Responsive competition law

Abstract

This article applies the lessons from the prior theory of responsive regulation in criminology to EU competition law and extends these lessons to argue in favour of an enhanced form of responsive competition law. First, it finds that EU competition law enforcement is already responsive in the traditional sense as it takes the reactions of undertakings into account when deciding which instrument to apply, in accordance with the enforcement pyramid developed by Braithwaite. An enforcement pyramid for EU competition law is presented. The objectives of competition law are found to be broad, and its key norms are open, facilitating responsiveness. This also allows competition law to develop to meet new societal demands, such as the need to control market power in the digital realm and to combat climate change. Next, the article examines the role of responsive and accountable behaviour by undertakings in competition law. First, it is found that in line with new forms of regulation concerning non-financial reporting, greenwashing, data protection, digital markets and services, and artificial intelligence, the special responsibility of dominant undertakings in competition law increasingly demands a pro-active approach to compliance. This also involves considering the interests of third parties and framing private governance in accordance with fundamental rights and legal principles. An enhanced degree of responsiveness of dominant undertakings results. Second, additional space is being created within competition law to accommodate undertakings that behave in a socially responsible manner, notably regarding sustainability. This is examined in relation to the issue of a fair share for consumers, and private enforcement by means of compliance agreements. After discussing potential objections to responsiveness in terms of democratic legitimacy, legal certainty, and redistribution of wealth, the article concludes that the developments sketched above indeed point towards the reinforcement of the responsive nature of competition law.

1. Introduction

Economic law reflects the important challenges facing our societies such as digital dominance, the concentration of market power, inequal wealth distribution and climate change. ¹ At the same time the Washington economic and political consensus has disintegrated and the neoliberal focus on markets and efficiency is no longer universally accepted. ² This changing context leads to new thinking about the role of the law in regulating markets, and on the responsibilities of undertakings vis-à-vis the public interest. I want to contribute to this debate by focusing on these issues in the context of competition law. Competition law is part of a system of public law norms that serve to discipline undertakings that are active in markets. In that context the general purpose of competition law is to prevent and/or control the negative consequences of the exercise of individual or joint market power. Cooperation that aims to keep our society liveable on the other hand deserves to be facilitated even in the presence of a degree of market power if the advantages outweigh the disadvantages.

My thesis is that effective enforcement of competition law requires that it must be (i) responsive and (ii) open to societal and economic change and respond to the behaviour of undertakings. Therefore, I plead in favour of responsive competition law. This requires a sufficiently flexible system of legal norms and objectives. Moreover, I argue that (iii) in this context something is required from undertakings as well, which is more than respecting the letter of the law after it has been tortured to the point where it can be claimed to be maximally efficient. Not every behaviour that is not explicitly prohibited is acceptable. Respect for the spirit of the law is required too, and this means taking account of the interests of third parties. Of course, the relevant norm should be cognisable, and consequently for novel or developing norms an injunction and penalty payments are more appropriate means to modify behaviour than would be a sanction.

Undertakings are responsible for their behaviour with respect to a growing circle of interested parties, for which they can be held legally accountable. In a broader sense this means there is a growing number of duties of care and a duty of accountability. We can increasingly expect undertakings to be required to act responsibly and to account for their behaviour. Think for instance of climate change liability such as was found by the District Court of The Hague in the Shell case in 2021. ³ This expectation influences how the norms of competition law are interpreted and enforced, not just by competition authorities and courts, but also by undertakings themselves.

In this paper I will focus on the two components of competition law that are quasi-criminal law in nature: first, the so-called cartel prohibition on restrictive agreements between separate undertakings; second, the prohibition of dominance abuse by (in principle) an individual undertaking.

Both prohibitions are enforced ex post, based on the relevant two provisions of the Treaty on the Functioning of the European Union (TFEU) itself (and a Council Regulation originally dating back to 1962). ⁴ Jointly they are sometimes referred to as antitrust.

The third component of competition law is the more administrative system of concentration control where mergers are vetted ex ante: here permission is required for an activity that is in principle considered to be legal but carries certain risks. This was not foreseen in the Treaty itself but has been the subject of a Regulation of the European Parliament and the Council (first adopted only in 1989). ⁵ It is concerned with preventing the emergence or strengthening of market power by means of transactions, although there is some overlap with the controls on dominance abuse, ⁶ and on restrictive agreements. ⁷ At national level concentration control also features certain proceedings where public interests are involved such as the possibility of the Minister of Economic Affairs to allow a merger that was previously blocked by the competition authority in Germany and in the Netherlands. For reasons of space and complexity I will not address the parallels and differences between concentration control and antitrust from a perspective of responsiveness in this article.

I want to address the following three questions: what is responsive competition law and how is this concept related to that of responsive regulation? If responsiveness requires competition law to leave space to accommodate societal developments, does it do so? What new demands are being made on the behaviour of undertakings in the context of new regulation in the EU, and to what extent is a similar trend reflected in competition law?

2. Responsive regulation and responsive competition law enforcement

A. Responsive regulation

To define responsive competition law, I will first address the concept of responsive regulation on which it is based. Responsive regulation has a different understanding of undertakings than classical economics. An undertaking is not simply a combination of capital and labour that is active in a market where supply meets demand. Instead, the undertakings function within a social context, and markets are social constructs that are shaped to a significant degree by law and regulation. This context involves social responsibility: every undertaking needs a so-called license to operate, in other words a degree of societal support that can help justify its existence. Where such support is lacking, an undertaking becomes socially undesirable, and its continued existence is at stake. The behaviour of undertakings and the degree to which it is acceptable play a role in acquiring and maintaining a license to operate.

On the one hand undertakings are therefore expected to behave within the boundaries of what is considered socially acceptable. On the other hand, undertakings themselves influence the perception of what is considered acceptable. This pleads in favour of a responsive view of the law and of its enforcement, which reacts to the attitude assumed by the undertakings concerned. Experience shows that most undertakings at least try to act in accordance with the law, a smaller group occasionally breaks the rules inadvertently, and a minority deliberately and systematically violates them if that is economically rational. Responsive regulation and responsive enforcement are elements of a theory of regulation that was originally developed in the nineties. This concept attempted to bridge the gap between rational and moral behaviour. It also tried to bridge the gap between scepticism about regulation that prevailed in those years and earlier ideas about the primacy of the state.

The Australian criminologist John Braithwaite has developed the theory of responsive regulation. ⁸ He stresses the importance of combining coercion and convincing arguments, collaboration and deterrence in a dynamic enforcement game that involves the entire market, not just individual undertakings. Responsive regulation reacts to the way in which undertakings deal with the enforcement of norms. That is to say that undertakings taking a cooperative stance primarily require information on how to act in conformity with the norms. Deliberate transgressions need to be met with tit-for-tat and repeat offenders deserve harsh sanctions. The intermediate cases need to be led back to the straight and narrow with gentle coercion. The degree to which a norm is cognisable plays a role here: authorities are actively involved in the development of the interpretation of these norms and communicating it. Given the regulatory context there is frequent interaction between the regulators and the undertakings concerned.

This is illustrated by Braithwaite’s enforcement pyramid. Apart from being an escalation and de-escalation model, this pyramid also reflects how undertakings are distributed according to their willingness to comply. The behaviour of the largest group is at or near the base of norm conformity and benefits primarily from persuasion. The parties in the middle require deterrence. The number of parties that break the norms narrows as transgressions become more extreme and systematic, and at the top of pyramid, they are removed from the market. An example of this enforcement pyramid is replicated below (Figure 1). Other examples emphasise voluntary self-regulation, self-regulation on demand, and command and-control.

Figure 1. Enforcement pyramid.⁹

The objective is to move the benevolent undertakings that took a misstep towards compliance at the broad base of the pyramid by first escalating and then giving them the opportunity to make amends. Those who deliberately scale the pyramid eventually lose their social license to operate and their actual license as well. However, Braithwaite qualified that ultimate sanction as a so-called ‘benign big gun’: ¹⁰ a threat so powerful that should suffice as a deterrent to obtain the desired effect. Because the model is based on activating undertakings’ learning it works particularly well for firms that are constantly exposed to regulation and are therefore repeat players. This does not appear to be a necessary condition however, because firms also learn from each other’s experiences.

Responsive regulation is both a more attractive model than the classical focus on the use of force, and more attractive than an approach based exclusively on economic rationality. It is dynamic and combines insights from various disciplines. Moreover, it promotes effectiveness and legitimacy because it is possible to apply instruments in a selective and proportional manner, and to build on self-regulation and other less costly methods of intervention.

Below I will examine to what extent European competition law is already responsive in the above sense. Next, I will see if there is room to extend or enhance this responsiveness. Before doing so, however, I provide a short section on the relationship with responsive law.

B. The concept of responsive law

In the remainder of this paper, I will not make a formal distinction between responsive law and responsive enforcement, as this is not necessary for my argument. However, as a matter of law in context, is worth noting that the general concept of responsive law was introduced by Phillipe Nonet and Philip Selznick in 1978, who framed it as part of a progressive development toward a more open legal system, following the more traditional categories of repressive and autonomous law. ¹¹ The foremost advocate of competition law as responsive law in this tradition is Stavros Makris. He uses the responsive law approach to explain competition law’s indeterminacy as a by-product of conflicts between openness and integrity, referring to (i) constructive interpretation, (ii) responsive enforcement and (iii) catalytic adjudication as the constitutive elements of responsive competition law. ¹²

Although my own approach to responsive competition law is compatible with that of Makris, it is much more a direct elaboration on Braithwaite (who did not address competition law as such) rather than based on Nonet and Selznick or indeed Makris. This is motivated by my basic concern with effective enforcement on the one hand, and with the accountability of undertakings on the other. Hence, I will focus on these aspects instead of developing the relationship with the theory of responsive law.

C. The responsiveness of competition law

Competition law is one of the most intrusive instruments of which society disposes to discipline undertakings. It features fines that can run into billions of Euros and can impose specific behaviour as well as the forced divestment of undertakings’ activities as an ultimate consequence. Moreover, in its milder registers it can be used to investigate markets, increase their transparency, provide undertakings and consumers with more effective choices, and reinforce changes in behaviour. It also forms the basis for private enforcement, especially actions for damages. This reinforces deterrence of norm violations while ensuring equitable compensation. An enforcement pyramid for competition law is sketched below (Figure 2).

Figure 2. Enforcement pyramid for competition law.¹³

At the baseline (but not part of) of this pyramid is its objective: competition on the merits, or effective competition. ¹⁴ Where this is threatened, we see an escalation from informing and warning to various types of decisions, starting with commitment decisions and injunctions under penalty payments to leniency and settlement decisions, and then moving to ordinary fining decisions in different degrees, with behavioural and structural remedies – forced divestment of parts of the business. The latter is the benign big gun or most extreme measure available because the ultimate remedy, licence revocation, does not exist in competition law. Obviously, not all steps are available alternatives in each case: apart from the nature of the response of the firm, the severity of the infringement plays an important role.

A good example of responsive competition enforcement is the use of commitment decisions (as was explored by Makris in his PhD). ¹⁵ In such cases a competition investigation does not lead to a finding of infringement and no sanction or injunction is imposed. Instead, the undertaking commits to a behavioural change that can be directly enforcement-based because it is set out in the commitment decision. This is only possible outside the area of hardcore cartels – which illustrates that not all steps in the enforcement pyramid are available or relevant in every case.

An illustration is provided by the European Commission’s commitment decision of 2021 vis-à-vis the Aspen pharmaceutical company. The Commission ceased its excessive pricing cases regarding Aspen’s essential cancer medicines in exchange for a commitment by Aspen to reduce prices for these drugs by more than 70 per cent throughout the European Economic Area (EEA) and supply guarantees of up to ten years. ¹⁶ The commitment decision included an extensive explanation why the risks of investments and the degree of innovation involved are essential to determining an excessive price in pharmaceuticals, as guidance for future cases. Moreover, the commitments are directly enforceable with fines of up to 10 per cent of Aspen’s worldwide annual turnover.

At the end of 2022 the Commission took a comparable commitment decision with respect to the abuse of data, preferencing of its own services and of providers using its supplementary services by Amazon. ¹⁷ Amazon committed (i) to keeping its competitors’ data separate from its own marketing activities for a seven-year period for the entire EEA, (ii) to ensure that its algorithms for the selection of offers would be non-discriminatory, and (iii) to no longer discriminate in favour of firms that used its complementary services. If Amazon violates these commitments the Commission can impose penalty payments of 5 per cent of its daily turnover and again a fine of 10 per cent of its worldwide annual turnover.

An example of responsive competition law outside the area of commitments is the Commission’s 2021 Ad Blue cartel decision. Here it imposed fines of 875 million Euros in total on German car manufacturers for restrictive agreements regarding selective catalytic reduction systems for their diesel passenger cars. The car manufacturers limited innovation in the reduction of NO_x emissions by constraining the range and size of tanks for the Ad Blue additive liquid for diesel vehicles. Combining punishment and encouragement, this fining decision was accompanied by a guidance letter about standardisation of selective catalytic reduction systems by the same car manufacturers that did constitute a legitimate effort to limit environmental damage. ¹⁸ This shows how the Commission is able to provide a calibrated response to different aspects of behaviour of the same (set of) firms, combining different steps in the enforcement pyramid.

As these few examples illustrate, competition law is already responsive today, even if it is not usually explicitly identified as such. This conclusion requires addressing the objection that –unlike sectoral regulation where undertakings are under constant supervision – competition law does not always involve repeat players, or parties that are regularly involved in the same or similar procedures. The latter is sometimes considered necessary for a learning effect to occur. However large platform companies, Big Tech, are obviously repeat players, as are other multinational companies.

Moreover, it is not necessary to be a repeat player to experience a learning effect regarding the application of competition law. Smaller undertakings especially may indeed start with an information deficit. However, they are targeted by awareness-raising information campaigns and advocacy. Professional advisers who follow the development of competition law closely play a role in conditioning the reactions of all undertakings. The largest undertakings even have a bespoke staff of competition experts, who direct a flexible layer of well-paid specialists. Finally, there is a continuum between different forms of sectoral regulation, such as that regarding electronic communication or digital gatekeepers and specific regulatory topics such as financial reporting, consumer protection and competition law. Hence, I think that the differences between regulation and competition on this point are limited.

The main argument regarding learning effects of responsive competition law, however, is that undertakings do not only learn directly from their own experiences, but also from each other, and therefore indirectly. Moreover, responsive regulation aims to control the market as a whole: there is interaction between supervision and all market parties collectively. It is therefore likely that when an undertaking takes decisions, the impact of regulation, including competition law, is entered into the equation and that communication by regulators or authorities influences the degree to which this is the case.

The responsivity of a regulator or authority in a commitment decision or by means of providing guidance is logical from the perspective of effectiveness and proportionality of the intervention, from a perspective of legitimacy, and to influence the behaviour involved. To be responsive and remain relevant in new markets competition law needs to monitor technical and economic developments closely. If competition law develops responsively, it can help tackle challenges like digital dominance and fighting climate change. I want to analyse whether this development is indeed possible and probable.

So far, I have discussed responsiveness in the sense of the reaction of the authority to the behaviour of undertakings within the enforcement pyramid, in line with the theory of responsive regulation. In the remainder of this paper, relying on a combination of theory and case law, I will focus on responsive competition law, as such, and, as postulated in the introduction, especially on two ways in which responsiveness may develop further within competition law. First, if the law is to be responsive this means that it must be sufficiently open to societal developments to be able to answer the main relevant questions. Therefore, I will first address the goals and norms of competition law and discuss whether these are at present sufficiently open to create space for responsiveness in this sense. Second, the question is what normative demands are made within competition law regarding the behaviour of undertakings. As we will see, these demands are evolving, both in modern regulation and under the competition rules. As a result of this development undertakings will increasingly have to pay attention to the way in which they deal with the interests of third parties.

3. The broad goals and open norms of competition law

As mentioned, I now want to examine the degree to which the objectives and norms of competition law are open in a sense that may promote its responsiveness to societal developments. The objective of competition law is not straightforward: there are different relevant objectives. When discussing them I will differentiate between, first, consumer welfare, second, market structure and competition as such, and third, broader goals of competition law. ¹⁹ Next, I will briefly address the norms involved.

A. Consumer welfare

This first goal derives from economics. In the wake of the neoliberal law and economics proposed by the Chicago School, and specifically Robert Bork, ²⁰ over the past decades the European Commission has repeatedly claimed that promoting consumer welfare is the primary goal of European Union (EU) competition law. ²¹ This concerns the benefits of the individual consumption of goods and services: in practice positive short term price effects for individual consumers.

A problem with this approach is that it leaves little opportunity to consider externalities. These are costs that are not reflected in prices such as the costs of the effects on the environment. Think of the cost of soybeans that are processed into animal feed to produce cut-price meat products and massive meat exports as in the case of bio-industry in the EU while the Amazonian rainforest is burnt down to cheaply farm the soy involved. Moreover, only the here and now is accounted for, and not the future or future generations. The circle of interested parties is limited to direct consumers.

All this makes the approach based on a narrow definition of consumer welfare ill-suited to take account of sustainability or fundamental rights. This narrow focus is, however, useful if the point is constraining and rolling back competition law, and therefore in the interest of large and powerful firms. Here a monomaniac focus on consumer welfare is often accompanied by a firm belief in the self-correcting ability of the market. In this vision excesses would not lead to harm but instead to market entry and innovative alternatives, so concentrations of power are rarely problematic.

B. The structure of the market and competition as such

This second goal derives from the context of European integration and the creation of a common internal market. In this context the European Court of Justice (ECJ) has tended to interpret the objectives of EU competition law more widely than the European Commission has frequently done, and apart from the interests of consumers and competitors points at the structure of the market and the system of undistorted competition. ²² Moreover, the ECJ has stated that there does not have to be a direct relationship with consumer prices, so these goals have a broad scope. ²³ Unlike the efficiency goal discussed above, the link to the case law here is very direct, and not theory but Treaty and teleology based. Finally, these goals are the basis for the special responsibility of dominant undertakings not to distort competition that will be addressed further below.

C. Broader goals

The literature has long identified related, but even broader goals for competition law. Since before the Second World War and during the formative years of the post-war German and European competition law the Ordoliberal school of law and economics emphasised the relationship between economic and political freedom. The Nazi era taught a costly lesson about the threat posed to freedom by the concentration of economic power and its collusion with immoral politics. ²⁴ The discussion on the influence of Ordoliberalism on European competition law carries on to this day. ²⁵

However, there are also more recent indications that broader objectives are involved. In 2018 the Oxford competition law professor Ariel Ezrachi provided the following list of competition goals for the digital society at the request of the European consumer organisation BEUC: consumer wellbeing alongside consumer welfare, efficiency and innovation, an effective competitive structure, market integration, plurality, economic freedom, and finally fairness. ²⁶ This list is highly similar to that presented by Competition Commissioner Margarethe Vestager in 2022 when she pointed out that a competition law based on the principles of the European Treaties can serve several objectives, including fairness and a level playing field, market integration, safeguarding the competitive process, market integration, consumer welfare, efficiency and innovation, and finally pluralism and democracy. ²⁷

All these goals can be traced back to their application in specific competition cases. ²⁸ For instance, the 1979 Hoffmann-La Roche Case on fidelity rebates for vitamins addresses the goal of undistorted competition in relation to the structure of the market. ²⁹ More recent cases, such as T-Mobile about restrictive agreements concerning the remuneration of distributors for concluding mobile subscriptions, ³⁰ and GlaxoSmithKline about the restraint of parallel trade in medicines, ³¹ (both dating from 2009) likewise identify the objective of the structure of the market and of competition as such.

Fairness already appears as an objective in the 1978 United Brands Case about excessive prices, ³² as well as in cases on predation such as TeliaSonera concerning broadband services in 2011. ³³ TeliaSonera also mentions preventing distortions of competition to the detriment of the public interest. There are similar examples for the other objectives mentioned above. This is not to suggest that all these goals are mentioned equally frequently in practice, but it does clearly suggest the existence of pluriform goals.

This pluriformity has been criticised. For instance the French economist and Nobel laureate Jean Tirole is afraid that the existence of pluriform goals leads to the fragmentation of efforts and eventually a limited effectiveness on all fronts. ³⁴ Therefore it is at least a good question how pluriformity relates to the adage that every policy instrument should have a single objective, and how, if this adage is ignored, a balance can be struck between the different goals. ³⁵ Another question is to what extent pluriform objectives are in fact discretionary objectives and what this means for the predictability of the law, and legal certainty. ³⁶ I will just note these questions in passing here.

Although these three (sets of) objectives apply in parallel, in the sense that they are each invoked from time to time, the third encompasses the other two. Hence, despite the abovementioned question marks, I will conclude based on the development and the broadening of the goals of competition law that this area of law is at least to some extent flexible and responsive in relation to the challenges that it must meet. This suggests these goals leave room for a responsive competition law. That raises the question whether the same applies for the norms of competition law.

D. The open norms of competition law

There is a fundamental difference between open and closed norms. An open norm would be that you must drive at a safe speed within built-up areas. This is context-specific and leaves room for judgement and interpretation. In the first instance this room is left to your own responsibility, but with an external check that is complex because a range of factors – such as the weather, the traffic situation, and your driving skills – need to be considered. A closed norm would be a general speed limit of 50 kilometres per hour, and in certain areas no more than 30 kilometres. Your responsibility is just not to exceed this limit, and the external check is straightforward. In simplified form one could say that open norms are flexible but difficult to enforce and closed norms are clear and therefore easily enforceable, but rigid. Open norms are multi-dimensional and can be customised. They rely more on the sense of responsibility of those who are subject to them and require these subjects to think about how they can comply. Closed norms are binary norms and less suited to argumentation. ³⁷

This brings us to the norms of competition law. As mentioned, these have been set out in a mere two articles of the TFEU. That does not necessarily imply they are clear. The interpretation and enforcement of these norms has been elaborated in Council and Commission Regulations and to a lesser extent Directives of the European Parliament and the Council, but especially in the extensive case law of the General Court and the ECJ, ³⁸ as well as in guidelines and interpretative notices of the European Commission. In the EU Member States these European rules are interpreted and applied by the national competition authorities (NCAs) and national judges, alongside their national competition rules. For instance, the Dutch competition law is based on this complex of European rules as a kind of summary of the European system, the interpretation of which must moreover be followed in national law. Alongside this formal setting there is of course a broad range of legal and academic literature on the topic, alongside more practical publications: an entire ecosystem of interpretation has evolved around the concise provisions of the Treaty.

This is unavoidable because the two Treaty articles concerned combine a multitude of concepts that require interpretation such as undertaking, effect on trade, object and effect, prevention, restriction or distortion of competition, dominant position, indispensability, and fairness combined with a non-exclusive list of examples of restrictions of competition. These open norms are defined in a functional manner – and must therefore be interpreted according to their objectives. They have been defined in the case law, and for that reason they are part of an ongoing development of the law.

In this context Ezrachi has described competition law as a sponge, with an open structure. This sponge draws its contents from its environment and renders them when pressed. ³⁹ The sponge has an economic membrane for a filter. This economic membrane too changes shape over time in line with the dominant economic insights of the period. In this way competition law is the result of the interaction between three elements: (i) the law; (ii) its economic interpretation; and (iii) their societal context.

This means that new norms can develop in competition law as the result of societal changes, including academic insights and economic developments.

An example is the interpretation of the exception to the cartel prohibition in Article 101(3) TFEU. Cartels are prohibited unless the advantages outweigh the disadvantages based on this exception, which features two positive and two negative conditions. The positive conditions are (i) contributing to an improvement of production or technological or economic progress, while (ii) passing on a fair share to consumers. The negative conditions are (i) that the restrictions of competition involved must be indispensable to achieving these benefits, and (ii) that not all competition may be eliminated. The interpretation of this norm is now developing further in the sustainability context under the pressure of the societal need to counter climate change and the legal obligations that ensue from this. This concerns, for instance, the assessment of which types of agreements contribute to production or technical and economic progress. It also concerns the question of what constitutes a fair share for the consumer.

A second example is provided by the gradual shift between by object restrictions and restrictions which require an assessment of their effects – in both cases in their factual and legal context. ⁴⁰ The trend in the case law on this topic is for the collection of by object restrictions to shrink while the group of by effect restrictions is growing. Again, there is a relationship here with the degree to which the norm is cognisable. It would be in line with a responsive approach if behaviour that has effects which are known well in advance were to be considered a by object restriction punishable with a sanction. Harmful behaviour with effects that are not known in advance should count as a restriction by effect. An injunction or penalty payments are then more appropriate.

Hence, I conclude that both the open norms and the broad goals of competition law undeniably leave scope for responsiveness in the sense that they are open to providing a proportionate response to new societal challenges. In fact, I would even suggest that it is necessary to fill the large scope for interpretation provided by these open norms and broad goals with a responsive approach to competition law enforcement. This enables enforcers to reflect on the responses of firms to norms and, as I will develop below, requires firms to reflect on and proactively comply with norms.

4. The responsiveness of undertakings

This brings me to the responsiveness that is demanded from undertakings given the developing expectations regarding their behaviour. These expectations have developed not only in competition law but in regulation, precisely in reaction to the new types of societal challenges just mentioned. Hence, I first want to note that outside the realm of competition law we see a proliferation of duties of care and accountability as well as supervision thereof. Therefore, I will briefly address the reporting requirements concerning social and economic goals, greenwashing, supply chain due diligence, data protection, and the regulation for digital platforms and artificial intelligence. This is not just relevant legal and social context, but confluence and concrete links with competition law are also involved. Next, I will address comparable developments in competition law itself.

A. ESG reporting requirements

Market based thinking has long determined our expectations of the behaviour of undertakings. The only duty a business had vis-à-vis society was to make a profit, in the well know adage of Chicago economist and 1976 Nobel prize winner Milton Friedman. ⁴¹ By being as efficient as possible, undertakings would contribute sufficiently to welfare to be excused from any other responsibilities. This point of view is by now outdated. A monomaniac focus on efficiency in a world full of externalities leads to overexploitation. Meanwhile the norm is developing in the direction of a socially responsible undertaking that is accountable for the way in which it pursues economic and social governance – or ESG. These goals are conceptually similar to the 17 Global Goals for sustainable development that were adopted by the United Nations in 2015. ⁴² Apart from action to combat climate change these include fighting poverty, hunger, and inequality, and ensuring gender equality, health care and education.

The EU has meanwhile adopted a regulatory framework for financial reporting that requires undertakings to give account of the degree to which they meet ESG goals. This regards especially the Directive on the Reporting of Non-Financial Information (NFRD) of 2014, ⁴³ and the Corporate Sustainability Reporting Directive (CSRD) of 2022. ⁴⁴ The idea behind these laws is that consumers and investors will be able to vote with their feet and their wallets, by choosing to buy from or invest in certain undertakings over others. In this way they can reward undertakings that set and achieve ESG targets and punish those that do not. Hence, there is pressure from both sides on undertakings to move in the right direction: to make money they need to convince consumers to continue choosing them, and to raise capital they must convince investors to do the same. And in both cases these choices will not be based exclusively on short term price effects or profits.

In this context it is important that the relevant norms are tightened step by step: from earlier voluntary compliance on ESG goals, with the NFRD we have moved to a comply or explain requirement. This relates at least to environmental matters, social and employee-related matters, respect for human rights, anti-corruption and bribery. ⁴⁵ The CSRD no longer contains a similar opt-out clause and introduces a fully binding obligation for reporting requirements with regard to the plans of undertakings to limit global warming to 1.5° C. ⁴⁶ Moreover while the reach of the NFRD was limited to large undertakings with 500 or more employees, the CSRD only exempts micro-undertakings.

B. Supply chain due diligence

The final instalment in this series of reporting Directives details the duty of care for supply chain due diligence. A careful investigation of the degree to which the supply chain respects the ESG goals, including abroad and outside the EU, is now required for the firms that fall within the scope of the Supply Chain Due Diligence Directive (CSDDD) that was adopted in June 2024. ⁴⁷ Implementing this piece of legislation is not only mandatory for the undertakings caught, but also includes a number of duties of care in particular including a duty to act in a norm-conform manner with regard to supply chains, in order to effectively respect environmental and human rights. Think of avocado growers in Peru, sweatshops in Bangladesh or Coltan mining in the Republic of Congo.

For this purpose, undertakings must exchange information and take collective action within the confines set – inter alia – by competition law. This raises the question whether as a matter of competition law it would be possible to go beyond collective purchasing agreements to a boycott of suppliers who violate national or international norms of environmental law or fundamental rights. The answer is affirmative, as addressed in more detail further below.

Undertakings are also bound to report how they intend to meet specific goals relating to climate change. This raises the question whether these reporting requirements can lead to corporate liability if such targets are not met. It does appear that apart from a duty to hold themselves accountable undertakings will indeed face increased liability for failure to attain concrete ESG goals, in particular respecting environmental and fundamental rights. Moreover, competition authorities may well have a role in supervising the application of the open norms that characterise the CSDDD.

C. Greenwashing

A downside of the trend behind ESG reporting is that undertakings increasingly engage in so-called greenwashing: to misleadingly present themselves as more sustainable than is the case. Regarding greenwashing, norms are likewise being tightened. Misleading consumers has long been prohibited based on 2005 the Unfair Commercial Practices Directive (UCPD). ⁴⁸ To increase the effectiveness of this Directive it will be tightened in relation to green claims by the Directive Empowering Consumers for the Green Transition that is currently under consideration by the EU legislator. ⁴⁹ This will qualify generic and non-substantiated environmental claims as by definition misleading under the UCPD.

For now, the legislative action against greenwashing will be crowned by the Green Claims Directive that is likewise still being considered by the legislator. ⁵⁰ This requires proactive substantiation of sustainability claims based on scientific evidence and the state of art of technology. Moreover, these claims must be verified in advance by designated third party conformity assessment bodies. This means there will be a far-reaching duty of accountability. Even in the absence of these measures, however, activists have succeeded in holding multinational corporations accountable for greenwashing claims. ⁵¹

D. Data protection and Artificial Intelligence

Hence, duties of care and a duty of accountability already exist in consumer law and the public reporting requirements of undertakings regarding sustainability and fundamental rights. This also holds for fundamental rights in relation to the General Data Protection Directive (GDPR), ⁵² and the recently adopted Artificial Intelligence Act (AI Act). ⁵³

In the GDPR we find inter alia a requirement of compliance by design and by default. ⁵⁴ This is a duty of care to control the risks at the design stage of a product or service and regarding its default settings with respect to the protection of third-party interests, in a way that respects not just the letter but also the spirit of the law. The AI Act likewise emphasises controlling risk by means of the design. ⁵⁵

This regulation does not just impose duties of care for undertakings but also provides for supervision thereof. The GDPR is applied by a network of national regulators as well as national certification bodies. The AI Act relies on national competent authorities and regulators and conformity assessment bodies. In both cases there is a central responsibility for the European Data Protection Supervisor.

E. The digital markets act and the digital services act

Digital platforms have been regulated more strictly recently given concerns about their market dominance as well as about illegal content, online disinformation, and other societal risks.

The Digital Markets Act (DMA), which entered into force in May 2023, ⁵⁶ establishes far-reaching ex ante obligations for the behaviour of so-called gatekeepers that are designated based on their size. This serves to guarantee the contestability and fairness of digital markets. These specific obligations can be seen as the manifestation of a duty of care that gatekeepers have in relation to purchasers who are also competitors, consumers, and finally the structure of the market. The obligations are a codification of the most important competition cases in this sector and they reach from sharing data with competitors to a prohibition on self-preferencing.

Regarding these obligations the DMA is characterised by an approach that could be qualified as comply and explain. Gatekeepers are not just required to ensure and demonstrate compliance with the relevant norms on their own account, ⁵⁷ but they also must be able to convince the Commission that their proposed approach is in conformity with the requisite norm. The DMA moreover contains anti circumvention provisions that require gatekeepers to act in the spirit of the obligations concerned. ⁵⁸ The measures that can be imposed to enforce the DMA do not just include penalty payments and punitive fines, but for repeat offenders also a prohibition on concentrations in the sector and even structural measures – the forced sale of parts of their business.

The obligations of the Digital Services Act (DSA) ⁵⁹ have applied since 17 February 2024. The DSA provides on the one hand a general framework for conditional immunity from liability and on the other hand rules on the duty of care for so-called intermediary services. The DSA imposes additional duties on so-called very large online platforms (VLOP) and very large online search engines (VLOSE) with proactive content moderation and risk assessment requirements to protect all manner of societal interests. Here too we find an obligation not just to apply the legal requirements but also to explain why the method and design that are applied would be appropriate. Just as in the GDPR and the AI Act compliance by design is demanded. ⁶⁰ Moreover under the DSA digital platforms and search engines must not only act in their own interest, but must also take the interests of their customers into consideration. ⁶¹

Both the DMA and the DSA are characterised by intense supervision, which for the DMA has even been centralised in the hands of the European Commission, ⁶² with a limited role for the national competition authorities. For the DSA on the other hand, VLOP and VLOSE excepted, this supervision has been entrusted to national regulators.

In sum, it does not appear far-fetched to state that in current European regulation duties of care play an increasing role, as does an accountability requirement, under enhanced regulatory supervision. Undertakings are expected to actively display behaviour that is in conformity with the applicable norms, and this is supervised. That makes the regulation concerned responsive to an enhanced degree. Especially in the case of the CSDDD and the DMA there is overlap and/or confluence with the competition rules. But what about competition law itself? What demands are made on the responsiveness of undertakings in this context?

5. The demands on the responsiveness of undertakings in competition law

As mentioned in the Introduction, responsive competition law takes account of the behaviour of undertakings. At the same time legal and regulatory burdens may be symmetrical or asymmetrical, and in competition law they are proportional to the degree of market power involved. Thus, all undertakings have responsibilities in relation to sustainability, but dominant companies have a special responsibility to bolster effective competition. Hence, I will now examine the thesis that competition law (i) imposes increasing demands on the responsiveness of undertakings in relation to the normative framework, and (ii) offers increasing scope for such norm-conform behaviour. For this purpose, I will first address the special responsibility of dominant undertakings, and then agreements that aim to implement the social responsibility of undertakings, using the example of sustainability agreements.

A. Responsiveness and special responsibility

Undertakings enjoying a position of economic dominance have a special responsibility in relation to their consumers (including purchasers), their suppliers and the competitors. The ECJ has stressed special responsibility since the 1983 Michelin I Case. ⁶³

The principle behind abuse of a dominance is that it is not illegal to enjoy a dominant position or to acquire one by way of independent growth, or even to force less efficient competitors out of the market ⁶⁴ Acquiring a position of dominance by means of takeovers or minority shareholdings can however be an infringement of competition law, ⁶⁵ and can be prevented by means of concentration control. And although intent is not a necessary requirement for a finding of dominance abuse, ⁶⁶ this does not mean that the behaviour of dominant undertaking is not subject to norms. The special responsibility implies that once an undertaking enjoys a position of market power that allows it to behave independently from (and unconstrained by) its counterparties in the market, it may not behave in such a way as to undermine genuine and unadulterated competition.

Private regulation by dominant undertakings that use their general conditions to establish a bespoke legal regime that profits them disproportionately, ⁶⁷ is limited by this special responsibility. ⁶⁸ This will increasingly involve a duty of care, proportionality and transparency, as well as principles of good governance and fundamental rights that discipline private regulation by such undertakings. ⁶⁹ They will be held accountable for the way in which they treat their business partners because of the asymmetrical and therefore unequal point of departure. Just like a government they will have to balance interests in a proportional manner. This appears reasonable, especially now governments have become increasingly dependent on these large firms. Moreover, I believe that – like morally conscious beings – under proportionality they should take the interests of their counterparts into account. ⁷⁰

An example can be found in the European Super League Case of December 2023. Football sports federations FIFA and UEFA had blocked the creation of a new European competition by means of the power they derived from the competitions they controlled themselves. The Court decided that where an undertaking controls access to a market it must wield its power based on material criteria that are transparent, clear, and sufficiently precise in a manner that is non-discriminatory and proportional, and subject to effective controls. This requires a framework of procedural rules. The absence of such guarantees implies the existence of abuse of dominance. ⁷¹ Future case law is likely to develop the demands that are made of dominant undertakings in this context further.

The Google Shopping Case of the General Court in 2021 on leveraging of market power and self-preferencing by Google regarding competing price comparison sites on its platform sets a comparable norm. ⁷² The special responsibility of Google is enhanced by the fact that this undertaking is an important gateway to the Internet. The General Court interpreted the fact that Google modified its behaviour to the detriment of its competitors as unnatural because open access and providing consumers with objective information were the original business model of its platform. Moreover, as a super-dominant undertaking Google’s behaviour must be in accordance with the general principle of equal treatment. The consequences of this are very concrete: the General Court upheld the 2.42 billion Euro fine that the Commission had imposed on Google. Recently this judgment was in turn upheld on appeal by the ECJ. ⁷³

This means that to respect the norms of competition law dominant undertakings at least will have to apply themselves to actively interpretating these norms, including the application of general principles of law. In this context behavioural changes are increasingly imposed on undertakings by findings of new types of abuse, with an emphasis on their potential effects. ⁷⁴

An example of a new type of abuse based on confluence with the GDPR is found in the German Facebook Case that led to the Meta ruling of the Court in July 2023. ⁷⁵ Here the ECJ found that when determining if there is an abuse of dominance a competition authority can decide whether there is an infringement of other rules such as the protection of consumer data under the GDPR that qualifies as such. The competition authority must, however, consult the specialised supervisory authority and may not take a decision contrary to the position of that authority. In this case that meant that forcing consumers to surrender their data was part of a competition infringement, and it thereby constituted a new type of abuse. This is an example of how the overlapping powers of competition authorities and other authorities are shaped in practice. ⁷⁶

The assumption that this is part of a trend where increasing demands are made on the behaviour of firms might seem to be contradicted by the ex-ante regulation of the largest digital platforms as gatekeepers under the DMA. After all, the latter categorically prohibits certain behaviour in advance by these largest and most powerful undertakings in the digital domain, like self-preferencing and combining consumer information from various sources. Defining a relevant market, establishing dominance, and finding an abuse ex post is no longer thought suitable: the problems of market power are such that we are in the realm of super-dominance and therefore of enhanced special responsibility.

Nevertheless, this is a form of escalation that even if it was not explicitly included in Braithwaite’s model appears to be compatible with it: for this exceptional category of undertakings there are stricter general rules of which the applicability does not have to be proven on a case-by-case basis, if an undertaking is big enough to be caught by the regime. The DMA can also be seen as a form of responsive regulation because it is a form of strict control that has been introduced because self-regulation based on the competition rules did not offer an effective solution. ⁷⁷ Moreover, the Commission can add new rules to the DMA, and it contains anti-circumvention clauses. ⁷⁸ In this manner the existence of credible enforcement including an ultimate remedy is guaranteed. This contrasts with competition law with its strict procedural guarantees and high burden of proof where enforcers struggle to achieve timely results when faced with deep pocketed repeat offenders.

Something similar applies to the co-called New Competition Tool (NCT) which enables market investigations to take place where there are structural problems, and imposing appropriate, and therefore necessary and proportional, remedies directly. ⁷⁹ This proposal was initially superseded by the DMA (albeit that the latter has a very different scope, it competed for the same limited resources both in terms of the need to invest political capital, and at DG Competition in terms of available staff), but has resurfaced in the recommendations of the recent Draghi Report. ⁸⁰ Here too structural remedies – the forced divestiture of parts of the undertaking – are possible. This means further scaling up the escalation ladder for certain types of problems and again doing so without the individual determination of a dominant position in a relevant market, or a finding of abuse. In contrast to the gatekeepers under the DMA where at least relatively straightforward quantitative criteria were used, in the NCT more qualitative criteria such as structurally weakened competition take central stage.

At the same time the distinction between the two can also be based on the degree to which the relevant norm is cognisable in advance, and therefore the degree of legal certainty: in the case of the DMA, we find clear norms that can lead to sanctions. The NCT does not require establishing an infringement, and therefore there is no liability or sanction either. Only proportional future-oriented obligations can be imposed. If competition law can be seen as a flexible shell around the hard core of sector specific ex ante regulation in the DMA, then the NCT forms an even more flexible shell with respect to the competition rules. ⁸¹

B. Reponsiveness and social responsibility

In addition to the special responsibility of dominant undertakings I now want to address the general social responsibility of undertakings and their relevance to the responsiveness of competition law. This dimension has been developed especially in relation to sustainability and the exception to the cartel prohibition.

Sustainability is a broad concept, and includes combating climate change and environmental damage, the efficient use of natural resources, protecting human rights and animal welfare. If undertakings are unable to take decisive steps to promote sustainability on their own accord, they may need to conclude agreements that restrict competition. This can for instance be necessary where there is a first mover disadvantage. The increasing costs that arise when externalities are accounted for can place pioneers at a disadvantage with respect to their competitors that do not take similar steps. But if competitors agree to take account of the CO₂ footprint of their products in their pricing, or to limit the use of plastic packaging, there is a level playing field. Such agreements between competitors can therefore be seen as responsive behaviour with respect to broader social interests. Just as in the case of supply chain due diligence this raises the question of what room there is for such agreements within competition law.

In line with earlier academic and societal criticisms, ⁸² the Authority for Consumers and Markets in the Netherlands has pleaded for an open approach to sustainability in competition law at least since 2020. ⁸³ Together with other national competition authorities it has played a pioneering role in this respect in the European context, challenging the European Commission’s more conservative approach. Since then, various competition authorities within and outside the EU – eg in the UK, Japan, Singapore, and New Zealand – have adopted guidance on sustainability, ⁸⁴ although resistance against such an approach notably in the US remains strong.

In the most recent version of its Guidelines on horizontal agreements of July 2023, which included a new chapter on sustainability, ⁸⁵ the European Commission has shifted its position as well. In the first place the Commission recognises the problem of externalities – the possible effects of behaviour on third parties – and thereby the relevance of a wider circle of interested parties, including the role of future generations. Moreover, the Commission in its Guidelines does not just refer to the European Green Deal, ⁸⁶ but also to international sustainability norms such as the sustainable development goals of the United Nations. ⁸⁷ The Guidelines have multiple layers. For instance, for agreements that aim to set sustainability standards that meet certain conditions, they create a soft safe harbour, and for agreements outside this soft safe harbour they elaborate a more relaxed version of the legal exception.

C. The fair share

The most contentious point in the debate on how EU competition law should address sustainability concerned the ‘fair share’ requirement of the Article 101(3) TFEU exemption: when evaluating agreements that restrict competition but have positive net sustainability effects the next question is whether these benefits are shared fairly with the consumers within the relevant market that is the subject of the agreement. By including the concepts of individual use- and non-use-benefits on the one hand and collective benefits on the other hand, it has become possible for the Commission to take broader interests into account in this context.

Individual use values take the shape of benefits in terms of price, quality, and choice, as has traditionally been the case in competition law. Examples are agreements to promote the availability of biological produce or the use of sustainable timber that provide better quality and more choice. Individual non-use values are novel, and relevant where we are ready to pay for more sustainability because we think it is worthwhile without obtaining direct use benefits in terms of price, quality, or choice. An example is the use of less polluting carburants. Collective benefits are much broader and new for competition law but are only considered where there is a substantial overlap with the circle of direct consumers. For instance, when a reduction in the emission of greenhouse gases for technology that during a start-up period leads to higher costs for consumers, who on the other hand do benefit from controlling climate change and/or better air quality. This provides room for a responsive approach to undertakings that take their social responsibility.

The approach taken by the Commission raises the issue whether eventually other benefits, notably those for third parties, should not be considered, and to what extent. The European development to date is only a cautious first step in the recalibration of competition law. ⁸⁸ For instance the Commission still requires full compensation for consumers in the relevant market. From the perspective of enforcement this approach is nevertheless responsive because room is offered for interests that go beyond efficiency, and with reference to international norms. Moreover, the Commission is prepared to provide proactive guidance to undertakings that seek possibilities to collaborate on sustainability. ⁸⁹ Finally, no fines are imposed on undertakings that enter this dialogue based on good faith, and even if illegal restraints of competition were to be involved, such problems are settled.

D. Polycentric regulation and private enforcement of norms

Under the Commission’s new approach to sustainability, space is offered to undertakings that are responsive regarding their social responsibilities. In line with the work of Elinor Ostrom, who in 2009 was the first woman to win the Nobel prize for economics, such action can be seen as a form of decentralised self-regulation, or polycentric regulation, by these undertakings. Based on her analysis of a wide range of case studies where scarce resources are managed privately yet collectively, Ostrom, in my view convincingly, argues that polycentric regulation based on experimental efforts at multiple levels is necessary to have any chance of reaching climate goals. ⁹⁰ Waiting for a unitary actor to take decisive steps – an effective world government or even maximally effective national governments – is after all illusory. ⁹¹

More recently, based on the legal concept of polycentricity introduced in the 1970s by Lon Fuller, ⁹² Ioannis Lianos has argued against the economics-based view of competition law with neoclassical price theory and an effects-based approach to consumer welfare at its centre. He sees this monocentrism as limited in scope to a bilateral relationship in an isolated relevant market that reduces individuals to one-dimensional consumers. Instead, Lianos advocates a polycentric competition law that operates within a web of interdependent relationships and can take multiple objectives and markets into account, inter alia by means of balancing. ⁹³

It goes beyond the scope of this paper to provide a systematic analysis and comparison of these two approaches to polycentricity – which share an awareness of the complexity of real world economic and social problems and consequently of the need for new solutions. My own argument on polycentrism within the context of competition policy is that private agreements are needed to advance social goals, and that we cannot just rely on governments to do so. Moreover, intervention to advance social objectives should take place at the lowest effective level, including that of private actors, to help solve problems of trust and reciprocity and allow experimentation and innovation. This is both in the interest of legitimacy and of speed, itself an important dimension of effectiveness. ⁹⁴

Against this background, I want to draw attention to the fact that the ACM has meanwhile explicitly opened the door to compliance agreements. By such agreements undertakings mutually undertake to enforce norm-compliant behaviour in line with sustainability legislation, not just amongst themselves but also vis-à-vis suppliers or customers.

A practical example regarding suppliers is the possibility of a collective boycott by Dutch garden centres of growers using prohibited pesticides accepted by the ACM, in the absence of effective public enforcement. ⁹⁵ This is necessary to prevent malevolent growers from distorting competition by offering an illicit competitive advantage to garden centres selling their cheaper but poisoned products. A garden centre avoiding such growers individually will suffer if its customers depart for less scrupulous but cheaper competitors. For similar reasons, growers will be tempted to continue violating the law if their competitors continue to do so.

Another practical example, this time regarding customers, is provided by Dutch waste collectors forcing separate delivery of waste streams on their customers: polluters who are generally already legally bound to do so. ⁹⁶ If waste collectors choose to compete by facilitating those who violate the law, the norm of waste recycling will remain a dead letter. Here too collective compliance agreements between competitors are necessary because there is not only no effective public enforcement, but hardly any public enforcement at all.

Hence, compliance agreements are necessary to remove perverse incentives from the system. In both cited cases, however, accepting this form of private enforcement is at odds with prior interpretations of EU law about the primacy of public enforcement, which are at least partly based on the optimistic assumption that public enforcement will by definition be effective. Moreover, the European Commission traditionally mistrusts undertakings that claim they are enforcing public norms for fear this will open the door to competition infringements. This poses a dilemma for NCAs between pursuing effective enforcement on the one hand and loyally following the Commission’s lead on the other.

However, EU law as interpreted by the ECJ when refining and elaborating on Hilti (1989) already provides the possibility to rely on this new form of private enforcement where public enforcement falls short, complex assessments that only public authorities can carry out our not involved and the restrictions concerned are proportional to the legitimate objective of ensuring compliance with a public norm, and are applied in observance of procedural fairness. ⁹⁷ Its objective is not the restriction of competition but rather norm-enforcement. In any event unfair competition that is at odds with national or international sustainability norms is not an interest that deserves legal protection. Moreover, private enforcement fills a gap when public enforcement, especially ex post, is ineffective – too little too late. Its relevance will increase further once compliance agreements are used to ensure the respect of environmental and human rights norms in the context of supply chain due diligence and enforcement of the CSDDD. That process is likely to influence the interpretation of competition law at large on this point.

Regarding such action as a form of private enforcement within the context of competition law is consistent with a modern view of compliance. ⁹⁸ It is an open question to what extent this phenomenon will remain limited to sustainability, environmental and human rights norms within competition law. In principle there is no reason why a broader application would not be possible, especially in cases concerning a clear norm that is not enforced uniformly and effectively, while private enforcement would meet those standards and be applied in a proportional manner. As happened in the garden centre example, this can involve meeting procedural standards such as a possibility to appeal before sanctions are applied, and reconsideration by a body that is at a distance from direct business interests. This is comparable to the demands that are made of private regulation in the dominance context, as we have seen in the European Superleague Case of December 2023. ⁹⁹

This leads me to conclude that EU competition law increasingly demands and facilitates responsive behaviour of undertakings with respect to their social responsibility. The consequences of this finding remain to be studied more closely but will influence the dynamics of the responsive enforcement of competition law.

Before concluding, I briefly want to address three objections to responsive competition law. These concern (i) the legal certainty and (ii) democratic legitimacy of responsive competition law as well as (iii) the legitimacy of redistribution by competition law.

E. Legal certainty

Regarding the first objection, I have already raised several arguments above. This objection concerns legal certainty and the principle of nulla poena sine lege: the existence of a prior binding norm is a requirement for the ability to impose a sanction. For the legitimacy of a responsive system this means that behaviour of which the noxious effects were not predictable in advance, a sanction should not be imposed, but instead for example an injunction combined with a forward-looking penalty payment. Where behaviour was in good faith it could even mean that only norm-transmitting guidance is applied. The same applies to norms that are still in development. Hence, legal certainty can be respected within a system of responsive regulation, although this aspect will require more attention when the theory is developed further.

At the same time, as I have noted, undertakings are increasingly required to behave responsibly when interpreting norms. This means that where the effects of their behaviour are predictable, they must act in accordance with the spirit of the norm, and must therefore consider the effects on third parties, and the development of societal views concerning the contents of the norm. This does not appear to be an excessive demand. In the end, undertakings are also fully engaged in influencing the context of norms by contributing technical and economic expertise, lobbying behaviour, the strategic disbursement of information and pursuing legal procedures. They also deliberately influence consumer behaviour by means of advertising and information provision that is increasingly based on the use of data and behavioural economic and psychological insights. It is not excessive to expect them to consider principles like proportionality and an objectifiable balancing of interests when doing so.

F. Democratic legitimacy

Regarding the second objection, it is possible to contest the democratic credentials of independent supervision based on norms characteristic of economic law, even if the latter is responsive. Or rather, this objection is even relevant if the norms concerned are open and the goals are broad, as I have argued. This applies especially where questions of redistribution are at issue. ¹⁰⁰ It can also be applied to the fair share of the consumer, whether in the relevant market or not. It goes too far to try to provide a comprehensive answer to this objection here. Instead, I will limit myself to a few remarks that I want to elaborate on in future work.

Part of the answer is that the correct context for this issue is the democratic state based on the rule of law, and not direct democracy. The rule of law allows for certain processes – such as the independent adjudication of norms by judges, but also by regulators – that have indirect democratic legitimation at best. From a perspective of checks and balances, so controls and institutional countervailing powers, this is precisely as it should be. The enforcement of public norms by definition has a sufficient degree of democratic legitimation insofar as these are national or international norms that have been adopted as the result of a democratic process or were confirmed by such a process. In the case of sustainability, including norms on climate change and fundamental rights, this standard will generally be met. In this case the problem does not arise, even in the case of compliance agreements. The democratic decision-making process has after all been respected.

G. Redistribution

The third question is whether it is legitimate for competition law to engage in redistribution. I must admit that I fail to see why this would be a pressing issue. Even if a broad definition of welfare is used it appears to me that redistribution as such is not an objective of competition law (with possible exceptions in developing countries ¹⁰¹ ), but rather a side effect of the behaviour of undertakings or agreements between them.

Here there is a distinction to be made between the evaluation or approval of redistribution with a sustainability objective by undertakings on the one hand and redistribution on its own account by a competition authority. In the first case, such as when the legal and hence democratically legitimised exception to the cartel prohibition is administered, the test of the redistributive effects involved will be sufficiently marginal to ensure that there is effectively no question of redistribution by means of competition law. Imposing redistribution in the context of an assessment of excessive pricing or unfair trading conditions is limited both by a demanding legal and economic burden of proof and a practice of extreme reticence. The latter is based on intervening only in exceptional cases and finally on the adage that preventing the emergence of market power and exclusion is the most effective remedy against exploitation. The solution therefore is normally sought in promoting effective competition, not direct redistributive intervention.

Finally, there is a difference between redistribution between certain groups of undertakings in a concrete dispute and a redistribution that applies more broadly or that is even of general application. In the first setting there is a better case to be made for the competence of competition law than in the latter. It is questionable however whether EU competition law will ever play a significant role in such general redistribution. This type of issue is more relevant to the application of an NCT than for ordinary and case-specific competition law. Hence this question should be addressed especially in the NCT context, if at all.

6. Conclusion

The concept and theory of responsive regulation are based on the interaction between regulators and undertakings that aim to make regulation more effective and legitimate. In the introduction I asked what should be understood as responsive competition law against this background. My conclusion on this point is that responsive competition law involves taking the behaviour of undertakings and their attitude with respect to the prevailing norms into account during enforcement, with the objective of keeping benevolent undertakings on the right path, correcting deviants, and disciplining repeat offenders with increasing degrees of severity, and the overall objective of directing undertakings towards compliance. As a matter of current practice therefore, competition law is already responsive.

Next, I have tried to expand this model of responsive regulation to a yet more responsive system of competitive law based on two theses. My first thesis was that truly responsive competition law also requires a system of broad objectives and open norms, to enable societal developments to find their way within the law as the basis for the demands that are then made of enforcement and compliance. The related question set out was whether such open objectives and norms are in fact found within competition law. Based on the overview provided above, I conclude that these conditions are met. The rapidly changing social context has an impact on the perception of the goals and thereby on the purpose of competition law, and on the way that its open norms are interpreted and applied.

My second thesis was that within competition law increasing demands are made on the responsiveness of undertakings themselves. The question used to test this was what demands on undertakings are made within new forms of regulation, and whether a similar trend is reflected in competition law. I have provided several examples to show this in fact occurs and involves new duties of care and accountability. More specifically, I have given examples of the way in which competition law (i) demands responsiveness of undertakings that have a special responsibility and (ii) treats undertakings that are in fact responsive regarding their social responsibility. The former are subject to increasingly strict demands and the latter are facilitated.

In this context cross-fertilisation appears to occur with regulation such as the GDPR and the DMA (which is itself developing into a prime example of responsive regulation), and the same is likely to hold for the CSDDD. Competition law itself initiates similar developments, as is exemplified by setting procedural conditions for private regulation by dominant undertakings, and the use of compliance agreements to mobilise social responsibility in the context of private enforcement. In answer to the three questions jointly, we can therefore conclude that an enhanced form of responsive competition law is emerging under the influence of theories and practices encountered in economic regulation. As a result, competition law is in constant development, which is necessary to meet the demands that modern society places on economic law: its context.

In terms of policy recommendations, this provides actual and potential new avenues towards enhanced effectiveness and legitimacy by means of more broadly distributed participation and responsibility that remain to be examined more fully. Simultaneously, it is important that competition law remains stable, predictable, and coherent. This, too, is necessary to provide an effective and legitimate framework for the behaviour and the decisions of undertakings. Developing established concepts such as necessity, proportionality, and legitimate objectives, can help to do so, as can policy guidelines and guidance in individual cases by competition authorities. Beyond informal guidance this could include seeking precedents by means of findings of non-applicability in formal Decisions by the European Commission. This should make competition not just more responsive but more future proof.