To quote this paper: V. GIOVANNINI, “The abuse of a dominant position on alternating current”, Competition Forum, 2022, n° 0038,

What future for the abuse of dominant position? The future of European competition law is not only written at the legislative level, far from it. It is also being written in the courts! Indeed, if the trialogue between the representatives of the European Parliament, the European Council and the European Commission led to a provisional agreement on March 24, Mrs. Stéphanie Yon-Courtin, Member of the European Parliament (hereinafter MEP), declared during a recent symposium that the first decisions on the Digital Markets Act (hereinafter DMA) should not be taken before the beginning of 2024[i]… Therefore, it is competition law as we usually understand it, and abuse of dominant position in particular, that will continue for some time to regulate behavior in the digital economy where platforms and data practices are intimately linked. However, data practices are not unique to digital markets, far from it. Indeed, these practices are also well known in ‘‘bricks and mortar’’ markets like the energy sector markets as illustrated by a relatively recent case that gave rise to a preliminary ruling on the interpretation and application of Article 102 TFEU, specifically with respect to exclusionary abuses. The stakes of this case were high, even beyond those of the parties to the dispute, for the abuse of dominant position because, in the words of Advocate General Rantos, it would firstly allow the Court to confirm its recent case law, in which the Court has shown itself open to a less formal and more effects-based approach to assessing the abusive nature of a behaviour[ii]. Secondly, this case offered the opportunity for the Court to clarify whether certain principles emanating from this recent case law on exclusionary pricing practices, and in particular the ‘‘as efficient competitor’’ test (hereinafter AEC test), could be applied in the context of exclusionary non-pricing practices[iii]. Finally, this decision could prove useful in the future for assessing data-related conduct under Article 102 TFEU[iv]. Thus, the tone was set and this case was already shaping up to be an important one!

Context. The case at issue, which took place in the context of the liberalisation of the market for the retail supply of electricity in Italy, opposed the Autorità Garante della Concorrenza e del Mercato (hereinafter AGCM) to three companies belonging to the same group: the subsidiaries Servizio Elettrico Nazionale SpA (hereinafter SEN) and Enel Energia SpA (hereinafter EE), as well as their parent company ENEL SpA (hereinafter ENEL). The AGCM had fined these companies a staggering amount of more than EUR 93 million for abuse of a dominant position, due to the implementation by the latter of a strategy of foreclosure aimed, in essence, at making it more difficult for competitors to enter the liberalised market. More specifically, the companies were accused of having used in a discriminatory way the data relating to customers of the protected market, which SEN had before the liberalisation, in order to launch commercial offers to these customers and thus to make them transit to the other subsidiary of the group operating on the free market, namely EE.

These companies subsequently filed individual appeals against the AGCM’s decision before the Tribunale amministrativo regionale per il Lazio, i.e. the Lazio Regional Administrative Court. The latter, while finding the existence of an abuse of a dominant position, partially upheld the appeals of SEN and EE with respect to the duration of the alleged abuse and the criteria used for the calculation of the fine, thereby reducing the amount of the fine to slightly more than EUR 27.5 million. However, the court of first instance dismissed ENEL’s appeal in its entirety. The three companies have therefore appealed to the referring court, the Consiglio di Stato, i.e. the Italian Council of State. While the latter has no doubts about ENEL’s dominant position on the relevant market, it has doubts about the interpretation to be given to the concept of abuse in general, and to this ‘‘atypical’’ abuse in particular.
The referring court has therefore referred no less than five questions to the CJEU for a preliminary ruling which, in substance, concern respectively: the interests protected by Article 102 TFEU (I); the anticompetitive effects (II); the intent to restrict competition (III); the concept of ‘‘abuse’’ (IV); and the liability of the parent company for the conduct of its subsidiaries (V). We will examine each of these points in detail[v] before concluding briefly (VI).

I.              On the interests protected by Article 102 TFEU

From philosophical to evidential interest. The first question the Court answered concerned, in substance, whether Article 102 TFEU protects consumers or the competitive structure of the market (A) and, depending on the answer given, the implications that could be drawn from it in terms of evidence to characterise an exclusionary abuse (B).

A.    On the determination of the ultimate purpose of Article 102 TFEU

‘‘Consumer welfare’’[vi] as the ultimate purpose of Article 102 TFEU. In this respect, the Court states unequivocally that ‘‘the welfare of consumers, both intermediary and final, must be regarded as constituting the ultimate purpose justifying the intervention of competition law in order to repress the abuse of a dominant position in the internal market or in a substantial part of it’’[vii]. This change is somewhat surreptitious because, until now, the Court had always refused to enshrine the term ‘‘consumer well-being’’ in its case law. It had also expressly given priority to the structure of competition over the interests of consumers in two decisions: T-Mobile[viii] and GlaxoSmithKline[ix]. In the latter case, the Court even went so far as to qualify this reference as an ‘‘error of law’’ on the part of the Court of First Instance (hereinafter CFI)[x]. Indeed, stated as it is, this response of the Court goes completely against the commonly accepted idea, inherited from the ordoliberals, that European competition law is more focused on the preservation of competition itself than on consumers; the former benefiting the latter more than it is supposed to protect them[xi]. In reality, this formula appears less surprising if one puts it in perspective with the Opinion of Advocate General Rantos, for whom the two objectives appear to be indissociable[xii]. Indeed, according to him, and this is common sense, the protection of the market only makes sense if it aims at protecting consumers[xiii]. In any event, this solution has the merit of anchoring European competition law a little more firmly in the so-called ‘‘more economic’’ approach, the modernisation of which began at the end of the 2000s, and which gives pride of place to economic analysis and the ‘‘effects-based approach’’, to the detriment of the outdated and no less decried ‘‘form-based approach’’. It also has the merit of bringing European competition law closer to its American counterpart, for whom ‘‘consumer welfare’’ is, since Robert Bork[xiv], the ultimate purpose in this field. That said, American antitrust law could, in the future, evolve in the opposite direction of ours. Indeed, recently, Assistant Attorney General Jonathan Kanter of the DoJ’s Antitrust Division gave a speech in line with the ‘‘neo-Brandesian’’ movement, condemning the use of the consumer welfare standard in antitrust cases and proposing its replacement by more structural considerations[xv], described by some as the ‘‘Competitive Process Test’’[xvi]. Or the concrete translation of the proverb that the grass is always greener on the other side…

B.   On the consequences in terms of evidence

Sufficient evidence of harm to the competitive structure. As for the evidentiary consequences of this ‘‘philosophical’’ discussion, the Court simply recalls that the demonstration of exclusionary abuse requires the authority to establish only the existence of an harm to an effective competition structure, but not the existence of a direct harm to consumers[xvii]; the harm to the market structure being supposed to automatically harm consumers. In other words, as paradoxical as it may seem, the assertion that the ultimate purpose of Article 102 TFEU is consumer welfare has no evidentiary consequences. This seems unfortunate to us, as consumers are often removed from the competitive process. To understand this, we must first agree on what is meant by the term ‘‘consumers’’. In competition law, the term ‘‘consumers’’ is not limited to ‘‘final consumers’’, but also includes ‘‘intermediate consumers’’, as the present decision points out[xviii]. However, these ‘‘intermediate consumers’’ are in fact ‘‘customers’’, or ‘‘suppliers’’, or even ‘‘trading partners’’ of the dominant undertaking, i.e. competitors of the latter, whose interests are not necessarily aligned with those of the ‘‘final consumers’’[xix]. Unfortunately, the CJEU did not follow the Advocate General’s view that it was wrong to interpret the Court’s case law as suggesting that a practice that affects the competitive structure of the market automatically harms consumers, and recommended that ‘‘competition authorities to demonstrate that such an exclusionary practice impairs the effective competition structure, while at the same time verifying that it is also liable to cause actual or potential harm to consumers’’[xx]. On the other hand, the Court indicates, in a rather classical way, that it is possible for a dominant undertaking to escape the prohibition of Article 102 if it demonstrates that the anticompetitive effects of the practice in question are counterbalanced, or even surpassed, by positive effects for consumers in terms of price, choice, quality and innovation[xxi].

II.            On the anticompetitive exclusionary effects

Procedural and substantive aspects of anticompetitive exclusionary effects. The second question referred to the Court of Justice concerned anticompetitive exclusionary or foreclosure effects and involved, more specifically, both procedural (A) and substantive (B) aspects.

A.    Procedural aspects

The obligation to take into account the actual effects of an abusive practice. Procedurally, first of all, the CJEU had to answer, in short, the question of whether a competition authority is obliged to take into account the actual effects of an infringement. To this, the Court answers in the affirmative, reiterating in this regard its 2017 Intel case law, which requires competition authorities to take into account evidence submitted by dominant undertakings that demonstrates the absence of the capacity to foreclose[xxii]; this requirement being, moreover, in line with the right to be heard, which is a general principle of European Union law[xxiii] . In short, this means that the analysis must not ignore the reality of the context in which the conduct takes place. In other words, ‘‘all the relevant elements of the case’’[xxiv] must be taken into account, as the Court summarises it. Moreover, it follows that the procedural safeguards applied in the Intel case would not be limited to exclusivity rebates and, more broadly, loyalty rebates, but would apply to all exclusionary abuses, whether or not they are price-related[xxv].

B.   Substantive aspects

Sufficient evidence of ‘‘potential’’ anticompetitive effects. On the substantive level, the Court had to determine whether the actual effects of a practice can invalidate the analysis of its potential impact on competition. First and foremost, the Court clarified that the characterisation of the abuse does not require the demonstration of concrete anticompetitive effects. This is not very surprising, as it is traditionally taught that if the effects do not have to be purely hypothetical, in order to establish a causal link between the abuse and the anticompetitive effects, the latter do not have to be actual, but only probable[xxvi]. Indeed, although Article 102 TFEU is concerned with both actual and potential effects on competition, this is consistent both with the fact that the analysis carried out in this area is prospective in nature – it would indeed make no sense to wait for the exclusionary effects to occur at the risk of irreparably harming competition[xxvii] – and with the fact that the anticompetitive nature of conduct must be assessed ex ante, i.e., at the time when the conduct takes place; the latter rule is also consistent with the principle of legal certainty[xxviii].

Insufficient evidence of ‘‘actual’’ effects to exclude the application of Article 102 TFEU on its own… In this vein, the Court goes on to state that, despite the fact that a long period of time has elapsed since the conduct took place, the absence of concrete anticompetitive effects is not necessarily linked to the fact that the conduct does not have that capacity, but may result from other causes, such as changes in the relevant market since the conduct took place or the dominant undertaking’s inability to carry out the strategy which gave rise to such conduct[xxix]. Consequently, the Court considers that this evidence alone is not sufficient to rule out the application of Article 102 TFEU[xxx]. In this respect, two remarks can be made. On the one hand, the Court found its inspiration in the Commission’s guidance communication, which is sufficiently rare to be highlighted. To be more exact, this is a first for the CJEU! Beyond the symbolism, this reference to this text will undoubtedly put an end to the recurrent accusation that this text has disappeared from the radar[xxxi]. On the other hand, all this is reminiscent of the General Court (hereinafter GC)’s judgment in Google Shopping. In that case, Google, relying on the Servier cartel decision, argued that, if the practices complained of were truly anticompetitive, harmful effects on competition should have materialized since they had been implemented over many years[xxxii]. However, the GC did not accept this argument, as this would have been contrary to the principle that an abuse cannot be excluded because the practice in question did not have the desired result[xxxiii].

… But relevant evidence? That said, the Court states that evidence of the absence of concrete anticompetitive exclusionary effects may be an indication of the absence of anticompetitive exclusionary effects of the practice, likely to make it escape the prohibition of Article 102 TFEU[xxxiv], provided that it is supplemented by additional evidence[xxxv]. Although the decision does not provide many details as to the additional evidence that the dominant undertaking must provide, the combined reading of paragraphs 54 and 56 suggests that it is up to the dominant undertaking to show that this lack of concrete effects is, inter alia, not the consequence of a change in the relevant market or the inability of the dominant undertaking to carry out the strategy behind the conduct[xxxvi], but rather that the conduct was incapable of producing such effects[xxxvii]. In this respect, it is doubtful that companies in such a situation would be able to provide such difficult evidence, which is negative by definition and therefore constitutes a probatio diabolica[xxxviii].

III.          On the intent to restrict competition

The (ir)relevance of the intent to restrict competition. In the following question for a preliminary ruling, the referring court asked the Court of Justice, in essence, about the relevance of the dominant undertaking’s intent to restrict competition for the purpose of assessing whether conduct is abusive. The decision on this point could be summarised by the Latin expression ‘‘Nihil novi sub sole’’, meaning ‘‘nothing new under the sun’’. Indeed, the CJEU recalls that abuse is an objective concept and that, as such, its characterisation is based entirely on the capacity of the practice in question to produce anticompetitive exclusionary effects. Consequently, it is not necessary to establish the existence of an intent to restrict competition on the part of the dominant undertaking. However, although not sufficient on its own, the dominant undertaking’s intent to restrict competition may be taken into account as an indication of the existence of an abuse of dominance[xxxix].

IV.          On the concept of ‘‘abuse’’

Some preliminary elements on abuse. The next question for a preliminary ruling concerned the concept of ‘‘abuse’’. In this respect, the Court first recalled that this concept is based on an objective assessment of the conduct at issue and that, in so doing, it may be unlawful under Article 102 TFEU even though it is lawful in other branches of law[xl].
Secondly, the CJEU gives some basic explanations, namely: that the abusive nature of the practice depends decisively on its anticompetitive exclusionary effects; that Article 102 TFEU does not prevent the exercise of ‘‘competition on the merits’’ by the dominant undertaking and that, consequently, competitors less efficient than the dominant undertaking are not necessarily destined to remain on the market in question; but that, in any event, the dominant undertaking has a ‘‘special responsibility’’.
Third, the CJEU considers that a practice which either ‘‘has no economic interest other than that of eliminating its competitors’’ or which cannot be reproduced by a hypothetical competitor as efficient as the dominant undertaking on the relevant market, does not fall within the scope of ‘‘normal competition’’ or of ‘‘competition on the merits’’.
Fourth and last, the Court recalls, in a classic manner, that the dominant undertaking has, among the defenses allowing it to escape the prohibition laid down in Article 102 TFEU, to allege either an objective justification or efficiency gains.

Clarifications. Having said that, we will leave aside most of the above-mentioned points, which are, after all, classic, and concentrate on the third one in particular, which is undoubtedly the most interesting of the decision, but also the most incoherent of all. A few remarks about abuse in general (A) and replicability in particular (B) are enough to convince us of this.

A.    On abuse in general

The abnormality of demonstrating the abnormality of the conduct. What is immediately striking about the concept of ‘‘abuse’’ described in this decision is that it requires, in addition to the demonstration of an anticompetitive exclusionary effect, the demonstration that the practice in question is contrary to ‘‘competition on the merits’’[xli]. This is particularly curious, since nothing of the sort is normally required. Traditionally, the demonstration of the company’s dominant position is sufficient, insofar as it places a ‘‘special responsibility’’ on the company, which implies a sort of duty of exemplarity on its part to behave as if it were not in a position of economic power on the market in question. The key idea here is that a commercial practice that is lawful in normal times is not necessarily lawful in such a position. In other words, the dominant position of the company and the particular responsibility that it has as a result of this position, allow it in principle to dispense with the need to demonstrate the abnormal nature of the practice in question[xlii].
In any event, the Court goes on to discuss the two types of practices committed by a dominant undertaking which, in its view, contravene ‘‘competition on the merits’’.

The first branch of the alternative, ‘‘prima facie illegal’’ practices. With regard to the first branch of the alternative, although there is no ‘‘restriction of competition by object’’ in the case of abuse, as is the case for cartels, certain abusive practices, sometimes described as ‘‘prima facie illegal’’[xliii], are almost presumed to have anticompetitive effects. While the decision cites the Akzo case law and predatory pricing in this sense, these practices are not limited to this single example. Exclusivity rebates are another example.

The second branch of the alternative, the ‘‘replicability test’’. With regard to the second alternative, the Court enacts a so-called ‘‘replicability test’’ which, in its words, consists of examining ‘‘whether it is materially or rationally impossible for a hypothetical competitor that is equally effective, but not dominant, to imitate the practice at issue, for the purpose of determining whether the practice is based on means of competition on the merits’’, and which it praises as relevant both to exclusionary price abuses, such as loyalty rebates, selective pricing, predatory pricing or margin squeezes, and to exclusionary non-price abuses, such as refusal to supply goods or services[xliv].

The application of the ‘‘AEC test’’ to exclusionary non-price abuses. At this stage, we will simply note that this test is clearly related to the AEC test. This is not surprising as the AEC test is generally regarded as the economic translation of ‘‘competition on the merits’’. In this veine, the Court indicates that it is generally in the light of this test, which consists in focusing the analysis on a hypothetical competitor who is as efficient as the dominant undertaking and who would risk being eliminated by the practices in question[xlv], that the above-mentioned exclusionary pricing practices are assessed[xlvi]. In the same way, the doctrine generally considers that this AEC test only covers exclusionary pricing strategies[xlvii]. However, referring to its previous case law, Post Danmark II, the Court recalls that this is only one means among others to establish the abusive nature of a pricing practice[xlviii]. It is true that a multitude of tests are applicable depending on the practice. Nevertheless, the Court does not fail to emphasize the importance that is generally accorded to it[xlix]. Moreover, the reference in this decision to the AEC test in relation to an exclusionary practice, in this case a non-price practice, reflects the fact that the quest for a single test for exclusionary abuses of all kinds, i.e. price and non-price, is not just a pious wish. In this respect, it agrees with Advocate General Rantos who, in his Opinion, stressed the relevance of the ‘‘underlying logic’’ of the AEC test, including with regard to non-price practices[l].

B.   On replicability in particular

A replicability test already well known in French competition law. The decision submitted to this commentary refers to an ‘‘atypical’’[li]  exclusionary abuse to designate this practice, which consisted in using customer data of the protected market in a discriminatory manner in order to prevent the growth or diversification of competitors’ offerings. The ‘‘replicability’’ test enacted by the CJEU in this decision to assess the abusive nature of the practice is less innovative than it seems, as revealed by an examination of the decisional practice of the Autorité de la concurrence, i.e. the French competition authority (hereinafter FCA). Indeed, in 2010, the FCA investigated the crossed usage of client databases (hereinafter cross-selling)[lii]. It is a ‘‘widespread practice whereby a company uses information about its own customers, collected in a given market, to market to those same customers another product in a separate market’’[liii]. Cross-selling is not considered an anticompetitive practice in all cases. However, there are situations in which it is likely to lead to foreclosure effects, i.e. to erect barriers to entry into one or other of the markets concerned[liv]. In order for the abuse to be characterised as such, it is necessary to determine, on the one hand, whether the dominant undertaking has acquired the data in the context of competition on the merits and, on the other hand, whether the data can be reproduced by equally efficient competitors on the market[lv] ‘‘under reasonable financial conditions and within an acceptable timeframe’’[lvi]. Ultimately, cross-selling can be considered abusive provided that the data is inaccessible to competitors and cannot be reproduced by them.
The FCA has ruled on several occasions in this regard, concerning operators who have benefited from a legal monopoly in the context of their public service mission. To give just one example, the facts of which resemble those of the Enel decision – and which the Advocate General laconically cited in a footnote in his Opinion[lvii] – FCA sanctioned EDF for having used the data available to it in its capacity as historical electricity supplier to facilitate the marketing of the offers of its subsidiary EDF ENR, between November 2007 and April 2009[lviii]. This demonstrates once again, if it were necessary, how far ahead of its time FCA is.
In any event, and to return to the present decision, the introduction of new criteria and/or new tests within the case law of the Union is not a bad thing in itself. Indeed, from our point of view, it does not seem questionable since the list of abuses set out in Article 102 TFEU is not exhaustive, and competition authorities and judges are therefore forced to innovate in order to repress behaviours they may encounter that are outside the box.

The questionable nature of the desire to generalise this test to all exclusionary abuses. On the other hand, what we contest more is the fact that the CJEU does not seem to limit itself to applying it to the practice in question, but to generalise it to all exclusionary abuses. Indeed, several uncertainties and inconsistencies stand in the way of the generalisation of the ‘‘replicability’’ test as advocated by the CJEU in matters of abuse.

General inconsistencies with abuse. In general, it should be noted that this test is not relevant in all cases. Indeed, as Advocate General Rantos mentioned in his Opinion, there are situations in which conduct that can be reproduced by an equally efficient competitor does not necessarily fall within the scope of competition on the merits, such as misleading statements to public authorities. Conversely, there are cases in which conduct cannot be replicated, but nevertheless falls within the notion of competition on the merits, such as the refusal to supply an essential input to a customer who would not be able to pay adequately or R&D activities involving patents[lix].
In this vein, the relevance of this test is questionable in the case where the dominant firm has acquired unique assets or resources through its investments and innovations[lx]. Indeed, paragraph 78 of this decision implies that any dominant undertaking using the resources or means in its possession that give it a competitive advantage would be contrary to competition on the merits[lxi]. But isn’t this case, on the contrary, the purest manifestation of competition on the merits? This is why this test seems better adapted to the specific facts of the case in which the constitution of such a database is not based on the merits of the dominant company, but on its former legal monopoly status[lxii].
An umpteenth ambiguity about this test lies in what it relates to. From paragraph 78, it seems clear that the test seems to focus on practice[lxiii]. But paragraph 83 suggests that the test applies to the resources available to the dominant firm[lxiv].

Specific inconsistencies with refusal to supply. More specifically, and notwithstanding the Court’s assertion, this replicability test is completely at odds with European decisional practice and case law on refusal to supply.
To be convinced of this, it suffices to state, first of all, that from a theoretical point of view, the test specifically applied to refusal to supply is not one of ‘‘replicability’’, but of ‘‘indispensability’’. In essence, as the Bronner case law indicates, a product or service is considered indispensable if ‘‘there is no real or potential substitute’’[lxv]. To be precise, the search for ‘‘real’’ substitutes requires the existence of products or services that can be substituted for the product or service to which access is requested. Failing that, and as a second step, it is necessary to question the existence of ‘‘potential’’ substitutes, which consists in determining the capacity to duplicate this product or service – it being understood that it is when this is impossible or extremely difficult that it will be considered indispensable. In doing so, some have suggested that this test has two branches based on the concept of ‘‘substitutability’’ for the first and ‘‘replicability’’ for the second[lxvi], but in fact ‘‘the whole ‘indispensability’ test (…) is based on the sole concept of substitutability’’[lxvii]. As Professor Ibáñez Colomo summarises, ‘‘the question is not whether an equally efficient rival can build a replica of the dominant firm’s infrastructure, but whether the infrastructure is indispensable to compete on the relevant adjacent market’’[lxviii].
Second, and more pragmatically, paragraph 102 suggests that the anticompetitive nature of the practice would not have been questioned even if competitors had been able to obtain ‘‘from third parties files containing data on customers in the protected market’’[lxix]. This point is in complete contradiction with the Bronner case law, in which the Court considered that the assessment of the indispensability of the product or service in question requires that alternative solutions be taken into account, even if they are ‘‘less advantageous’’[lxx]. These explanations once again tend to show that the Enel decision would certainly gain in consistency if it were explained by the specificity of the facts of the case[lxxi].

V.             On the liability of the parent company for the conduct of its subsidiaries

The burden of proof and the duty to state reasons in relation to the presumption of decisive influence. The last question to which the Court of Justice gave a preliminary ruling concerned, in broad terms, the allocation of the burden of proof between the parties and the duty of a competition authority to state reasons in a situation where the undertaking abusing its dominant position belongs, together with several subsidiaries, to the same economic unit.

On the burden of proof. As regards, first, the burden of proof, the Court considers, as usual, that the holding of all or almost all of the subsidiary’s share capital by the parent company is sufficient to consider that it personally participated in the abusive practice, without it being necessary to provide additional evidence[lxxii].

On the duty to state reasons. As regards, secondly, the duty to state reasons, the Court recalls, first of all, that this is a general principle of Union law[lxxiii] that must be satisfied in order to ensure respect for the right to an effective remedy[lxxiv]. In order to do so, it is up to a competition authority to adequately explain the reasons why the elements alleged to rebut the presumption of decisive influence were not sufficient[lxxv], without the authority being required to take a position on each of the elements put forward by the parent company[lxxvi]. In this case, these elements were not considered sufficient to rebut the presumption[lxxvii].

VI.          Conclusion

Enel as a ‘‘landmark case’’? In conclusion, we mentioned at the beginning of this commentary that the expectations for this case were such that it had all the makings of an important case. But is it a ‘‘landmark case’’? In this regard, Professor Bosco, speaking about the CJEU’s Intel decision of September 6, 2017, quoting Professor Lequette, explained in substance that what makes a ‘‘landmark case’’ are, on the one hand, its ability to structure the matter, and, on the other hand, its longevity over time[lxxviii]. As far as the time factor is concerned, there is still too little hindsight to say whether or not the Enel decision is a ‘‘landmark case’’. On the other hand, the inconsistencies, uncertainties and other ambiguities highlighted by this commentary suggest that the Enel decision does not tick all the boxes that make it a ‘‘landmark case’’. 

Vincent Giovannini

[i] Stéphanie Yon-Courtin, ‘Présentation du texte’ (Digital market act : Droit de la concurrence et géants du numérique, Paris Nanterre University, April 2022).
[ii] Case C-377/20 Enel, Opinion of AG Rantos, para 6.
[iii] Ibid.
[iv] Ibid, para 7.
[v] To be precise, these preliminary questions were not put to the Court in that order. However, for the sake of clarity, we prefer to follow the order in which the decision considered these questions.
[vi] The decision has not yet been published in English, but in the Advocate General’s Opinion, the term ‘‘bien-être des consommateurs’’ is translated as ‘‘well-being of consumers’’, whereas the usual term is ‘‘consumer welfare’’. However, as revealed by one author, these terms do not have the same meaning: ‘‘‘consumer welfare’ is a technical term for the purposes of economics, while well-being is not’’. In reality, the wording is mostly due to translation. Indeed, the term ‘‘bien-être du consommateur’’ is the French economic term for ‘‘consumer welfare’’. This is why the use of the term ‘‘consumer welfare’’ is probably preferable. See Victoria Daskalova, ‘Consumer Welfare in EU Competition Law: What Is It (Not) About?’ [2015] 11(1) The Competition Law Review, 150.
[vii] Enel (n 3), para 46.
[viii] Case C-8/08 T-Mobile Netherlands [2009] ECR I-04529, para 38.
[ix] Case C-501/06 P GlaxoSmithKline Services Unlimited [2009] ECR I-09291, para 63.
[x] Ibid, para 64.
[xi] Louis Vogel, Traité de droit économique – Tome 1/1 – Droit de la concurrence – Droit européen, (3rd edn Bruylant 2020), 18.
[xii] Opinion of AG Rantos in Enel (n 2), para 96.
[xiii] Ibid.
[xiv] Robert H. Bork, The Antitrust Paradox: A Policy at War With Itself (1rst edn Basic Books, 1978).
[xv] Jonathan Kanter, ‘Assistant Attorney General Jonathan Kanter Delivers Remarks at New York City Bar Association’s Milton Handler Lecture’ (The 2022 Milton Handler Lecture, New York, 18 May 2022) accessed 10 June 2022.
[xvi] Einer Elhauge, ‘Should The Competitive Process Test Replace The Consumer Welfare Standard?’ ProMarket (24 May 2022) accessed 10 June 2022.
[xvii] Enel (n 3), para 47.
[xviii] Ibid, para 46.
[xix] Pinar Akman, ‘‘‘Consumer’’ versus ‘‘Customer’’: the Devil in the Detail’ [2008]   accessed 10 June 2022, 10-11.
[xx] Opinion of AG Rantos in Enel (n 2), para 108.
[xxi] Enel (n 3), paras 47-48.
[xxii] Case C-413/14 P Intel / Commission ECLI:EU:C:2017:632, paras 138-140 ; Enel (n 3), para 51.
[xxiii] Ibid, para 52.
[xxiv] Ibid.
[xxv] Dimitrios Katsifis, ‘The judgment of the Court of Justice in Servizio (C-377/20) – Part I’ (The Platform Law Blog, 23 May 2022) accessed 10 June 2022.
[xxvi] Catherine Prieto and David Bosco, Droit européen de la concurrence – Ententes et abus de position dominante (Bruylant 2013), 887-888.
[xxvii] Nicolas Petit, Droit européen de la concurrence (3rd edn 2020), 412.
[xxviii] Opinion of AG Rantos in Enel (n 2), para 114.
[xxix] Enel (n 3), para 54.
[xxx] Ibid, para 55.
[xxxi] Recently, this communication was cited by the General Court in the Intel renvoi judgment and in the Opinion of Advocate General Rantos. On the possible reasons for the reference to this instrument, see Pablo Ibáñez Colomo, ‘The (growing) role of the Guidance Paper on exclusionary abuses in the case law: the legal and the non-legal’ (Chillin’Competition, 9 February 2022) accessed 10 June 2022.
[xxxii] Case T-612/17 Google Shopping (GC, 10 November 2021), para 426.
[xxxiii] Ibid, para 442. In the same vein, see Katsifis (n 24).
[xxxiv] GC (n 31), para 440.
[xxxv] Enel (n 3), para 56.
[xxxvi] Ibid, para 54.
[xxxvii] Ibid, para 56.
[xxxviii] Katsifis (n 24).
[xxxix] Enel (n 3), paras 59-64.
[xl] Ibid, para 67.
[xli] Ibid, para 103.
[xlii] Pablo Ibáñez Colomo, ‘On Case C-377/20, Servizio Elettrico Nazionale (I): overaching framework or case-specific judgment?’ (Chillin’Competition, 16 May 2022) <> accessed 10 June 2022. However, other authors have suggested that, since in the Google Shopping case the Commission had to show that Google’s behaviour deviates from competition on the merits in addition to establishing at least potential anticompetitive effects, the AEC test should be replaced, where it is not relevant, by a focus on whether or not the practice in question falls within the scope of ‘‘competition on the merits’’. See Dimitrios Katsifis, ‘Servizio, competition on the merits, and Google Shopping’ (The Platform Law Blog, 7 June 2022) <> accessed 10 June 2022.
[xliii] Pablo Ibáñez Colomo, ‘Legal Tests in EU Competition Law: Taxonomy and Operation’ [2019] accessed 10 June 2022.
[xliv] Enel (n 3), para 79.
[xlv] Prieto and Bosco (n 25), 883.
[xlvi] Enel (n 3), para 80.
[xlvii] Prieto and Bosco (n 25), 884.
[xlviii] Enel (n 3), para 81.
[xlix] Ibid, para 82.
[l] Opinion of AG Rantos in Enel (n 2), para 73.
[li] Enel (n 3), para 24 and 27.
[lii] FCA, opinion 10-A-13 of 14 June 2010 relative to the crossed usage of client databases.
[liii] Ibid, para 3.
[liv] Ibid, para 20.
[lv] Ibid, para 19.
[lvi] FCA, decision 14-MC-02 of 9 September 2014 relative to a request for interim measures submitted by Direct Energie in the gas and electricity sectors, para 146.
[lvii] Opinion of AG Rantos in Enel (n 2), fn 4.
[lviii] FCA, decision 13-D-20 of 17 December 2013 concerning the practices implemented by EDF in the photovoltaic solar power sector.
[lix] Ibid, fn 52.
[lx] Pablo Ibáñez Colomo, ‘On Case C-377/20, Servizio Elettrico Nazionale (II): does the replicability test really work?’ (Chillin’Competition, 18 May 2022) accessed 10 June 2022.
[lxi] This same paragraph also seems to suggest, in fine, the establishment of a genuine causal link between the dominant position and the abuse, whereas European decisional practice and case law currently require, at the very least, only a ‘‘connectedness’’ between the two. See:  Petit (n 26), 423-425.
[lxii] Ibáñez Colomo (n 59).
[lxiii] Enel (n 3), para 78.
[lxiv] Ibid, para 83.
[lxv] Case C-7/97 Oscar Bronner GmbH & Co. KG contre Mediaprint Zeitungs- und Zeitschriftenverlag GmbH & Co. KG, Mediaprint Zeitungsvertriebsgesellschaft mbH & Co. KG and Mediaprint Anzeigengesellschaft mbH & Co. KG [1998] ECR I-07791, para 41.
[lxvi] Nicolas Petit, ‘Theories of Self-Preferencing Under Article 102 TFEU: A Reply to Bo Vesterdorf’,  [2015] accessed 10 June 2022, 13.
[lxvii] Pinar Akman, ‘The Theory of Abuse in Google Search: A Positive and Normative Assessment Under EU Competition Law’ [2017] 2017(2)  Journal of Law, Technology & Policy, 314.
[lxviii] Ibáñez Colomo (n 59).
[lxix] Enel (n 3), para 102.
[lxx] Bronner (n 63), para 43.
[lxxi] Ibáñez Colomo (n 59).
[lxxii] Enel (n 3), para 114.
[lxxiii] Ibid, para 115.
[lxxiv] Ibid, para 116.
[lxxv] Ibid, para 118.
[lxxvi] Ibid, para 119.
[lxxvii] Ibid, para 122.
[lxxviii] David Bosco, ‘L’arrêt Intel, un « grand arrêt » ?’ (2017) 11 Contrats Concurrence Consommation.


Submit a Comment

Your email address will not be published. Required fields are marked *