On December 9, 2021, Advocate General (“AG”) Rantos delivered his opinion on the questions referred to the Court of Justice (“ECJ”) by the Italian Consiglio di Stato in case Servizio Elettrico Nazionale.[1] The Consiglio di Stato is seeking clarification of certain aspects of the concept of “abuse” under Article 102 TFEU.[2] The alleged abuse concerns the discriminatory use of customer data to avoid losing those customers to rivals in a newly liberalized market. The case gives the Court the opportunity to address the issue of abusive conduct based on a competitive advantage lawfully inherited by a former statutory monopolist from that very same position. The case also touches upon the value and potential replicability of a set of data as a means of competition. Above all, it allowed the AG to revisit the Court’s existing case-law on exclusionary non-price abuses, which he proposes to clarify and categorize for future guidance.
Background
Upon the liberalization of the retail market for electricity distribution in Italy, Enel group’s (the former statutory monopolist) activities were unbundled, among others, into Enel Energia SpA (“EE”) (active in the liberalized market) and Servizio Elettrico Nazionale SpA (“SEN”) (active in the so-called protected market).[3] In 2017, following a complaint by an industry association, the Italian antitrust authority (“AGCM”) opened an investigation into Enel SpA (“Enel”), SEN and EE.
In December 2018, the AGCM adopted its final decision in which it held that EE and SEN, coordinated by their parent company Enel, had abused their dominant position in the energy distribution sector in Italy in violation of Article 102 TFEU. The incriminated conduct lasted more than five years from 2012 until 2017. The AGCM was concerned that SEN had allowed EE to use a non-public database of SEN customers to target those customers with commercial offers by EE to incentivize customers to switch to EE and avoid a mass departure to rival distributors following the opening of the protected market. At the same time, SEN restricted access by third parties to its customer list.[4] The AGCM held that the discriminatory use of SEN customer data constituted an abuse by the Enel group of its dominant position. Following a partially unsuccessful challenge in first instance, the parties appealed to the Consiglio di Stato, which referred five questions to the Court of Justice:
- What are the elements necessary for finding an exclusionary abuse of a dominant position?
- What is the underlying purpose of Article 102 TFEU, consumer protection or protection of the competitive structure of the market (in an interest to identify the evidence to be taken into account for the assessment of whether a conduct is abusive)?
- Can the dominant undertaking rely on ex post evidence to prove the absence of anticompetitive effects?
- Is anticompetitive intent relevant in the finding of an abuse?
- Does belonging to the same corporate group justify holding liable a company that did not take part in the prohibited behavior?
AG Rantos’ proposed response to the Consiglio di Stato’s first question will be particularly noteworthy for assessments of exclusionary non-pricing abuses, if endorsed by the Court.
Clarification of the concept of ‘competition on the merits’ in exclusionary cases
The AG seeks to systemize the concept of abuse in exclusionary cases based on previous case-law.[5] By the first question, the referring Consiglio di Stato asks whether a dominant undertaking’s business practice that is considered ‘perfectly legal’ outside the context of competition law can be classified as an abuse under Article 102 TFEU solely based on its (potentially) restrictive effect on competition, or whether such finding requires an additional element of unlawfulness.
AG Rantos confirms that conduct can amount to an abuse under Article 102 TFEU if it is capable of causing (potential) harm to competition.[6] A different reading of Article 102 TFEU would risk under- enforcement in cases where anticompetitive conduct does not infringe any other laws.[7] Yet, a restrictive effect, whether actual or potential, is not necessarily anticompetitive, as expressed by the “as efficient competitor” (“AEC”) test.[8] AG Rantos confirms that conduct that equally efficient rivals can replicate is generally not abusive.[9] This is consistent with the general premise that less efficient rivals may be driven out of the market as a natural outcome of competition ‘on the merits,’ and are therefore not meant to be protected by Article 102 TFEU.[10]
AG Rantos acknowledges the limits of the AEC test in this case, as it would be materially impossible for competitors to replicate Enel’s exact same strategy before the liberalization of the market.[11] Nonetheless, AG Rantos suggests applying the “underlying logic” of the AEC test “which seeks, in essence, to estimate whether a dominant undertaking was in a position in which it was able to foresee, on the basis of data known to it, whether a competitor could, despite the conduct in question, have stayed competitive on the market operating in an economically viable way.”[12]
In this context, AG Rantos seeks to clarify the meaning of ‘competition on the merits’ acknowledging that the Court’s inconsistent terminology in previous cases may have led to misunderstandings.[13] In his opinion, resorting to an assessment of the ‘normality’ of the conduct may be confusing. Antitrust authorities should rather focus on analyzing the effects of the conduct, in accordance with the case law, in particular Intel.[14]
The assessment of the effects of the behavior and whether the latter consisted of ‘competition on the merits’ must not be carried out in the abstract. The analysis should consider the relevant economic and legal context of the behavior. For practical guidance, AG Rantos notes that ‘competition on the merits’ is generally (i) based on economic or objective reasons;[15] (ii) benefits consumers with lower prices, improved quality and/or larger choice of new or better products and services; and (iii) preserves the ability of competitors to imitate the conduct of the dominant undertaking.
Conversely, the manifest deviation from normal commercial practice can be a relevant indicator of a conduct’s abusive nature.[16] Ultimately, the conduct at issue must be assessed in close correlation with the “special responsibility” that rests on dominant firms not to allow their conduct to impair effective competition in the market. This “special responsibility” however does not prevent dominant firms from protecting their legitimate commercial interests.[17]
In practice, to determine whether the Enel group’s conduct was abusive AG Rantos considers the Consiglio di Stato should assess (i) the competitive significance of Enel’s lists, (ii) to what extent access to these lists was discriminatory against rivals and in favor of EE, e.g., in view of maintaining the contract with the supplier; and (iii) finally, and irrespective of the previous point, the replicability or the availability of similar lists (in terms of price and content) in the Italian market. If the commercial value of competitors’ lists is similar to the SEN lists, then no potential foreclosure effect of this strategy would be attributable to Enel.[18]
Consumer welfare vs protection of a competitive market structure
The second question referred to the Court of Justice asks whether the aim of Article 102 TFEU is to maximize consumer protection or rather preserve the market’s competitive structure. The AG sees both objectives as closely linked: the ultimate aim of increasing consumer welfare encompasses their protection not only against direct harm (such as higher prices or diminished choice) but also against indirect harm caused by an alteration of the market structure that would otherwise have guaranteed effective competition.[19]
AG Rantos recalls that there is no room for the application of Article 102 TFEU absent consumer harm,[20] and hence, no reason to protect the competitive market structure in the abstract where it does not contribute to the ultimate objective— and, indeed, the “leitmotiv of competition law more generally”—of consumer welfare protection.[21]
As a result, if endorsed by the Court, competition authorities relying on theories of indirect consumer harm will need to demonstrate that an exclusionary practice not only impairs the competitive market structure but also causes actual or potential harm to consumers.[22]
Ex-post evidence of absence of actual effects is a relevant factor in assessing whether alleged foreclosure could have actual effects
The Consiglio di Stato’s third question will allow the Court to discuss the relevance of ex post evidence that undertakings may rely on to prove a lack of actual anticompetitive effects. It is established that competition authorities do not need to prove actual anticompetitive effects in the market to establish an infringement of Article 102 TFEU.[23] As AG Rantos notes, any such requirement would be contrary to the ratio of Article 102 TFEU, which is preventive and forward-looking in nature, and which would be thwarted if competition authorities would have to wait for the detrimental effects to occur in the market.[24]
Nonetheless, and especially when the behavior under scrutiny is dated, elements advanced by the dominant undertaking during the administrative procedure to evidence the absence of actual foreclosure can be a relevant circumstance for the assessment of the conduct’s capacity to restrict competition in cases where the competition authority’s theory of harm is based on actual, and not just potential, anticompetitive effects.[25]
The third question offers AG Rantos the opportunity to address the relevant threshold of effects to find an abuse. Is it enough to demonstrate that exclusionary effects appear more likely than not? Or should the likelihood of the restrictive effect be considerably more than a mere possibility? As did Judge Wahl, AG Rantos recognizes that the terms ‘probability,’ ‘likelihood,’ or ‘capability’ to restrict competition are used interchangeably.[26] AG Rantos’ further considers that the degree of probability will depend on the legal, economic, and factual context of each case and on “hard facts” e.g., the duration and the coverage of the practice. Accordingly, the type of abuse at hand would impact the threshold of effects required: the threshold for below-cost pricing cases would be lower than that of margin squeezing cases.[27]
In response to the Consiglio di Stato’s fourth and fifth questions, the AG recalls, first, that abusive foreclosure practices do not require demonstration of a specific intent but rather of “the economic logic of the behavior in question, as it objectively results from the characteristics of that behavior and its context,”[28] and second, that according to the settled case law of the Court, a parent company that holds 100% of the capital in its subsidiaries is presumed to exert decisive influence on its subsidiaries, unless it adduces sufficient evidence that the subsidiary acted independently in the market.[29]
Conclusion
The Opinion is a welcome advancement of a more systematic approach to exclusionary non-pricing abuses under Article 102 TFEU. Nonetheless, it provides little innovation. AG Rantos’ summary of the existing case-law is helpful but remains retrospective in nature. Should the Court engage with such fundamental discussions as regards the ultimate goal of Article 102 TFEU, it may be useful to maintain flexibility to include, for example, aspects of environmental and social sustainability that are at the center stage of current antitrust debates and that might soon emerge as serious sources of (indirect) consumer harm, if not as fundamental goals in themselves.
It remains to be seen whether and to what extent the Court of Justice will endorse the Opinion. The outcome may have repercussions beyond recently liberalized sectors, including for the data-related conduct at issue.
Editors: Conor Opdebeeck-Wilson and Thorsten Schiffer
[1] Servizio Elettrico Nazionale and Others v. Autorità Garante della Concorrenza e del Mercato (Case 377-20), Opinion of Advocate General Rantos, EU:C:2921:998 (“AG Rantos Opinion”).
[2] Treaty on the Functioning of the European Union, OJ 2012, C 326, paras. 47–390.
[3] The liberalization of the Italian energy sector followed a two-step process. In a first step, Italy opened electricity distribution to competition for so-called “eligible” customers (mainly larger enterprises), while smaller customers (SMEs and individuals), referred to as “captive” customers, continued to benefit from a protected service (“servizio di maggior tutela”) under which prices and conditions remained strictly regulated. In a second step, the Italian legislator opened the market also for captive customers, which, after several postponements, was put into effect as from January 2021 (for SMEs) and as from January 2022 (for households).
[4] SEN requested two separate consents to its customers: a first consent to receiving commercial offers from undertakings of the Enel group, and a second one to receiving commercial offers by third-parties. In practice, clients more often refused consent to sharing their information with third parties but consented to SEN using their data assuming this was necessary for proper management of their contractual relation with SEN.
[5] TeliaSonera (Case C-52/09), EU:C:2011:83, Post Danmark I (Case C-209/10), EU:C:2012:172, Post Danmark II (Case C-23/14), EU:C:2015:651, Intel (Case C-413/14), EU:C:2017:632, Generics (UK) (Case C-307/18), EU:2020:52.
[6] AG Rantos Opinion, paras. 32–38.
[7] Whether the opposite, i.e. an abuse under Article 102 TFEU due to an infringement of other laws (in that case the General Data Protection Regulation), is true is currently pending before the Court of Justice in the Facebook case, see Bundeskartellamt decision B6-22/15 of February 6, 2019, as reported in our February 2019 German Competition Law Newsletter and the referral to the Court of Justice by the Regional Court of Appeals (Oberlandesgericht) Düsseldorf on April 22, 2021,Facebook (Case C-252/21).
[8] AG Rantos Opinion, paras. 43–45.
[9] AG Rantos Opinion, para. 69. In footnote 52, the AG nonetheless admits that this is not an absolute rule as, in particular circumstances, a behavior can be replicated by an equally efficient competitor but still not be considered as competition on the merits (e.g., misleading statements to public authorities as was the case in AstraZeneca (Case C-457/10 P)).
[10] It is the that inefficient competitors.
[11] The Court’s case-law has established the AEC test as an appropriate test for price-related abuses, showing obvious limits to the AEC test for instance in cases where the structure of the market makes the emergence of an as-efficient competitor practically impossible, e.g., due to a (former) statutory monopoly and high barriers to entry, see, e.g., Post Danmark II, paras. 57–61.
[12] AG Rantos Opinion, paras. 73–74.
[13] AG Rantos Opinion, para. 53.
[14] AG Rantos Opinion, para. 54.
[15] AG Rantos Opinion, para. 62; TeliaSonera, para. 88; Akzo (Case C-550/07), EU:C:2010:512, para. 71; AstraZeneca (Case C-457/10), EU:C:2012:770, para. 130.
[16] AG Rantos Opinion, paras. 61 and 128.
[17] AG Rantos Opinion, paras. 58–60. United Brands (Case C-27/76), para. 189.
[18] AG Rantos Opinion, paras. 75–81.
[19] AG Rantos Opinion, para. 96.
[20] As clarified by the Court in Post Danmark I and Intel.
[21] AG Rantos Opinion, para. 99.
[22] AG Rantos Opinion, para. 108.
[23] AG Rantos Opinion, paras. 110–113.
[24] AG Rantos Opinion, para. 110.
[25] AG Rantos Opinion, para. 119. The absence of actual effects can also be a mitigating factor in the calculation of the amount of the fine.
[26] AG Rantos Opinion, paras. 113 and 118.
[27] In Akzo, the Court suggested that “plausibility” is sufficient (Case C-62/86, paras. 71–74), whereas the Court applies a higher threshold in Deutsche Telekom (Case C-280/08, paras. 250–254).
[28] AG Rantos Opinion, para. 130. The existence of this intention can be taken into account as a simple factual circumstance and does not entail any reversal of the burden of proof.
[29] AG Rantos Opinion, paras. 149–155.