Competition Stories: November & December 2021

Welcome to the Competition Stories – a bimonthly exploration of recent courts and competition law agencies’ decisions. Authored by Makis Komninos, a renowned expert in the field, this new column aims to go through the latest and most important developments in competition law of the last two months. We call them “stories” because Makis has promised to include some anecdotes from time to time, and not just stay at the black letter. Enjoy!


What an astonishing end of the year! November was the month of the Google Shopping judgment and December was dominated by AG Rantos’s Opinion in Servizio Elettrico Nazionale, which is likely to stay on as one of the most fundamental competition law Opinions ever. There were also a number of other interesting judgments, such as Stichting Cartel Compensation and Visma. Finally, we now know that the Intel judgment of the General Court will come out on January 26. It will be interesting to see whether the General Court will confirm or annul the Commission’s Decision. Irrespective of what happens at the General Court level, I think the pronouncements of the ruling of the Grand Chamber of the Court of Justice will remain relevant: they are the cornerstone of Article 102 TFEU and the Servizio Elettrico Nazionale Opinion is a perfect reminder of this.

I. General Court

Google Shopping

This was a judgment eagerly awaited. It is the first in a “trilogy” of Google cases (the Android and AdSense cases are currently pending) and raises important questions relating to the legal test and the theory of harm that the Commission had followed in this case. Although the words appear neither in the Decision nor in the judgment, this became known as the “self-preferencing” abuse. Full disclosure: I acted as counsel for the intervener supporting Google…

The facts of the case relate to the treatment that Google accorded to comparison shopping service providers vis-à-vis its own comparison shopping service, when consumers enter a query into the Google search engine. According to the Commission’s theory of harm, Google demoted rival comparison shopping services in its search results, while at the same time favouring its own comparison shopping service, which was not subject to Google’s generic search algorithms. As a result, Google’s comparison shopping service was more visible to consumers in Google’s search results, whilst rival comparison shopping services were less visible. This resulted in anti-competitive effects.

Google inter alia argued that the Commission should have satisfied the Bronner test (including the requirement of indispensability), whereas the Commission viewed Google’s conduct as merely an example of anti-competitive leveraging, which was against competition on the merits. As the readers will see, the General Court’s judgment amounts to a resounding victory for the Commission. There was only an annulment of a minor finding (see below) and the Court did not touch the EUR 2.42 billion fine and considered that “Google’s conduct constituted a particularly serious infringement” (para. 690). Indeed, the Court at times goes further than the Commission Decision and “rewrites” it. While the judgment deserves a closer reading, I will limit myself to the legal test, which, in my view, is the most important part of the judgment.

The Court starts in paras 128-135 with the applicable standard of judicial review. A friend of mine says that he is always suspicious when he sees judgments that begin by explaining what they are bound to do in terms of their review of the Commission’s decision. Indeed, these rather wordy and somewhat defensive paragraphs may be “explained” by the fact that – in my humble and biased view – the Court in reality may have gone further than what para. 135 states, i.e. that the General Court cannot “substitute their own reasoning concerning the assessment of the facts for that of the author of the contested act”. In any event, these paragraphs as such are correct and can be contrasted with the equivalent paragraphs in Microsoft, where the Court had fallen into an error, by devising a different standard of review for “technical appraisals” (para. 88) that supposedly warrant giving the Commission more discretion. This is something that the Court of Justice never confirmed and, indeed, did not include in its pronouncements in KME, Chalkor and GALP.

Then, the Court goes into the substance and the question of the legal test. I will focus on what I consider to be the seven most important elements of the judgment in this regard.

First, in paras 182-183, the Court explicitly espouses the concept of “super-dominance” (or “ultra-dominance” in para. 180) and links it with the concept of “gateway to the internet”. The Court also says that this can materially change the principle of “special responsibility” (para. 183 – “stronger obligation”), i.e. it’s not just a slogan but there are legal consequences. This is clearly a first and something that was not in the Decision. This theory is reminiscent of Article 106 TFEU, which applies to State undertakings or undertakings with special and exclusive rights (theory of “automatic abuse”). I would, however, resist this “ultra-special responsibility” notion in an Article 102 TFEU case, where the undertaking concerned has gained its strong position in the market as a result of competition on the merits and not because of any State measures.

Second, in upholding the findings of abuse, the Court discovers a “general principle of equal treatment, as a general principle of EU law” (para. 155), while relying on non-competition and irrelevant case law (Arcelor Atlantique et Lorraine). To that end, the Court also relies, in para. 180, on secondary EU law. However, the references to the net neutrality and the mobile roaming Regulations are problematic. Beside the fact that roaming is just too different from search, the fact that the legislature made a choice and introduced certain principles, does not mean that we can start introducing new concepts into Article 102 TFEU. Secondary legislation cannot change primary law, as interpreted in the Court of Justice’s case law. Article 102 TFEU catches primary or secondary line discrimination, only if the conditions of the case law are fulfilled (see MEO and AG Wahl’s Opinion in that case). So these paragraphs point in effect to a lex specialis for digital platforms, stricter than in other areas of 102 TFEU.

Third, in reality, the Court analyses the case at hand through the prism of exclusionary discrimination, something that Decision had shied away from. To be honest, I find the Court’s approach more intellectually honest than the Commission’s, on this point. The truth is that the Commission ran in reality a discrimination case. The issue, however, is that before Google Shopping we never had a precedent where the mere favourable treatment of a dominant company’s downstream activity was sufficient in itself to qualify as an abuse of Article 102 TFEU. Indeed, there has never been a precedent of the EU Courts finding a self-standing abuse of a dominant position on exclusionary discrimination grounds only (primary line discrimination). Deutsche Bahn and GT-Link were Article 106 TFEU cases and hence related to State measures, not pure conduct of a private undertaking. The same goes for Aéroports de Paris, to which para. 155 refers. As for Irish Sugar (again mentioned in para. 155), I am not sure it’s a relevant precedent, since this was a pricing case with elements of internal market restrictions. Absent the existence of some particular type of abuse (e.g. margin squeeze, anti-competitive discounts and rebates, introducing obstacles to the internal market, etc.), there is no Article 102 TFEU case suggesting that the sole allegation that a dominant company discriminates in favour of its own downstream activity is sufficient to make an abuse case. This is the novelty of Google Shopping.

Fourth, the Court agrees with the Commission that the conduct at stake amounts to anti-competitive leveraging, albeit after making some welcome clarifications on the concept of leveraging in paras 162-163:

The mere extension of an undertaking’s dominant position to a neighbouring market cannot in itself constitute proof of conduct that departs from normal competition, even if that extension leads to the disappearance or marginalisation of competitors. In addition, as is apparent in essence from the judgment of 25 October 2002, Tetra Laval v Commission, leveraging is a generic term in relation to the impact which a practice identified on one market may have on another market. The term may designate several different practices that are capable of being abusive, such as, in particular, tied sales as in the case giving rise to the judgment of 17 September 2007, Microsoft v Commission, margin squeeze practices as in the case giving rise to the judgment of 29 March 2012, Telefónica and Telefónica de España v Commission, or loyalty rebates”.

In other words, “leveraging” is a neutral and generic concept and can be both pro-competitive and anti-competitive. This is, indeed, a very positive clarification that comes at a critical point in the development of EU competition law enforcement in the digital sector.

Then, in paras 166-168, the Court says that certain leveraging practices can be anti-competitive and, in any event, the Commission was not relying only on leveraging as such but also on the “favouring” and “demotion” aspects of Google’s conduct (paras 166-168; see also para. 372). Then, the Court goes on to say (para. 168) that what made a difference here is that the conduct in question “was liable to lead to a weakening of competition on the market” because of “three specific circumstances – namely (i) the importance of traffic generated by Google’s general search engine for comparison shopping services […]; (ii) the behaviour of users when searching online […]; and (iii) the fact that diverted traffic from Google’s general results pages accounts for a large proportion of traffic to competing comparison shopping services and cannot be effectively replaced by other sources […]”. This is a mantra repeated many times and actually was also a central point mentioned in the Court’s press release.

Fifth – and this is where the “rewriting” of the Decision is evident, the Court introduces in para. 176 et seq. a rather intriguing and nebulous concept: “abnormality”. Look for example at para. 179: “the fact, assuming it to be established, that Google favours its own specialised results over third-party results, which seems to be the converse of the economic model underpinning the initial success of its search engine, cannot but involve a certain form of abnormality. It follows that, […] it is for the person responsible for that difference in treatment to justify it in the light of competition law”. This points to a sort of “no economic sense” standard but this is the first time we see this in case law, plus the last sentence amounts to a shift in the burden of proof – remember: we are at the first stage (establishing the abuse) and not at the objective justification stage.

In paras 181 and 184, the Court also seems to be relying on an approach that a situation of venire contra factum proprium is in itself suspect (see also para. 197 – “suspect elements”): “the deviation in relation to competition on the merits of the conduct at issue, assuming it is established, is all the more obvious as it follows a change of conduct on the part of the dominant operator. Google did indeed change its conduct on the market for general search services.” I find this part problematic. If there is a rule that a (dominant) company must always justify changes in its business strategy, we would be creating disincentives for dominant companies to adapt their business models and their overall conduct in order to better serve the consumer’s needs.

Sixth, the Court rejects Google’s Bronner pleas. In paras 219-222, the Court echoes the Slovak Telekom approach (without citing it though) and says that this is a case about access “conditions”, not about refusal to supply. Then, in paras 230-240, the Court suggests that the higher standard of Bronner does not apply to cases where there is no explicit request to be supplied and the dominant company simply engaged in unfair treatment. These findings place emphasis on form (i.e. whether there was an “explicit” request for supply, etc.) and therefore strike me as rather formalistic and inconsistent with economics. The overall approach of the Court also creates serious disincentives for both dominant companies and their competitors. The former may prefer not to supply rivals in the first place, for fear that their supply may later be “regulated”. The latter may prefer not to compete on the basis of innovation and rather enjoy a guarantee of support in perpetuity. Finally, the Court rejects in paras 244-247 the argument that the Commission should always rely on Bronner when the remedies at stake are typical “essential facilities” remedies, i.e. “transfer[ring] assets or enter[ing] into agreements in order to bring an infringement to an end” (para. 247).  Honestly, although I think the distinction between “passive” and “active” refusals to supply is nonsensical and formalistic, I do not generally disagree with that point. The substantive legal test cannot be dependent on the purely procedural choice that the Commission makes, when it decides for example whether to go for a behavioural measure of this or that nature.

Interestingly, however, the Court does go on to say that the Bronner conditions would be fulfilled here! Paras 223-229 clearly amount to a rewriting of the Commission’s Decision – effectively saying that the Bronner test would be passed in this case because of the very special circumstances. The Court says in para. 224 that “Google’s general results page has characteristics akin to those of an essential facility […], inasmuch as there is currently no actual or potential substitute available that would enable it to be replaced in an economically viable manner on the market”. At the oral hearing, the President of the Chamber had made this very point in a question posed to the Commission, i.e. why it chose not to rely on the “essential facilities” case law while taking the specific circumstances of the case into account.

Seventh, I should also mention that the Court seemed totally unimpressed by the argument that this is conduct that has brought technical improvements. The Court goes on to say that such conduct can be “an autonomous form of abuse where that improvement results in the dominant undertaking favouring its own product or service through recourse to methods different from those governing competition on the merits and that conduct is capable of having anticompetitive effects” (para. 188). Then, para. 314 goes even further: “If the funding model of an undertaking leads it, as in the present case, to take part in an abuse of a dominant position, there is nothing to preclude that funding model being caught by the prohibition under Article 102 TFEU. It is, moreover, a feature of many abuses of a dominant position that the aim is to improve an undertaking’s sources of funding.” I think we will see these passages being repeated quite a lot in the future… One thing is for sure: it would be unfortunate if they were to be read as an “efficiency offence”.

You may think that my commentary is one-sidedly critical. But the judgment has some positive elements, too, and it would be unwise for readers to consider it as a carte blanche for antitrust enforcers. Without wishing to enter into the merits and whether the Court got everything right in its assessment of the Commission’s effects findings, it remains the case that both the Decision and the judgment are very much focussed on an effects-based analysis. Indeed, the judgment has two parts: the first part is about the legal test (see above), but there is a second part (para. 356 et seq.), where the Court performs a rather detailed appraisal of the effects findings of the Commission. Indeed, the Court even annuls the part of the Decision that found (potential) anti-competitive effects on the market for general search services (paras 457-458). So this is certainly not a “by object abuse” case. Enforcers should see the whole picture and not concentrate on the first part of the judgment.

Another positive element is the Court’s constant focus on the causal link between conduct and effects (e.g. paras 382, 392 et seq., 412), while addressing Google’s arguments. The fact that the Court addresses these points in their substance is very welcome, because it implies that the Court does accept the importance of causality. This was not to be taken for granted, since the case law is rather mute on that point – there is only an obscure reference in para. 47 of Post Danmark II. If the Court believed that there is no need to prove any causal link as a matter of EU law, it would not have done all this assessment.

That being said, not everything is positive (from a systemic point of view) in that part. For example, I find the judgment’s part on the counterfactual analysis highly problematic. First, while paras 369-376 give the impression that the Court accepts the necessity of a counterfactual analysis, para. 377 et seq. contradict that impression and also seem to reverse the burden of proof (para. 378) in favour of the Commission. The Court seems to imply that a counterfactual analysis is less important when the Commission builds its theory of harm on potential, and not actual, effects. If that is the meaning of paras 377-379, then that must be wrong. An effects-based analysis is an effects-based analysis and there is no difference of substance or degree based on whether one relies on potential or actual effects (see also below AG Rantos’s Opinion in Servizio Elettrico Nazionale). Then, another point of contention was whether a realistic counterfactual is the absence of both elements of the abuse (“favouring” and “demotion”) or simply the absence of the “favouring” part of the conduct at stake. The Court opts for the former (paras 374, 376). But it was undisputed that each of the two elements separately would not necessarily have amounted to an abuse. If this is so, shouldn’t a credible counterfactual be the absence of just one element?

On a more positive note, I like the fact that the judgment refers to the Commission’s Guidance Paper at least twice (paras 443 and 615). If I am not mistaken, this is the first time a ruling of the EU Courts refers to the Guidance Paper in the course of the reasoning of its findings, i.e. it sees the Guidance Paper as a source of inspiration. This is certainly very welcome, especially at a time that the Guidance Paper is under attack by the formalists.

Finally, I would like to focus on paras 539-540, where the Court rejected Google’s arguments about the “as efficient competitor test” (AEC test). In reality, what happened here is that the Court confused the arguments made. Google was not arguing that the Commission should have employed the methodological-numerical tool known by the Guidance Paper as AEC test. Indeed, that particular price-cost test can only be applied in pricing abuses. Google was relying on the broader AEC principle, which the Intel case law considers fundamental in Article 102 TFEU.

Still, the Court had some rather positive things to say about the AEC test obiter and I think these are very useful pronouncements for pricing abuses. In particular, I read rather positively para. 538 which goes as follows:

The use of the as-efficient-competitor test is warranted in the case of pricing practices (predatory pricing or a margin squeeze, for example), in order, in essence, to assess whether a competitor that is as efficient as the dominant undertaking allegedly responsible for those pricing practices, and which, in order not to be driven immediately from the market, would charge its customers the same prices as those charged by that undertaking, would have to do so at a loss and accentuating that loss, causing it to leave the market in the longer term”.

In my view, the affirmative expression “is warranted” (“se justifie” in French) is an incremental improvement vis-à-vis Post Danmark II. I also welcome the explanation in para. 539 about the “as efficient competitor” being “a hypothetical competitor”, although the Court has overdone it and made a mistake by referring to the hypothetical competitor charging “its customers the same prices as the dominant undertaking”, whereas it should have only referred to costs. Be it as it is, still this are very positive obiter dicta.

So this was certainly a major development and we will all be holding our breath, while Google decides whether or not to appeal to the Court of Justice!

II. Court of Justice

1. AG Rantos’s Opinion in Servizio Elettrico Nazionale

On 9 December 2021, AG Rantos delivered his Opinion in Servizio Elettrico Nazionale, which is bound to remain as one of the foundational texts in Article 102 TFEU. It is also the strongest ever pronouncement from Luxembourg in favour of the effects-based approach and the Commission’s Article 102 TFEU Guidance Paper, to which the Advocate General refers on numerous occasions.

The Opinion reads like a scholarly article. This is because of the highly abstract and somewhat theoretical nature of the reference questions that were drafted by the referring court, the Consiglio di Stato (Italian Supreme Administrative Court). In fact, when I first read the questions, I thought that there was a risk that the Court of Justice might consider them inadmissible because of their overly abstract nature.

I will limit myself to the most fundamental points. Since the Opinion is not yet available in English and I know that many readers are anxious to read it, I will break with the tradition and I will quote extensively from the French text, producing my own unofficial translation, so that non-French speakers can appreciate the strength of the original text. It is regrettable that as of 1 January 2022, no English version has appeared on the Court’s webpage, because – I am told – English interpreters are swamped.

AG Rantos is fully aware of the importance of his Opinion, bearing in mind the chance offered by the nature of the reference questions. Paras 6-7 of the Opinion read:

The present case will permit the Court to crystallise its recent jurisprudence consisting of cases such as TeliaSonera, Post Danmark I and II, Intel, Generics (UK) and Deutsche Telekom II, in which it showed openness towards a less formalistic approach, while assessing the abusive nature of certain conduct, based on an appraisal of effects and taking into account the legal as well as the economic context of the conduct concerned. More specifically, this case offers the Court the opportunity to confirm whether certain principles emanating from that recent jurisprudence on pricing exclusionary abuses, in particular the “as efficient competitor” test, can be applied to non-pricing exclusionary practices, like the one of the present proceedings […] Finally, the present case is particularly interesting in that it relates to conduct that relies on the abusive use of data, which is now considered a very important indicator of power in certain markets, even outside the confines of the digital sector. The Court’s answers can therefore prove useful in the future, when it comes to analysing practices that relate to the use of data under Article 102 TFEU.

In other words, the Opinion should be of utmost relevance to the application of Article 102 TFEU to the digital sector.

The very abstract nature of the reference questions means that it is not critical to have a good understanding of the facts of the case, which are usefully set out in some detail in the working document that accompanies the preliminary reference. Suffice it to say that the case at issue relates to conduct by the former electricity State monopoly in Italy. After the liberalisation of the electricity market, generation and sale were opened to competition, whereas transmission and distribution are still operated under a monopoly system. The former State monopoly – Enel – was structurally 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) condemned the Enel Group for illegally using commercially sensitive information by operators to transfer customers from SEN to EE, in view of the announced market change. In particular, SEN obtained the consent of users to receive commercial proposals in a discriminatory manner, involving a request for initial authorisation for the processing of personal data by companies within the Enel Group and a second request in favour of third party operators. The customers usually gave the initial consent, believing that this was necessary for managing the existing relationship with their supplier, while they usually refused the second consent, intended for other operators. 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.

After dealing with some obvious questions, such as whether the legality under different laws of certain conduct affects the legality of the same conduct under Article 102 TFEU (the answer is obviously not), the Advocate General delves into the really interesting issues. In para 41 et seq., he proceeds to a powerful endorsement of the effects-based approach and the broader “as efficient competitor” principle in Article 102 TFEU. Relying on TeliaSonera and Generics (UK), he reminds that a conduct is abusive only if it is likely to restrict competition, in particular to lead to the anti-competitive exclusion of competitors that are as efficient as the dominant company, and this only after an “holistic appraisal” (see also para. 38 in fine) of all circumstances. A concrete and actual anti-competitive effect is not required, so a potential effect would suffice, but it cannot be “totally hypothetical” (para. 41). So far so good – the Advocate General simply puts together the Court of Justice’s pronouncements in a systematic way. But the Advocate General goes further, certainly in terms of emphasis:

42.  As a result, the crucial element in order to characterise a behaviour as abusive is the behaviour’s ability to produce a (potentially) restrictive result in the relevant market, such as an anti-competitive exclusionary effect.

43. However, […] solely the fact that several types of behaviour are capable of excluding a competitor does not render the market less competitive and, even more so, such behaviour abusive under Article 102 TFEU. Hence, a distinction should be made between a simple foreclosure risk and an anti-competitive foreclosure risk, which is the only one prohibited by Article 102 TFEU.

Here the Advocate General cites para. 19 of the Guidance Paper and goes on to refer primarily to Post Danmark I and Intel, which stress that it is not the objective of Article 102 TFEU to protect inefficient competitors (para. 44).

The Advocate General then stresses that anti-competitive foreclosure can only be based on an analysis of effects. According to para. 49, “the characterisation of a behaviour as ‘anti-competitive’ cannot be distinguished from the assessment of such a behaviour’s effects”. Indeed, the Advocate General stresses that “the Court has excluded the notion of the ‘per se abuse’, i.e. the existence of an inherently abusive practice, regardless of its anti-competitive effects” and that the “modern view in EU competition law [is] to favour an effects-based approach instead of a form-based approach” (para. 55). I imagine there may be other Advocates General who may not be entirely at ease with that “modern view” 😀…

Then, AG Rantos goes into what I consider the most interesting part of his Opinion. He attempts to give an answer to the question about the meaning of “competition on the merits” and the difference between “normal” and “distorted” competition. He states boldly that proving that a dominant company has engaged in conduct that is not consistent with “competition on the merits” is not different from proving anti-competitive effects (para. 48). He further demystifies the concept of “competition on the merits” by stressing the following in para. 49:

The analysis of the company’s behaviour plays a determinative role in the qualification of the effects as anti-competitive. Thus, these two elements belong to one and the same analysis. More specifically, in order to analyse whether the exclusionary effect is (actually or potentially) anti-competitive, it is equally necessary to analyse whether the elements of the behaviour pertain to competition on the merits. At the same time, in order to analyse whether the behaviour accords with competition on the merits, it is necessary to know whether the practices present a capacity to foreclose, in other words to cause (actual or potential) exclusionary effects on the market. In effect, the qualification of certain behaviour as ‘anti-competitive’ could not be part of an analysis that is distinct to the analysis of the effects of that behaviour. The two are thus intrinsically linked and must be appraised in light of the totality of the pertinent factual circumstances of that behaviour.

In para. 53 et seq. the Advocate General continues along the same lines and proposes a number of parameters that are crucial in the overall analysis of the problematic conduct. I find noteworthy the following two paragraphs:

62.  [T]he practices that do not accord with the concept of ‘competition on the merits’ are generally characterised by the fact that they do not rely on evident economic reasons or objectives. Therefore, ‘competition on the merits’ can for example cover practices that reduce the costs of the dominant undertaking, by increasing efficiency through another way, that have as effect the increasing of consumer choice, by bringing in the market new products or by increasing the quantity or quality of those already offered. On the contrary, if the conduct is characterised by no other justification than inflicting harm on competitors, it cannot be covered by competition on the merits.

This is an interesting paragraph because we are not in the stage of objective justification but still in the prior stage of establishing a behaviour that is restrictive of competition. I think this idea is also consistent with the Court of Justice’s most recent case law in Article 101 TFEU. We shall see if the Court picks this up in its ruling.

63.  [C]ompetition on the merits, in the context of applying Article 102 TFEU to exclusionary practices, refers generally to a competition situation, which benefits consumers through lower prices, better quality and a wider range of new or advanced products or services”.

In other words, Article 102 TFEU does not punish efficiency. On the contrary, Article 102 TFEU rewards it. This must be read in conjunction with what the Advocate General says about the general principle of the “as efficient competitor” and how this can practically be put in effect in the analysis of exclusionary behaviour. These are clearly important parts of the Opinion. In para. 93, the Advocate General notes the following (with my own emphasis):

[I]t derives from the now settled case law that Article 102 TFEU does not aim, or to be more accurate, no longer aims, in an absolute way, at protecting competitors against a dominant undertaking. Indeed, following the 2005 Commission’s revision of competition policy with regard to abuse of dominance, the Court has now recognised that Article 102 TFEU does not prohibit an undertaking from acquiring, by its own means, a dominant position in a market and, consequently, does not aim at securing that competitors, which are less efficient than the dominant undertaking, will remain in the market. This approach has been adopted by the Intel judgment, by means of which the Court clarified that ‘competition on the merits may, by definition, lead to the departure from the market or the marginalisation of competitors that are less efficient’. The perception that Article 102 TFEU does not aim, as a matter of priority, at protecting the (not as efficient) competitors, which is besides shared also by the Commission, seems to comprise part of the acquis of the Union’s competition law.

I think the Advocate General’s reference to the “as efficient competitor” principle now being part of the EU competition acquis, will characterise this Opinion. The Advocate General refers here to the AEC principle, as opposed to the AEC test. We should not confuse the two (as the General Court did in Google Shopping – see above). The AEC test is just the quantitative methodological tool used in assessing pricing practices. The AEC general principle, which forms part of the EU competition acquis, excludes from the protective scope of Article 102 TFEU the protection of inefficient competitors offering higher prices, worse products and less innovation.

However, the Advocate General does not limit himself to that. He also proposes concrete practical tests to evaluate the conduct of dominant companies, which are inspired from the AEC general principle. In so doing, the Opinion makes a distinction between pricing and non-pricing practices. As regards, first, pricing practices, the Advocate General gives his wholehearted support to the application of the AEC test (para. 70). But the Advocate General scratches below the surface and interestingly speaks of a more general idea behind the AEC test, which is of course linked to the general AEC principle. In his words, “the Court’s case law confirms […] that the exclusionary practices that are applied by a dominant undertaking and can be replicated by as efficient competitors, do not comprise, in principle, practices capable of resulting in anti-competitive foreclosure and, hence, are part of competition on the merits” (para. 69). This “replicability” criterion has relevance also for non-pricing practices, according to the Advocate General. Even for non-pricing behaviour, such as the behaviour at stake, in case a particular practice can be “replicated” or imitated by an as efficient competitor, no anti-competitive conduct may be ascertained. The readers are referred to paras 71-80, which contain a number of useful factors to assess replicability in the specific case.

As for the “dogmatic” question of the objective pursued by EU competition law and Article 102 TFEU, the Advocate General again takes a bold and principled position: protecting the structure of competition cannot be an objective alternative to “consumer welfare” (paras 94-95). The second objective is more central in his view, although it’s unwise to view the two objectives as unconnected to each other. Indeed, “Article 102 TFEU aims at the end of the day at maximising consumer welfare, in particular through the protection of the competitive structure of the market. Therefore, that protection may well constitute an objective pursued by Article 102 TFEU, not autonomously but only if in a specific case is integrated in the ultimate rationale of the (direct or indirect) protection of the consumers. Thus, the two objectives that the referring court seems to consider contradictory, in reality they are not so” (para. 100). And the Advocate General adds that effects on the structure of competition on a specific market can be considered a good “proxy” of the effects that a certain exclusionary practice may have on consumer welfare (para. 106). But an effect on the structure of competition does not necessarily mean a reduction of consumer welfare.

The Opinion continues and deals with other interesting questions, such as the role of intention and other factors, but I will mention only one last point, which I consider very interesting: the role of actual exclusionary effects. The Advocate General accepts that it is not necessary for actual effects to have taken place. However, AG Rantos notes that it is absolutely legitimate for a dominant undertaking to produce evidence showing the absence of actual exclusionary effects (paras 116-117):

[F]rom a procedural point of view, evidence of absence of anti-competitive effects that is produced ex post should be taken into account by a competition authority, in particular when that evidence relates to conduct that has ceased well before the adoption of the decision finding an abuse. Indeed, in the Intel judgment, the Court has explained that in case a dominant company produces, during the administrative procedure, evidence showing that its behaviour did not have the capacity to distort competition, the Commission is required to examine a number of elements in order to analyse the capacity of that behaviour to exclude from the market competitors that are at least as efficient […] In that context, economic evidence proving, after the contested conduct has ceased, the absence of foreclosure effects must be allowed when the dominant undertaking seeks to demonstrate that the conduct was not capable of restricting competition. Moreover, the anti-competitive effect of a given practice is not sufficient to be purely hypothetical, and the evidence disproving the capacity to restrict competition may corroborate such a hypothetical nature. Therefore, from a procedural standpoint, to the extent that the competition authorities bear the burden of proving anti-competitive foreclosure effects, I consider that those authorities are required to take carefully into account the evidence adduced by the dominant undertaking, even ex post, in the context of their analysis on the existence of an abuse.


The competition principle, depending each time on the relevant context, should take into account this type of evidence as an element potentially corroborating the lack of a possibility to restrict competition.”

As regards the probative value of such elements, AG Rantos highlights that “the lack of actual effects may, in my opinion, indicate that the practice was not capable, not even theoretically, of harming competitors, having as a result, the possibility of damage to be proven entirely hypothetical. This is the case in particular when the lack of such actual effects cannot be affected by subsequent factors, which are not related to the behaviour under dispute” (para. 119). So, again, a voice of reasonableness about a particular point that only makes sense and on which – unfortunately -competition authorities uncritically and invariably have shown complete inflexibility.

So will the Court of Justice follow AG Rantos’s Opinion? Can I risk a prediction? In my view, the Court will not contradict the Opinion and will follow closely the substance of the Advocate General’s proposed responses to the reference questions but the Court’s ruling will be of a different nature. It will not be as “scholarly”. The very abstract and theoretical nature of the reference questions may have given us a scholarly Opinion but I expect the Court to resonate in a different way and be more “practical”. So those that anxiously await to see whether the Court will tick each of the many boxes of the Opinion will be disappointed. But if that happens, I would be extremely happy, because we will have a very rich Opinion that has not been undermined in the least.

2. Stichting Cartel Compensation

For the (rather complicated) facts of this case, the readers are referred to my May-June Stories, where I presented AG Bobek’s Opinion. As I had predicted, the Court followed the AG, notwithstanding the Commission’s urging to the contrary. So the Court concludes that the so-called “transitional regime” for air transport services, during which Articles 104 and 105 TFEU applied, does not affect the direct effect of Article 101 TFEU and the ensuing competence for the national courts to deal with private enforcement cases.

In reaching that conclusion, the Court reminds us of the fundamental and institutional role played by the private enforcement of the EU competition rules. Paras 47 to 52 include references to the foundational case law of the Court of Justice, starting from Courage v Crehan. In short, private antitrust enforcement is a matter of both effective judicial protection and effectiveness of the competition rules. Then, the Court explains that Articles 104 and 105 TFEU only apply to the administrative enforcement of Article 101 TFEU and do not affect private enforcement, as AG Bobek had opined (para. 53). Very importantly, the Court also reminds us that Article 6 of Regulation 1/2003, which refers to the competence of national courts to apply the EU competition rules, does not change the legal reality, i.e. it does not affect the direct effect of Articles 101 and 102 TFEU (para. 54). Article 6 simply mentioned this “pour rappel”. In my monograph of 12 years ago, in p. 87, I had made exactly the same point: “The direct effect of Articles 81 and 82 EC is a direct consequence of the Treaty itself; Article 6 of Regulation 1/2003 therefore does not change the legal reality, but merely clarifies it.”

And, finally, the Court in effect reverses the old Asjes and Ahmed Saeed case law, where it had established that the national courts were denied competence to apply Article 101 TFEU to the air transport sector during the “transitional regime”. The Court, in paras 53-67, “distinguishes” that case law, because it was the product of a different era, when a different regime applied. In conclusion, the fact that the Commission could not apply Article 101 TFEU to the specific conduct in question, does not mean that the national courts could not.

3. Visma

This was a preliminary reference from Latvia. This is the fourth time that the Latvian courts have sent cases to Luxembourg, while reviewing the Latvian competition authority’s decisions (Maxima Latvija, AKKA/LAA, Remonts and Visma). This is a very nice record. I only hope other national courts were as open. On the substance, the Court again (as in Maxima Latvija) seems to disagree with the overly formalistic approach of the Latvian authority, which apparently finds too easily “object” infringements of Article 101 TFEU or of the equivalent national provision. The agreement in question was between a producer and a number of distributors and provided that the distributor who was first to register a potential transaction with the producer enjoyed priority in progressing the sale process with the end user for 6 months from that registration, unless the user objected. Indeed, my first impression when reading this was that I could not see how this could amount to an infringement of Article 101(1) TFEU “by object”.

The Court effectively instructs the national court to reject the Latvian authority’s “object” characterisation, after engaging in a pedagogical reminder of the basics behind the distinction between “object” and “effect” infringements of Article 101 TFEU and the respective analysis required, referring to its latest case law, in particular, Budapest Bank and Generics (UK).

The welcome novelty of this case is the Court’s confirmation that restrictions of intra-brand competition tend to be less serious than restrictions of inter-brand competition (para. 61). Para. 78 of the judgment is even more explicit and goes further: “Vertical agreements are, in principle, liable to be less harmful of competition than horizontal agreements. Thus, a restriction of intra-brand competition is, in principle, problematic only when effective inter-brand competition is weakened” (unofficial translation from the French original). If I am correct, this is the first time we have such a pronouncement from the Court of Justice and it is very welcome!


Citation: Makis Komninos, Competition Stories: November & December 2021, CONCURRENTIALISTE (January 6, 2021)

Read all the competition stories over here: link

Related Posts