Competition on the merits in liberalised electricity markets: AG Rantos’ Opinion in Servizio Elettrico Nazionale

The Opinion of Advocate General Rantos in Case C 377/20 Servizio Elettrico Nazionale (available in English here) concerns the use of customer data (legitimately collected by the ENEL group at the time it enjoyed a legal monopoly) to target offers to those same customers following Italian energy market liberalisation. The AG’s Opinion on when such data use might be considered abusive under Article 102 TFEU has been the subject of well-reputed competition law and policy blogs. This is not surprising. In this detailed and thorough Opinion, AG Rantos goes to the core of Article 102 TFEU: what is an abuse, which type of conduct is abusive, when an abusive conduct takes place, what are the goals pursued by the prohibition of abuse of dominance, and who is liable within a corporate group.
The Opinion sheds an important light on non-price-related competition issues that can arise in partially liberalised markets where some market segments are open to competition, and other segments of the market are still heavily regulated. The electricity and gas markets are a case in point. Data is a competitive resource and access to formerly ‘captive customers’ in liberalised markets is critical for competition.
In this note we focus on the wider relevance of the Opinion for highly regulated and liberalised markets. Servizio Elettrico provides an opportunity to reflect on competitive risks associated with network industries.

1. Background

On market liberalisation the former State monopoly – Enel – was separated in three companies: Enel Energia (EE), electricity supplier for the deregulated market; Servizio Elettrico Nazionale (SEN), supplier of the “enhanced protection service”; and e-distribuzione, concessionaire for electricity distribution activities. The Italian Competition Authority (ICA) fined the Enel Group for illegally using commercially sensitive information to transfer customers from SEN to EE. In particular, SEN obtained the consent of users to receive commercial offers in a discriminatory manner, sending first a request for initial authorisation for the processing of personal data by companies within the Enel Group and then a second request to the benefit of third party operators. Customers usually gave the initial consent, while they usually refused the second consent so that consent for third-party operators was only given in 30% of cases. The ICA considered this a “sui generis abuse” and imposed fines on EE, SEN and the parent undertaking. Those companies appealed to the Italian courts and the Council of State referred a number of questions for a preliminary ruling to the ECJ.

2. The Opinion in a nutshell

The Opinion is likely to become a mandatory read for the study of the interaction between EU competition law and sector-specific regulation law. It is thorough and rich in references to case law. For the most part, it does not invite controversy. The Opinion underlines the importance of economic analysis in abuse cases and the ‘more economic approach’ to competition law. It shows a clear preference for what is known as an ‘effect-based’ rather than a ‘form-based ‘ approach to competition issues (para. 55) [1].

3. On the notion of abuse under Article 102 TFEU

First, AG Rantos confirms that conduct that might be legal under a different field of the law, for example, sectoral energy regulation might still be classified as abusive (paras 32 – 81). However, to be abusive, a pure exclusionary (actual or potential) effect in the market by and in itself is insufficient. That a practice drives a competitor from the market does not make it anticompetitive. The conduct may entail a risk of foreclosure (based on the merits and as a response to competition on the merits) or an anticompetitive foreclosure. Only the latter is anticompetitive. Competitive foreclosure may be justified as it benefits consumers in the form of price, choice, quality, or innovation (para.44). Thus, abnormal competition (i.e., not economically justified) is needed for abuse to take place.
For AG Rantos, this assessment does not need to be separated from whether the conduct has a restrictive effect (para. 48). In his words, “the ability of a practice to produce an anticompetitive effect, on the one hand, and the use of means that do not come within the scope of normal competition, on the other, are conditions that come under the same assessment” to determine if a practice is abusive (para. 50). Normal competition has several synonyms: fair competition, competition on the merits, and competition based on quality. AG Rantos suggests using ‘competition on the merits’. We agree with this terminological proposal. In his view, this is an abstract concept that ought to be assessed on a case-by-case basis (para. 55). However, common elements exist. In particular, dominant players have a special responsibility, and, therefore, their conduct is subject to much tighter scrutiny than non-dominant undertakings. The form of the behaviour is not relevant to classify it as abusive (para. 61). It ought to restrict or be capable of restricting competition. Conduct not economically justified is likely to not be ‘on the merits’. Further, competition on the merits will lead to lower prices for consumers, better quality, and wider variety and choice (para. 63).
The Opinion also suggests that the ‘as-efficient competitor’ test – usually used in abusive pricing cases might be a good tool to determine if the conduct results from competition on the merits (para. 44 and paras. 66 to 74). If equally efficient competitors can replicate the conduct, it would not lead to abusive foreclosure (para. 69). Thus, a practice that is replicable by competitors in an economically viable way would not be conduct that leads to anticompetitive foreclosure.

4. Goals pursued by Article 102 TFEU

Second, the Opinion discusses what values are protected by Article 102 TFEU. He analyses in-depth whether the goal is to protect consumers, protect the competition process, or both. Here, the Opinion summarises the history of EU competition law [2], and emphasizes that EU competition law does not protect (less efficient) competitors (para. 93).
AG Rantos merges the two objectives into one. Protecting the competitive structure is not a separate goal from safeguarding the well-being of consumers. Protection of the market, however, is only relevant if this is to prevent harm to consumers (para. 96). AG Rantos also stresses that harm to consumers is “an essential element for the application of Article 102 TFEU”. This is not controversial but is more nuanced than it seems, as consumers are not always easy to identify. Also, this concept is broader than the end consumer [3]. In black and white terms, AG stresses that

“Article 102 TFEU is aimed at maximizing consumer well-being, inter alia by protecting the competitive structure of the market. That protection may, therefore, indeed be an objective pursued by Article 102 TFEU, not independently but only when, in a specific case, it contributes to the ultimate objective of the (direct or indirect) protection of consumers.” (para. 100).

Additionally, the Opinion is illustrative of the evidence of harm needed to trigger the application of Article 102 TFEU. AG clarifies that the case-law has consistently shown that Article 102 TFEU is triggered not only in cases of direct consumer harm but also in the light of conduct “detrimental to them through their impact on an effective competition structure” (para. 104). Thus, the evidence needs to show whether the conduct has had a restrictive effect on competition and not (only) if there has been harm to consumers.

5. Evidence and intent

Third, the AG deals with the notion of evidence to determine the existence of abuse. The Opinion also clarifies that there is no obligation to show actual effects for a conduct to be abusive – potential harm is sufficient, but as the case law confirms, it cannot be purely hypothetical. The key factor is to demonstrate that the conduct is capable of restricting competition by excluding as efficient competitors. An absence of effects in the market may not necessarily mean that the conduct was not abusive (para. 117 and ss). Conduct might still be capable of restricting competition. However, the absence of such effects over a sufficient period may make it reasonable to conclude that the “practice was not even theoretically capable of harming competitors” (para. 119). Absence of effects may instead be used to assess the gravity of the conduct and reduce the size of the fines to be imposed (para. 120). So – a yes but no, and be careful, regardless.
Fourth, the Opinion addresses whether intent plays a role in abuse cases. AG Rantos clarifies that abuse is an objective concept that may be triggered absent any fault (para. 127), so evidence of intent is not necessary. However, evidence of intent may be a factor to help determine if abuse has taken place. For AG Rantos, it is economic evidence that matters most as “abusive exclusionary practices are seldom established by demonstrating a specific subjective intention on the part of the dominant undertaking to restrict competition, but on the basis of the economic rationale underlying the conduct in question as it appears objectively from the characteristics of the conduct and its context” (para. 130). Determining that there has been intent does not reverse the burden of the proof. It is for competition authorities to prove the abusive nature of the conduct (para. 139).

6. Why this case in this sector and the importance of data

AG Rantos highlights the importance of data in a liberalisation process (para. 76). Customer data was used to avoid losing clients once the regulated market disappeared. In his view, using data to contact customers is nothing but defensive conduct by the ENEL Group in light of the market liberalisation. This is an entirely normal practice, and one that its competitors would also engage in (para. 67). We agree with this. Companies strive to keep clients to obtain revenue and profits, even former monopolists. Implementing business strategies, including the use of client data to retain a client base, would not in itself constitute abuse (para. 66).
However, the Opinion also stresses that not all measures might be a legitimate means to keep customers. Competitive advantages (such as a previous statutory monopoly?) must not be used with the effect of foreclosing rivals as efficient as the dominant undertaking (para. 69). The fact that a practice removes a competitor from the market is not sufficient for it to be abusive, even in newly liberalised markets.
The Court in Luxembourg and the referring court in Italy have a challenge ahead: access to captive/previous customers is an advantage derived from the previous statutory monopoly. AG Rantos argues that new providers can contact customers, and that electricity providers can find alternatives to gather client data. Furthermore, being captive does not mean that customers are tied to the previous and monopolist service provider (para. 76). The Opinion disregards the risk that consumers are often not motivated enough to switch suppliers.
This case highlights the importance of competition law as an additional tool to ensure competition in network-intensive sectors. Competition must not only be made possible (this being the main objective of regulatory law); it must be conducted on the merits (and that assessment may well lie outside of the scope of regulatory law). However, this is the model adopted in Europe for the co-existence and co-application of sectoral regulation and competition law [4]. This ‘division of tasks’ is all the more important when the conduct that takes place is novel – in this case, the use of personal data to ‘secure’ clients – and/or when it does not relate to pricing.

7. More than a special responsibility and the notion of equal footing

In para. 60, AG Rantos stresses that ‘inheriting’ a dominant position does not exclude the need to assume a special responsibility in the market. This is uncontroversial. The Opinion emphasizes that the same applies to operators with a public service obligation (electricity providers, transmission operators and energy distributors). Could we question whether, in the case of an inherited monopoly ‘graced’ with a pre-competition advantage (here the previously captive customers), it is expected to discharge an even higher level of special responsibility? AG Rantos’ Opinion points to the contrary. The AG stresses that even incumbents:

“from the moment they are subject to free competition should seek to maximise their profits inter alia by means of retaining their customer base. Indeed, winning customers is an essential element of normal competition. Thus, the Enel Group is surely fully allowed, even expected, to implement practices that seek to improve its goods and services in order, inter alia, to remain competitive and retain its customer base. It therefore seems to me to be entirely in accordance with normal competition that a dominant undertaking, such as Enel, should wish to retain its customer base, even in the context of liberalisation.” (para. 66).

What a legacy monopolist should avoid, according to the Opinion, is to compartmentalize the market and it should “not adopt practices which, by exploiting the advantages stemming from the statutory monopoly, are capable of having exclusionary effects on new competitors considered to be as efficient as it is itself.”
Thus, former national monopolies are entitled to fight for their previous consumers and keep them, based on competition on the merits. They are not expected to play less hardball when competing in newly liberalised markets, even if they have a public service obligation or prior advantage.
In the words of AG Rantos, “Enel must not adopt practices which, by exploiting the advantages stemming from the statutory monopoly, are capable of having exclusionary effects on new competitors considered to be as efficient as it is itself” (para. 67). However, for AG Rantos, even if the incumbency advantages might make it impossible for competitors to replicate a similar strategy (because they did not have the monopoly), this does not preclude the possibility for competitors to use customer lists available in the market, similar to those lists put together by the Enel Group.

8. Parent liability – any food for thought for Vertically Integrated Utilities (VIU)?

The Opinion also sheds light on the concept of parent liability in competition law. Enel has traditionally been organised as a vertically integrated undertaking (VIU) with a monopoly in the generation and distribution markets (para. 11). The parent company, ENEL SpA, may be imputed with a subsidiary’s conduct if the latter is not able to determine its business activity independently.
In the energy sector a vertically integrated group or a vertically integrated undertaking (encompassing production, transmission, and distribution of energy) is usually the target of unbundling rules as required by EU energy law. Unbundling seeks to prevent possible competition distortions typically arising from simultaneous control within a VIU over the network (transmission and/or distribution) for generators or suppliers.
Although in this case the regulatory restrictions imposed through ownership or structural unbundling rules were not relevant, Servizio Elettrico Nazionale is a timely reminder of the risks accompanying vertical and horizontal integration, and that legal unbundling alone might not suffice to address competition concerns beyond the scope of regulatory law. The Opinion is of particular relevance for energy network sectors especially as companies seek to diversify into ancillary markets such as hydrogen networks or EV charging infrastructure where access to data will play a key role.

9. Some final thoughts

The Opinion of AG Rantos in Servizio Elettrico Nazionale clarifies various points concerning the concept of abuse in EU competition law in regulated sectors. AG Rantos raises the importance of ensuring that all entities operate on a liberalised market on an equal footing. He offers an important benchmark: in case a particular practice can be “replicated” or imitated by an as efficient competitor, no anti-competitive conduct may be ascertained on the part of an incumbent in a liberalised market. Indeed, although the case before the Italian court concerns a former incumbent monopoly, there is no reason that the AG’ reasoning can apply to data leveraging by any firm found to be dominant in one market to gain a position on an adjacent market. The burden of proof will be on the (former) incumbent to show that it competes on the merits in new and/or adjacent markets.

While the Opinion clarifies many points, it also uses broad, conceptual language. Would the ECJ render a more ‘practical’ answer to the referring court? It is still to be decided by the national courts (at Italian level) if the practices by the Enel Group effectively limited data available to competitors in the free market, foreclosing them as efficient competitors to the ultimate detriment of competition and the welfare of consumers.
It remains to be seen if there now are clear parameters for courts to determine what constitutes competition on the merits, or if it is to be left to economic analysis on a case-by-case basis. On what basis do we evaluate the transfer of customer lists and data from the regulated market to the free market? Maybe some of these difficult-to-answer issues will be addressed by ECJ in the months to come. We will stay tuned.

[1] See, for example, Pablo Ibañez-Colomo’s views in: AG Rantos’s Opinion in Case C-377/20, Servizio Elettrico Nazionale: a clean framework capturing the essence of the case law (I), and (II); Assimakis Komninos in: Competition Stories: November & December 2021; Carmen Puscas’: AG Rantos: What Is The Legal Framework For Analysing Data Leveraging Abuses Under Article 102 TFEU?

[2] AG Rantos remarks on the ordoliberal influence when it comes to the importance of a market structure. As a side note, ordoliberalism does not promote the protection of inefficient firms (some authors related to the Freiburg School might) – and here Rantos does not seem to argue in the contrary. One of us has written on Ordoliberalism and Competition Law, for those interested in the topic: Herrera Anchustegui, Ignacio, Competition Law through an Ordoliberal Lens. Oslo Law Review. Vol. 2, No.2. (2015). Available at: https://www.idunn.no/doi/10.5617/oslaw2568.

[3] This is the case in buyer power scenarios, for example.

[4] The discussion between the concomitant application of competition and regulatory law from a comparative perspective is an interesting issue. One of us has written about it regarding the electricity sector here: Herrera Anchustegui, Ignacio, Transmission Networks in Electricity Competition: Third-Party Access and Unbundling – a Transatlantic Perspective (Acceso a las Redes de Transmisión de Electricidad y Separación Efectiva: Una Perspectiva Transatlántica), in Ruiz Peris, Juan Ignacio, Cerdá Martínez-Pujalte, Carmen (ed) “Competencia en mercados con recursos esenciales compartidos: telecomunicaciones y energía” (Thomson-Aranzadi, 2019). 37 pages. Available in: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3159458. In the US see: Verizon Communications Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 U.S. 398 (2004) and Credit Suisse Securities (USA) LLC v. Billing, 551 U.S. 264 (2007).