If a large part of competition law academics seems to agree on the necessity of the NCT – we can read that today the question is no more if we need this new tool, but how we could implement it[5] – its eventual adoption should not be taken lightly. The NCT would constitute an unprecedent evolution in the EU competition law history. It would indeed empower competition authorities to pronounce structural or behavioural injunctions not on the account of any illegal behaviour of the undertakings, but due to the competitive features of the market on which they are operating. In other words, the competition authorities’ intervention would have no regard to the merits of the undertakings, the legitimacy of the position they have acquired nor to their most fundamental rights: liberty to act on the market and property. It is undeniable that the economy is modernizing and that this evolution is a major issue for competition authorities. Nevertheless, the end does not justify all the means. It is necessary to consider with the greatest prudence the consequences that the enactment of such a tool could have. In our view, it could become a real Trojan Horse to implement an interventionist policy which could compromise the competition process itself and overlook the respect of fundamental rights and compliance with the Rule of Law. The NCT would establish a hybrid model between regulation and law[6]: it would not be a sectorial regulation to the extent that no precise behaviour would be prohibited[7]; nor a general legal rule as the NCT would not define any determined or determinable behavioural norm enabling undertakings to escape public intervention.

This legislative proposal did not appear from nowhere: it is the direct consequence of an evolution ongoing since the enactment of Regulation n° 1/2003. The modernisation of competition law, with the objective of strengthening the efficiency of the antitrust rules enforcement, may have had positive effects in the short term, but has, at the same time, compromised its sustainability in the long run (Part II). Nowadays, one would be blind not to see that competition law is at the end of its rope, however the NCT raises so many concerns that it could perhaps prove to be a cure worse than the disease (Part I) compromising compliance to the Rule of Law (Part III).

I. Main concerns raised by the New Competition Tool

The list of the concerns raised by the NCT is long: loss of legal certainty (1), disproportionality and potential inefficiency of the remedies (2), significant weakening of fundamental rights (3) and risk of disparate application (4).

1. An imprecise standard compromising legal certainty

During the above-mentioned Webinar, speakers have highlighted that it is necessary to provide details of the NCT’s intervention threshold and standard of appreciation. Isabelle de Silva pointed out that the NCT would be used in case of “distortion of competition”, admitting this expression remains rather vague. Can we expect this standard to be explained further? Not really, if we refer to the one used by the CMA. After eighteen years of decision-making practice, “the substantive boundaries of the regime remain unclear” and the Market Investigation regime is described as “a highly flexible policy instrument”[8]. Here is what the UK guidelines provide: “[a] market investigation may examine any competition problem and identify the featurecausing the problem. It aims only to see if competition within the particular market under review is working well or can be improved and is not seeking to establish general rules and obligations for firms. (…) As well as being able to look into the conduct of firms, the [Competition Commission] can probe for other causes of possible [adverse effects on competition], such as structural aspects of the market (including barriers to entry and expansion) or the conduct of customers” [9]. Regarding market analysis, we learn that “[t]he extent to which the [Competition Commission] will seek to quantify particular effects (…) – and the degree of precision with which this is attempted – is likely to vary from case to case”[10]. What should be stressed is that “Competition Commission is not required to treat any aspect of market structure or behaviour as ‘given’, but rather has an extremely wide brief to identify any aspect of the market to which changes could be made that would significantly improve competition”[11]. In the end, to determine if the functioning of the market could be improved, the CMA conducts a counterfactual analysis: “[i]n identifying some features or combination of features of the market that may give rise to an [adverse effect on competition], the [Competition Commission] has to find a benchmark against which to determine how the market may be judged to be performing. In the absence of a statutory benchmark, the [Competition Commission] defines such a benchmark as a ‘well-functioning market’ (…) – i.e. one that displays the beneficial aspects of competition (…) but not an idealized perfectly competitive market. The benchmark will generally be the market envisioned without the features”[12].

The question revolves around the enigmatic definition of a “well-functioning market”. Up to now, competition was considered healthy provided it was not compromised by anticompetitive behaviours. Merger law was, for its part, conceived as a complementary strategy of prevention of the subsequent occurrence of such behaviours. With the NCT, the idea is quite different: as soon as the intervention of a competition authority can “make things better”, it becomes relevant. Needless to say, the competitive functioning of almost every market could be “improved”. Such a standard would thus leave considerable scope for competition authorities which would be able to intervene as they please[13], thus broadening their power to enforce a real competition policy. This would dangerously reduce legal formalism which hitherto ensured the predictability of competition authorities’ intervention. Undermining legal certainty in such a way would be all the more serious as the range of remedies foreseen by the NCT is wide and open.

2. The potential inefficacy and disproportionality of the remedies

As the French competition authority itself agrees[14], the design and monitoring of the remedies is a major issue for competition authorities. The difficulty becomes even greater when remedies are to be applied on complex markets. It is then very challenging to appreciate, at the time of their conception, all the effects they are likely to produce in the medium or long term. Indeed, remedies can lead to a weakening of competition, have a negative impact on consumers or even deter investment in dominant undertakings. These problems have already been pointed out about the CMA’s Market Investigation decisional practice[15].

Moreover, the French competition authority advocates for the absence of hierarchy between behavioural and structural remedies. This would constitute a major difference from what is foreseen in the field of antitrust law by Regulation n° 1/2003 which provides as follows: “[s]tructural remedies should only be imposed either where there is no equally effective behavioural remedy or where any equally effective behavioural remedy would be more burdensome for the undertaking concerned than the structural remedy”[16].It should be noted that the CMA is endowed with a wide margin of appreciation regarding injunctions: “[h]aving established a competition problem, and identified its causes, the Competition Commission is able to impose a wide range of legally enforceable remedies that typically focus on making the market more competitive in the future and make recommendations for remedial action by other public bodies”[17]. Choice is, to say the least, very open! The CMA has already made use of its faculty to impose structural remedies, for instance in the case of the London airports in which it ordered the BAA company to sell several of its establishments[18]. It seems very surprising, if not incoherent, to consider on the one hand that a structural injunction would generally be disproportionate in the antitrust field, whereas in this case an infringement is characterized; while considering on the other hand that such an injunction could be proportionate if pronounced on the basis of the NCT in the absence of any wrongdoing. This criticism has already been addressed to the Commission’s decision-making practice regarding commitment procedure, as soon as commitment decisions making legally binding structural commitments are far from being uncommon[19].

Granting such a wide and non- hierarchical range of remedies to competition authorities would be all the more preoccupying as we can reasonably expect the judicial review of the decisions founded on the NCT to be superficial, which would contribute to the weakening of the undertakings’ fundamental rights.

3. The significant weakening of the undertakings’ fundamental rights

In spite of the letter of Regulation n° 1/2003 which details at Article 23(5) that decisions taken in application of antitrust rules are not of a criminal nature, these belong to the criminal field in the sense of the European Court of Human Rights (ECtHR)[20]. It should nevertheless be specified that the decisions assimilating antitrust proceedings to criminal charges are legally binding only for Member States. Indeed, since the European Union has not yet become a party to the European Convention of Human Rights (ECHR)[21], the ECtHR is not competent to operate direct control on the European Court of Justice (ECJ) case-law. Nevertheless, driven by the conclusions of several Advocates General[22], the ECJ has recognized the benefit of the main guarantees provided by the ECHR, thus ensuring de facto compatibility between the case law of the Strasbourg and Luxembourg Courts[23]. As a result, the rights of defense, the presumption of innocence and the principle nullum crimen, nulla poena sine lege benefit companies in antitrust proceedings[24]. Even though this protection is far from being perfect[25], it must be acknowledged that the ECJ case law has progressively improved the respect of fundamental rights.

The adoption of the NCT could jeopardise these achievements. Indeed, according to the Engel criteria[26], the main criterion used to determine if a proceeding constitutes a criminal charge is its repressive purpose[27]. Thus, if the ECtHR once considered that a simple blame is a criminal sanction[28], it is mostly on the basis of its punitive goal. Conversely, the question was raised before the ECtHR as to whether a Russian sectorial competition law pertained to the criminal sphere. The ECtHR considered that this Russian proceeding did not fulfill the criteria of a criminal charge on the grounds that (1) it was limited to certain sectors and not universally applicable as would be a criminal law; (2) it aimed at preventing distortions of competition but not at punishing offenders and (3) it allowed undertakings to justify their behaviours in cases where they had beneficial effects on the market, whereas a criminal wrongdoing cannot be subject to an utilitarian justification[29]. This reasoning could undoubtedly be transposed to the NCT. An additional argument confirms this idea: the ECtHR did not consider that the implementation of the Russian competition law was a criminal charge even though it established an infringement and a sanction consisting in the restitution of the illicit gain. As long as the NCT would not establish any infringement nor foresee any sanction and that its aim would not be repressive at all, one can already be sure it would not be considered as a criminal proceeding by the ECtHR. In conclusion, despite the significant economic impact for companies that could have injunctions pronounced on the basis of the NCT, undertakings would certainly be deprived of the guarantees linked to criminal matters.

It is worth mentioning that if the UK government tacitly acknowledged that Market Investigation proceedings shall respect the principle of a fair trial foreseen by Article 6 of ECHR – which applies in criminal as well as in civil matters – the respect of this principle is deemed to be satisfied provided that undertakings have a right to appeal before the Competition Appeal Tribunal (CAT)[30]. It nevertheless remains to be seen whether, beyond formal compliance to the principle, it is reasonable to consider that a superficial judicial review is truly satisfying in terms of effective protection of the rights of defense.

The only constraint to which authorities would be subjected to in the enforcement of the NCT would be the respect of the principle of proportionality – as it is a general principle of EU law – between the identified competition feature and the prescribed remedy. However, let us not be naive. What would the judicial control of proportionality consist of? To answer this question, a parallel can be drawn with the control of legality exercised over commitment decisions. Since the NCT and the commitment decisions would have in common that they are to be taken without characterizing any infringement, the arguments developed by the ECJ in the Alrosacase could probably be transposed to the future proceedings based on the NCT. The ECJ’s reasoning in the Alrosa case is as follows: as enforcement decisions and commitment decisions have different mechanisms, means of action, objectives and underlying concepts, it justifies a different standard of control of proportionality by the Court, limited to a “manifest error of the Commission”[31]. Yet, how could the Commission or any other competition authority be manifestly mistaken regarding the extent of the margin of appreciation conferred by the NCT? It is worthy of note that during the 2010-2020 period, the CMA successfully defended most challenges before the CAT, winning 10 of the 12 cases – the two defeats being founded on procedural issues and not substantive ones[32].

The modernisation of competition law aimed to avoid competition authorities’ over- intervention by the adoption of a “more economic approach” which goal was to prevent type I errors (false positives). Such an ambitious approach is nevertheless more resource-intensive for authorities. Indeed, the analysis of increasingly complex markets has resulted in an exponential rise in the length of the Commission’s decisions[35] and in the cost of antitrust enforcement. It was thus necessary to find a way to alleviate this difficulty, and the competition authorities quickly found a ready and powerful means to circumvent the problem: Article 9 of Regulation n° 1/2003 establishing commitment procedure. Allowing the authorities to dispense with the need to conduct an in-depth investigation and to characterise an infringement, this procedure was conceived to deal more efficiently with cases already covered by an established case law and for which the competition authorities would not have considered the infliction of a fine. It should be pointed out that the design of this procedure is particularly inconsistent. Here is the paradox: when an established case law exists, competition authorities inflict fines in the great majority of cases. The only cases in which a merely symbolic fine is pronounced are precisely those which are complex and unprecedented. It thus seems difficult, if not impossible, to implement the commitments procedure in accordance with its original spirit. The next part of the story was inevitable. Competition authorities have fully seized their power to initiate commitment proceedings in order to quickly “negotiate”[36] commitments with docile companies fearing the idea of engaging in a long and potentially expensive quasi-criminal litigation, including in complex cases where the existence of an infringement was uncertain in the light of existing case law. That is to say, precisely in the cases where it would have been necessary to develop a sound theory of harm; to provide for a strong legal argumentation regarding the existence of an infringement; and last but not least, to submit this reasoning to genuine judicial control.

This evolution, far from being limited to a few marginal cases, is a widespread phenomenon. Following the entry in force of Regulation n° 1/2003, “commitment decisions subsequently emerged as the default antitrust enforcement tool in the modernization era and are still today becoming increasingly frequent”[37]. Thus, instead of fostering a more balanced and better reasoned enforcement of antitrust rules, the reform paradoxically led to the decline in the number of enforcement procedures based on Article 7, including in the most complex cases[38]. Beyond the concerns raised by the increasing use of commitment procedures regarding the undertakings’ fundamental rights, this soft and “efficient”[39] decision making practice put a halt to what was essential: the evolution of the case law. The gradual adaptation of antitrust rules, which are sufficiently flexible to seize new cases, has been sacrificed for the sake of a short-term efficiency policy. If antitrust rules now appear to be insufficient to address issues raised by the new economy, it is partly because for the last 17 years, competition authorities have renounced on allowing them to sufficiently evolve. Indeed, if they had more often addressed complex cases, they could have developed a solid theory of harm about new issues and contributed to the improvement of the definition of antitrust infringements, in particular of the notion of abuse of dominant position. This would eventually have led to a gradual lightening of the competition authorities’ burden of proof – thus fostering efficiency.

This massive use of commitment procedure resulted in (1) compromising the deterrent effect of antitrust rules, (2) denying victims of anticompetitive behaviours possibility to engage in private litigation without having to prove the existence of an infringement and (3) depriving a substantial part of Commission’s decisions – including the most questionable – of satisfying judicial review, which reduced the role of judges and strengthened the discretionary power of competition authorities. The introduction of the NCT would complete this evolution. Indeed, commitment proceedings paved the way for market shaping experiments. If additional resources are not provided, the enforcement of the NCT would stem from a trade-off between the different tools available to the authorities. The NCT could either replace the commitment proceedings or contribute to the further reduction of the number of antitrust litigations. This is particularly regrettable as in the last years, the Commission has opened several antitrust litigations, which are still pending, against the GAFAM. In our opinion, the Commission may have doubted that the ECJ will agree with it on these complex cases and thus tried to find another way to tackle big tech companies. The NCT standard, which would undoubtedly remain very vague, would also give competition authorities the full discretion to implement an ambitious policy, based on (and largely guided by) a blind faith in economic “science”. This is the real crux of the issue. There is no disputing that competition law has a political raison d’être and foundation. It is also undeniable that there is a part of political choice in the enforcement of competition rules, but this one remains acceptable only provided that it is constrained by the respect of legal formalism and submitted to a satisfying judicial control.

The idea here is not to promote a laissez-fairepolicy, but to advocate for a public intervention respectful of the Rule of Law. It is the duty of the jurist to sound the alert: we must ensure that a legitimate discretion does not turn into an arbitrary power compromising our liberties.

This policy would go against the liberal idea at the foundation of the EU construction. Indeed, the German ordoliberal doctrine, which played a major role at the time of the adoption of the Treaty of Rome[47], aimed at the establishment of an “economic constitution” to prevent restrictions of competition coming from unfair behaviours of private undertakings but also from undue public interference in the free flow of trade[48]. The purpose was to enact intangible rules which would be able to guarantee the protection of the market process but also the preservation of the expectations of companies when facing coercion risks arising from choices dictated by immediate interests of the sovereign[49].

If up to now we could consider that, thanks to this “economic constitution”, the EU competition law enforcement was more sheltered from the influence of political and electoral cycles than in the US, it finally appears that the current interventionist inclination of the EU institutions could deeply impact the European system. It must be acknowledged that the NCT would be a most efficient tool as it would avoid necessity to enact new laws each time the authorities identify a “problem”. It would undoubtedly satisfy the authorities’ eagerness to find ready-made solutions for the sake of European consumers. If their main objective has become to be “acting fast”, in our view, the real emergency is to slow down.