategic needs of the UK's economy’.⁵⁸

C. The Emergence of Digital Pragmatism and Strategic Autonomy Factors

The rise of ideas about political and economic globalisation has coincided with the emergence of the Internet. The digital revolution fitted perfectly into the ‘end of history’ matrix. There was a consensual view that authoritarianism is impossible without the control of information and censorship that an Orwellian reality is only possible in dystopias with limited access to information. The rapid spread of digital technologies was directly correlated with global democratisation. The Internet companies, designing the new online world, were seen not only as commercial entities, opening new channels of communication for societies, but also more broadly as the pioneers of free speech, human rights and other fundamental values of liberalism; they were perceived primarily not in their economic capacity, but as important institutions of the digital democracy. This approach was further reinforced with the glorification of the miracle of self-made garage-style entrepreneurship, and the promotion of a vision of the new economy with impressive social lifts for developers, where most products and services became accessible with one click, and free to consumers. This explains why in the first phase of the development of the digital society most democratic jurisdictions opted for a non-interventionist approach, prohibiting only the most drastic instances of harmful content, radicalism, and piracy. There was a liberal consensus that the societal transfer to digital modality would take place quicker and more easily if the main drivers of the digitisation were offered a green regulatory light.

The decline of the end of history euphoria has coincided with the decline of the period of Internet Optimism. On a political level, it has become evident that many authoritarian regimes have adapted to the digital revolution with remarkable efficiency, transforming the power of the new technologies to their own benefit and not losing control over the minds of their people. In addition, the pervasiveness of digital technologies in all societies has triggered a number of new systemic challenges. The perception of the biggest online companies is transforming from one of heralds of freedom to new systemic oligopolies, capable of controlling, steering, designing, and exploiting everyone's digital life. The globalised period of Internet Optimism is being replaced by the more individualised period of Internet Pragmatism.

Public authorities are becoming more vigilant about the consequences of the pervasive digital transformation. The new wave of legislative initiatives aims to recalibrate the permissive and encouraging approach to Big Tech, trying to develop a new modality of their relationship with the public authorities, competitors, business- and end-users. The reforms include all sectors of the digital society, and the economies relying on the democratic format of governance are in search of a delicate balance between the Scylla of the effectiveness of the regulatory measures and the Charybdis of their compliance with the principles of liberalism.

D. The Pivotal Role of Digital Markets in Shaping the Economy and Society

The next feature, explaining the decline of universalistic competition policy and the emergence of autonomous digital competition law, refers to the reassessment of the role that digital technology plays in nations’ strategic development.⁵⁹ It is often portrayed by using the terminology of the 4^th Industrial Revolution.⁶⁰ The central idea of this vision is that the systemic transformation, and the totality of control and influence that digital technology is capable of reaching, makes it the central element of national security and strategic macroeconomic development. The success of the polities in implementing these mechanisms in all public and private sectors would define the role of that polity in the global division of labour.⁶¹ This explains the ubiquitous comparison of the role of data today with the role of natural resources in the period of the 3^rd Industrial Revolution, and the role of data controllers with the role of transport and infrastructure.

Discussions about the 4^th Industrial Revolution are less concerned with the vertical relationship between public authorities and big digital companies, and more with the horizontal relationship between the polities themselves. Due to the inherently strategic status of these interests, these relationships are usually perceived as a zero-sum game. There could be only ‘x’ number of polities capable of designing digital societies. The race to become global and regional leaders, or at least of being able to control and steer their own sovereign space, is fierce, as the position each polity takes right now is very likely to be decisive for the place of this polity in the global division of digital labour over the coming decades.

Given the remarkable ability of those controlling digital users to scale up by synergising data and modelling the behavioural patterns of one group of users over the others, and given the totality of the Internet of (Every-)Things,⁶² the real digital power will only be accumulated in the hands/algorithms of the very few biggest players. Such a trend is becoming even more categorical with the gradual transition of the user's habits from display- to voice-search, which usually offers only one of a variety of available search results, and with the exponential increase of the data generated by various digital assistants and similar IoT and AI LLM technologies. Some societies reasonably believe that they give to the new economy more in terms of big data, active users and scientific ideas than they take out of it, and the fact that consumers in these societies feel satisfied with the status quo is not decisive in shaping the real interests of the polities aiming to reach a stronger position in the global digital race.

As with the previous feature, the EU and UK face a delicate ideological labyrinth between pragmatically and systematically pursuing its strategic interests on the one hand and maintaining a level playing field on the other. Paying insufficient attention to the former would likely make the strategic position of European polities less advanced than it is expected to be. Compromising on the latter would endanger the very ideological and cultural DNA of both European polities.

E. Inner Attributes of the Digital Economy

The last important factor explaining the emergence of digital competition law concerns the distinctive features of the digital economy on the micro- rather than macro-level. As widely reported and observed,⁶³ competition in the digital markets is susceptible to being tipped as the pervasive network effects encourage single homing. Users move to the platform capable of connecting them with other users. Such platforms are capable of synergising and synthesising the data generated in one segment with all others, satisfying and designing the user experience in its totality. This leads to the phenomenon of competition for the market rather than competition in the markets, provoking a winner-takes-all situation.

In addition, traditional competition rules address individual conduct ex post. By definition, they are not capable of offering systemic remedies to market failures.⁶⁴ Even those case-specific remedies,⁶⁵ adopted within the traditional competition law enforcement are often sporadic and inefficient.⁶⁶ As the new rules aim at complementing, and not replacing, the existing mechanisms, the justification for their adoption is inherently lower. The existing model does not have to be proven as broken to legitimise an improvement. The improvements are particularly required for unilateral conduct (be it within the ambit of abuse of dominant position or mergers).

The biggest platforms are the main beneficiaries of this systemic feature of the digital markets, not least because they managed to reach their strategic market status in the much more permissive regulatory era of Internet Optimism. Potential newcomers to the entrenched digital markets face not only the need to match the quality of their services to those with the highest capitalisation in the world, but also the requirement to do so on a sustained basis within a much more data-mindful and proactive regulatory environment and in a way that would comply with a long list of legislative limitations developed in the emerging age of Internet Pragmatism.

Paradoxically, the current incumbents, while being the main addressees of the new legislative initiatives aiming to tame their pervasive growth, may suffer less than the newcomers; not least because they have reached a position that is much easier to protect than to get to, and because they have reached such a strategic position using the mechanisms of markets and thus by being objectively the best in terms of offering the experience needed by users (whether the experience is organic or shaped by these companies is a secondary question). The unsinkability of the gatekeepers is also explainable by their exclusive knowledge and strategic understanding of the functioning of the technology. They—not the public authorities—are the masters of the digital reality.⁶⁷ This explains their remarkable ability to convert any legislation to their benefit,⁶⁸ using the powerful rhetoric of consumer welfare as a pretext for their own strategic planning.⁶⁹

This feature of the gatekeepers may also explain the reluctance of the Commission to encompass in the DMA any form of efficiency defence. Having their products and services be so diverse and cross-fertilising, proving an efficiency of otherwise prohibited practice, may become a matter of a technical skill. After all, gatekeepers are by definition efficient. Additionally, from a procedural perspective, allowing an efficiency defence may undermine the consistency of the DMA and its speedy enforceability. On the other hand, the supporters of the efficiency defence in the DMA argue that such option should be available at least in principle, submitting that ‘[t]he rules from EU competition law can […] also be used to develop an efficiency defence in the DMA. […] The undertakings would remain bound by the rules of conduct in Articles 5 and 6 of the DMA until a decision is taken on an exemption’.⁷⁰ It remains unclear how the efficiency defence may be introduced without a spillover effect on the need for introducing the mechanism of market definition, necessitating the introduction of other mechanisms of ex post competition rules. It appears that the proposal is intentionally drafted in a terminology that avoids any meaningful terminological overlaps with the existing competition rules. Such semantical and syntactical overlaps would unnecessarily embed the new rules into the existing framework—with all its advantages, but also with numerous limitations. Remarkably, the UK equivalent of the DMA—the DMCCb does envisage mandatory efficiency defence for fairness—(Sec 29 of the Bill) (though not contestability— (Secs 44–54 of the Bill)) related obligations. This appears to be a systemic flaw of the DMCCb⁷¹ inasmuch as it would allow for exempting—or delaying—pretty much all fairness-related obligations as all of them while being an obstacle to competition in digital markets in some sense are simultaneously efficient in the other.

IV. The Main Regulatory Features of Digital Competition Law

However rapidly the legislators react to the changing reality, and however future-proof the new rules become, it is systemically impossible to eliminate the gap between the undertakings trend-setters and the enforcers without redesigning the very format of the rules. Such a redesign was attempted by the Commission with the introduction in December 2020 of the Digital Markets Act proposal—an essential part of digital competition law. After intense discussions the Regulation has been finalised. This Part discusses the elements of the Act, which would allow for the asymmetric discretional implementation of the new digital enforcement modality.

Among its distinctive characteristics of the are the following three: (1) complementing the protective function of competition law with a more robust proactive, procompetitive dimension;⁷² (2) asymmetric (and in some instances bespoke) selection of the addressees of the new rules;⁷³ and (3) dialogical and individualised approach to defining the scope of obligations of the asymmetrically selected addressees; future-proof, opacity-by-design open-endedness of the obligations. The following three subsections conceptualise of all these aspects.

A. From Market Protection to Market Design

The revival of a more interpretive approach to law, the debunking of the myopic belief that any uncertainty in and of law is its flaw, leads to another specificity of digital competition model, namely the search of its conceptual foundations and legitimacy in common law. This trend is visible both in theory⁷⁴ and in the UK Government's current public policy initiatives.⁷⁵ The gist of this interpretive turn is about delegating ‘greater flexibility to regulators to put the principles of agile regulation into practice, allowing more to be done through decisions, guidance and rules rather than legislation’.⁷⁶ This implies a greater discretion on the part of ‘the UK's regulatory bodies, removing many of the detailed rules in the existing statutory frameworks to make them less prescriptive (replacing them with outcomes to be achieved), and allowing the regulatory regime to be shaped more by case law’,⁷⁷ acknowledging a greater role of interpretation, relative indeterminacy and regulatory experimentation as well as recognising the higher probability of regulators losing cases in courts. The law is becoming crystallised in adversarial judicial proceedings, and regulators should not be discouraged from taking an action simply because there is a high possibility of their decision being annulled in a courtroom. One of the main mechanisms for such recalibrated competition policy would be market inquiries. ‘Government is concerned that the market inquiry process, and market investigations in particular, are overly cumbersome and significantly underused’,⁷⁸ and thus a greater engagement in various forms of regulatory dialogue and a tendency toward the individualisation of obligations.

This explains another important feature of digital competition law: an extension of its function of being mainly protective to becoming a harmonious combination of both protective and proactive. The idea that competition is only self-discovering and should only be protected—never proactively shaped or promoted—is deeply embedded in the normative non-interventionist narrative. Under such a vision, markets will always self-correct and self-improve, unless undertakings conspire, or abuse their dominant position, or merge, or receive State aid. For these situations the traditional protective competition toolkit is mostly sufficient. Only for some rare cases of systemic market failures may a sectoral intervention be acceptable.

The phenomenon of digital competition is not exhausted and cannot be fully embraced by this narrow, technical perception of ex post competition policy, particularly if its main mechanisms are interpreted in a non-interventionist way of market self-correction.

It is in the sovereign power of countries to shape and promote specific features of the digital markets which are in the interests of the people, businesses and industries of those polities—even if some elements of such a proactive intervention may not be in full conformity with the categorically axiomatic, scientifically absolutist claims of the mathematised wisdom of Law & Economics analysis. The institutional traditions of liberal democracy, in which the EU and UK are deeply embedded, will always serve as a reliable and sufficient safeguard against mimicry of the proactive design of competitive markets into a dirigistic protectionism.

B. Asymmetric/Bespoke Designation of Addressees of the Rules

Digital markets are inherently susceptible to tipping and network effects; they are almost always mono- or oligopolistic. Strategic market status is usually achieved by the remarkable ability of the gatekeepers to tailor individual content to end users, their remarkable skills of matching producers with consumers, suppliers with distributors, advertisers with publishers. Such an individualised approach may serve as a suitable metaphor (not a legal ground but an ironic symbolism) for the transposition of such ‘bespokeness’ from being a feature of the digital markets to being a feature of regulatory response: the new modality proposes fighting fire with fire by assigning to the enforcers unprecedented competences of intervening in market processes.

The DMA proposal embodies this trend. It establishes a list of three cumulative criteria for designating gatekeepers. The model is binary. The obligations are applied in their entirety only to those digital undertakings which meet all three criteria. Going below any of the criteria implies the full non-applicability of the DMA to that undertaking. The criteria have their qualitative—Article 3(1)(a)(b)(c) DMA—and quantitative—Article 3(2)(a)(b)(c) DMA—dimensions. Under Article 3(2) DMA, quantitative thresholds are presumptions, and as such can supposedly be adjusted by the Commission ad hoc without formal amendment to the Act and as long as the concrete criteria continue meeting qualitative aspects of designation.

Qualitatively, the first element of designating gatekeepers requires an undertaking to have a significant impact in the internal market. Its quantitative aspect has two possible avenues. Both are two-step requirements. The undertaking has to achieve ‘an annual EEA turnover equal to or above EUR 7.5 billion in the last three financial years’⁷⁹ and provide a core platform service in three or more Member States. Alternatively, the undertaking has to have market capitalisation of EUR 75 billion in the last financial year, and again provide a core platform service in three or more Member States.

The second requirement concerns the operation of a core platform service (‘CPS’) ‘which serves as an important gateway for business users to reach end users’.⁸⁰ The proposal envisages a number of such services,⁸¹ with many commentators suggesting an extension of the list to other important services (eg browsers or streaming). Under Article 3(2)(b) DMA, the requirement of a CPS as important gateway for business users to reach end users is satisfied where the CPS is provided for more than 45 million active end users per month and 10 million business users per year.

Under the third threshold, the CPS provider must either enjoy or be in the position of potentially enjoying ‘an entrenched and durable position in its operations’.⁸² Quantitatively, this would be the case if the undertaking has met the previous quantitative parameter in each of the last three years.

These three criteria allow only the selection of the most important incumbents in the digital markets. The first intake of designated gatekeepers confirms this statement. The selected gatekeepers are then regulated in a way that differs paradigmatically from the one applied to their competitors, customers and consumers. The structure and the substance of these obligations are discussed under the last Section of this Part. It is important at this juncture to look further at the specificities of the process of designating gatekeepers.

Defining the precise quantitative scope of the criteria plays a pivotal role in the discussions about the DMA. Setting the thresholds too high would on the one hand make the eventual new entries more plausible as the challengers would remain outside the scope of the DMA for longer, permitting them more time for scaling up. On the other hand, this may put below the radar those CPS providers who play a systemic gatekeeping role in some of the digital markets. Such excessively high thresholds may also encourage a tactical move by an incumbent to decrease some of its operations in order to fly below the DMA's radar.

Setting the quantitative thresholds too low would include in the category of gatekeepers too many undertakings who are essentially challengers to the biggest market players rather than being themselves the biggest and most strategically important.⁸³ On the other hand, this would make it impossible for the biggest players to attempt to sink below the DMA radar.

As the relationship between over- and under-qualification is antithetical, it appears to be impossible to completely avoid the side effects of any selected option. When quantitative borders are delineated, borderline cases are unavoidable. The question is less about the theory and the format, and more about the practice and concrete numbers. This explains why the Commission is permitted to fine-tune the precise quantitative (but not qualitative) parameters if the circumstances require.

The DMA contains another important safeguard for protecting the position of new entries. It concerns the requirement to have an EEA annual turnover of about EUR 7.5 billion for at least three years (an increase in market capitalisation above EUR 75 billion waives the three-year window, as it may indicate that the market sees a greater in the new entrant to play a significant role in the digital economy). The same three-year period is applied to the second quantitative condition concerning the number of business- and end-users. Such a period allows an opportunity window to the newcomers after they reach a size that requires them to become subject to the DMA.

The discussed UK model goes in some sense even further.⁸⁴ The process of designation of the addressees of the new regime (undertakings with strategic market status or uSMSs) the CMA is expected to undertake several procedural steps. Unlike the DMA, the bill does not identify core platform services, liberating thereby the CMA from unnecessary, lengthy, and thorny process of drawing a clear line between different core platform services and giving it a competence to designate the uSMS regarding any digital activity provided the qualifying conditions are met. This is a very welcoming and commendable approach.

The bill is also more flexible in respect of defining the jurisdictional thresholds of users: instead of identifying exact number of business and end users in the relevant jurisdiction, which presumably meets the second designation requirement of the DMA, it merely requires the presence of a significant number of UK users or the potential uSMS must carry on business in the UK in relation to that digital activity or the relevant digital activity is likely ‘to have an immediate, substantial and foreseeable effect on trade in the United Kingdom’.⁸⁵ This criterion is also more flexible than its DMA counterpart.

The requirement of ‘strategic significance’ is wide and flexible enough for being adopted without any changes. It is good for the provision to be met if only one of the four requirements is met—with a possible caveat that the requirement of Sec 6(1)(a) is tautological and could be re-drafted: ‘(a) the undertaking has achieved a position of significant size or scale in respect of the digital activity; (b) a significant number of other undertakings use the digital activity as carried out by the undertaking in carrying on their business; (c) the undertaking's position in respect of the digital activity would allow it to extend its market power to a range of other activities; (d) the undertaking's position in respect of the digital activity allows it to determine or substantially influence the ways in which other undertakings conduct themselves, in respect of the digital activity or otherwise’.⁸⁶

Finally, the turnover requirements are equally satisfactory: £25 billion of global annual or ‘the total value of the UK turnover of an undertaking or, where the undertaking is part of a group, the UK turnover of that group in the relevant period exceeds £1 billion’.⁸⁷

The bill offers a very prudent way of imposing requirement upon the designated uSMSs. The DMA had opted for a wide scope of loosely definable obligations with a majority of them being categorised as ‘obligations susceptible of being further specified’ – this formula allows to calibrate the scope of the obligations to specific business model of each gatekeeper, as it was widely agreed that the future-proofed structure of the obligations is the only meaningful way to keep them relevant.

The DMCCb went—prudently—even further, allowing the CMA to tailor obligations which would be either active (mandating the uSMS to act—eg to ‘trade on fair and reasonable terms’⁸⁸) or passive (mandating the uSMS to refrain from action—eg not to use data unfairly).⁸⁹ This is a very elegant and efficient formula granting the DMA the necessary discretion.

Another type of obligations which the CMA may impose on the uSMS in addition to conduct requirements is the mechanism of pro-competition intervention. The idea of the need for a new pro-competition regime underpins the very rationale of the reform. It is its core, its quintessence. The logic is that contrary and in addition to the existing ex-post rules, which are designed to penalise an anticompetitive conduct and by penalising and remedying the harm to restore competition, the new ‘pro-competition’ modality goes further than this—not only restoring but actually creating and shaping competition in digital markets. The idea being that a mere restoration is by far insufficient to remedy systemic failures in digital markets.

The current version of the pro-competition interventions (PCIs) is characterised by three remarkable advantages in comparison to the DMA obligations and one significant shortcoming.

The first advantage is its open-textured nature, allowing the CMA to design concrete parameters of the pro-competition order following a PCI depending on the specificity of each digital market and on the objectives the authority aims to achieve.

The second advantage is that the CMA will be allowed to make recommendations ‘to any person exercising functions of a public nature about steps which the CMA considers the person ought to take in respect of the designated undertaking or the digital activity, or otherwise’. This is the best example of the vertical and horizontal regulatory dialogue of the CMA with the Government and its specific parts as well as with other digital regulators. The CMA may only make recommendations, and it would be ultra vires to assign the CMA with a more interventionist mandate. This instrument of soft power may be seen as a successful example of the regulatory forum – particularly, if other digital regulators, where necessary, would be provided with a similar option in the relevant legislation.

The third advantage is that contrary to the mechanism of conduct requirements, the PCI is not circumscribed by the mandatory efficiency defence: ‘In considering whether to make a PCI, and the form and content of any PCI, the CMA may have regard to any benefits to UK users or UK customers that the CMA considers have resulted, or may be expected to result, from a factor or combination of factors that is having an adverse effect on competition’.⁹⁰ This approach mutatis mutandis has to be transposed to Section 29 dealing with countervailing benefits exemptions for conduct requirements.

Another remarkable indicator of the CMA discretion is the mechanism of pro-competition interventions (‘PCI’). The original idea of the pro-competition interventions (called back then ‘pro-competitive’ interventions) was offered in the Government ‘New pro-competition regime for digital markets’ public consultation, shaping the conceptual contours of the DMCC Bill.

Chapter 4 of the Bill is dedicated to this mechanism. The rationale for pro-competition intervention is very welcoming. As the semantic of the term suggests, it aims to promote rather than merely restore competition in digital markets—evidently, this is the only possible way of recalibrating the dynamics of the UK digital economy.

The Bill covers this instrument in the following way: ‘the CMA may make a pro-competition intervention after conducting a PCI investigation. The PCI investigation may be launched when the CMA ‘has reasonable grounds to consider that a […] relevant digital activity may be having an adverse effect on competition’ and that ‘making the PCI would be likely to contribute to, or otherwise be of use for the purpose of, remedying, mitigating or preventing the adverse effect on competition’.

A fundamentally important element of the PCI is the mandate of the CMA to issue ‘an order imposing on the designated undertaking requirements as to how the undertaking must conduct itself, in relation to the relevant digital activity’ and/or ‘recommendations […] to any person exercising functions of a public nature about steps which the CMA considers the person ought to take in respect of the designated undertaking or the digital activity, or otherwise’.

Both provisions of Section 44(3) are quintessential for the Bill. They are presented in sufficiently wide and flexible scope, allowing the CMA to design the remedies depending on the factual circumstances.

The remaining subsections of Section 44 concretise the substance of a PCI, specifying that it may ‘include provision for the purposes of remedying, mitigating or preventing any detrimental effect on UK users or UK customers that the CMA considers has resulted, or may be expected to result, from the adverse effect on competition to which the PCI relates’, and explaining that a factor ‘or combination of factors relating to a digital activity has an adverse effect on competition where it prevents, restricts or distorts competition in connection with the relevant digital activity in the United Kingdom’.

While being very welcoming in principle, it is important to make the semantic concretisation: the rationale of pro-competition intervention should not be seen exclusively in the restorative terms. The current version of the relevant provisions of the DMCCb does have the tendency of interpreting the logic of pro-competition intervention exclusively in the restorative terms. This significantly restricts the potential of the toolkit and makes the overall regulatory initiative of the UK much less ambitious.

The very term ‘pro-competition’ refers not only to restoring competition, which is harmed by the relevant undertaking with SMS. The idea is mainly—or at least it is also—about promoting competition, triggering it where possible in those segments of digital markets, where it is missing most. The restorative function may indirectly and incidentally contribute to the promotion of competition, but it neither exhausts nor indeed directly aims at this (still very underexplored) dimension of competition policy.

Despite offering different quantitative formulas and criteria, all of the proposals indicate a common philosophy: an asymmetric economic power requires an asymmetric regulatory response. Such an asymmetry is not exhausted by designating the addressees. It is also present in the obligations as such.

C. Opacity-by-Design, Individualisation, and Open-Endedness of Obligations

The last important feature of digital competition law complements the asymmetric approach to selecting its addressees with the open-ended, interpretative nature of the obligations as such. The concrete scope of many of these obligations is supposed to be calibrated in the course of the regulatory dialogue between the enforcer and the incumbent. The very idea of selecting specific undertakings to be the subject of ex ante obligations is not new. Many sectoral rules impose stricter, quasi-public utility duties on the incumbents in specific industries. Usually, the fact of being subject to strict ex ante rules does not exempt the incumbent from compliance with ex post competition law. However, these extended obligations are always mitigated by a very important feature: their clarity and unambiguity. While ex post competition law is general, principled and leaves room for ad hoc interpretation, the ex ante regulations are usually clear and precise. This is explained by the fact that the market failures it aims to remedy are systemic and not attributed to any infringement of law.

This correlation between ex ante precision and ex post generality is being substantially recalibrated by the DMA. The ex ante obligations of the gatekeepers are made intentionally wide and open to interpretation. The purposive nature of this opacity-by-design is explained by the fact that 11 out of 19 DMA obligations are united in Article 6 DMA under the umbrella-term ‘obligations for gatekeepers susceptible of being further specified’. They are unclear, as they are not designed to be clear in the first place.⁹¹ Their precise scope is intentionally supposed to be calibrated in the course of the regulatory dialogue between the Commission and each individual gatekeeper.

In addition to the selectivity of the addressees and the open-endedness of the obligations, another remarkable feature of the DMA is that the liability for non-compliance with the provisions of Articles 6 and 7 DMA may begin from the moment of reaching a compromise between the Commission and the gatekeeper (later), or from the moment of designation of the undertaking as gatekeeper (sooner). The liability, in other words, begin ex nunc if the regulatory dialogue is successful, or ex tunc if it is not.⁹² The arbiter defining the success of the dialogue is a participant in the dialogue—the Commission. This is an instance of juristic innovation. The formula requires the gatekeeper to comply with a very broad list of very vague obligations. It allows calibration via the regulatory dialogue but fully leaves the determination of the dialogue's outcomes to the Commission's discretion.

Remarkably, the Draft IMCO Report proposes not to apply the mechanism of commitments suggested in the original version of the proposal.⁹³ The most plausible reason for this may be the fact that most of the obligations would become the subject of a regulatory dialogue, which essentially overtakes and expands the function envisaged for the mechanism of commitments, making the latter redundant. On the other hand, there are tactical differences between these two administrative procedures, and reserving both options for the enforcer does not appear to bring about any meaningful shortcomings.

There may be a reason behind extending the mechanism of regulatory dialogue to all obligations of gatekeepers.⁹⁴ The format of the regulatory dialogue appears to be more suitable than the mechanism of commitments, as the latter only allows remedying the harm, caused by an individual instance of non-compliance. The scope of the regulatory dialogue may be broader and more systemic.

Essentially, the rules are designed as a two-tiered responsibility. The second tier concerns the liability for non-compliance. Because the rules are open-ended and very demanding, because they will be applied to a very narrow group of undertakings, and because the definition of the precise scope of the rules is assigned to the enforcer, the very compliance with the rules may essentially be seen as the first tier of responsibility. The mission of the rules is not exhausted by the positive implications for business- and end users. The mission is broader, and it may also be seen as, if not as punitive, at least as constraining the uncontrolled growth of the economic, societal and political power of the digital gatekeepers.

The most obvious reason explaining and justifying such a juristic formula is that the digital economy evolves very quickly, and the rules aiming at regulating it effectively must be open enough to remain relevant for a longer time. These rules can only succeed if they are designed to be future-proof. Furthermore, the model can only work if the limited enforcement resources are not spent on endless litigations on the procedural legal technicalities of the trivial formalities.⁹⁵ This is the reason why the liability can begin ex tunc, and why the DMA does not allow for the mechanism of objective justifications, which would establish a list of precedents, limiting the enforcer's flexibility in future cases.

The final explanation of such asymmetric individualisation is related to the level of stakes being strategically played in the digital economy. The markets shaped at this stage will influence and determine the digital society for decades to come. This is another reason why the EU, while being primarily focused on the protection and promotion of intra-platform competition and consumer sovereignty,⁹⁶ also closely monitors the situation in the non-ecosystem inter-platform competition. The designing of such pro-enforcer rules does not prove with a deterministic certainty the strategic decision of the EU to shape the non-ecosystem inter-platform competition. Success in pursuing such a policy, however, would only be possible with the presence of such strong pro-enforcer rules.

V. Conclusion

Competition law experiences ‘a Copernican transformation’.⁹⁷ While never been ‘a tool for the optimization/fine-tuning of market outcomes in light of more or less idealised benchmarks’, it is now expected to begin playing this very role.⁹⁸ This is a significant momentum for the discipline. Our society faces a number of cacophonic challenges and a number of polyphonic opportunities in the area of digital markets, and the emerging digital competition law is designated with an important function in this regard.

Shaping the digital markets in a way that mitigates the challenges and reinforces the opportunities is a duty of responsible competition governance. The reform proposes a radical transformation of the discipline, going far beyond the letter of its substantive provisions. It manifests and embodies an important trend, paradigmatically changing the digital universe as well as the very discipline of competition law.

The transition from mono- to polycentricity of competition policy can be seen by some as a disciplinary crisis. However, even if seen as a crisis, it concerns only the epistemic reassessment of the methodological, normative and political foundations of competition. The situation is a challenge rather than a crisis, an opportunity rather than an obstacle. In terms of its impact, competition policy is topical as never before.

Economic competition as an inherently societal phenomenon cannot be reduced to single metrics. Abandoning the holistic and absolute reliance on the axiomatic mathematical and normative determinism of competition metric, makes the policy more context-specific and more context-adaptable. Each polity designs some elements of competition in its various markets and may wish giving priority to one shade over the other. This preference may be strategic or tactical, long- or short-term, clear or implicit. In all cases it derives not from the mathematical formulas but from societal interests.

Different methodologies and different ideologies will continue competing for being selected and prioritised by the decisionmakers.⁹⁹ Paraphrasing a famous socio-legal metaphor, the transition from mono- to polycentricity implies only that ‘competition law is too important to be left to competition lawyers and economists’ (and even more so to law-&-economists). It has a greater societal dimension and impact. The DMA epitomises this trend, shaping a new socio-legal modality of EU digital competition law.

The pendulum of competition law is moving in the direction of openness and polycentricity. Many jurisdictions show examples of successfully tackling the problems of the digital economy with traditional competition tools. The availability of new tools would make their task logistically easier and more meaningful in a broader strategic sense. It remains to be seen however if these initiatives will manage to redesign digital markets, making them more pro-competitive; only contribute to incremental changes and improvements; or even prove to be ineffective, not succeeding or succeeding very little in bridging the gap between gatekeepers and all other digital undertakings.