Competition Stories: November & December 2020

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!


I. Court of Justice

  1. Canal+ (commitments)

2020 ends with many important developments. Let us first start with the new case law from Luxembourg. From a policy point of view, the most important judgment of the last two months is Canal+. It amounts to a dramatic annulment (the first ever) of a commitments decision of the European Commission. The ECJ decided to follow AG Pitruzzella’s Opinion (I was actually surprised – I thought the Court would not follow the AG) and set aside the General Court’s judgment and annulled the 2016 Paramount commitments decision on cross-border pay-TV services. As a reminder, this was a decision accepting commitments by Paramount that aimed at addressing the Commission’s competition concerns raised by certain clauses in film licensing contracts for pay-TV between Paramount and Sky UK that prevented cross-border services.

Canal+ challenged the Commission’s decision arguing that the commitments may have resolved the Commission’s concerns but encroached on its own contractual rights based on its agreement with Paramount. Initially, the General Court rejected the challenge, following a rather cryptic reasoning. Essentially, the way I interpret that ruling is the following: The rights that Canal+ bases on its existing contracts with Paramount can still be pursued before the national civil courts. The latter cannot order specific performance (since this would go against the Commission’s decision) but could award (for example) damages for non-performance. Paramount bears the risk of such possible damages exposure since it was its own decision to offer these commitments to the Commission. The General Court’s ruling was a rather ingenious solution to a typical problem of conflict between overlapping legal orders (supranational administrative competition enforcement v. national civil litigation).

The Court of Justice, however, probably thought of this as too fictional and preferred a rather “monist” approach. The Court found that the third party’s (Canal+) pre-existing rights were affected because the national courts could not go against a Commission decision (reading Article 16 Reg. 1/2003 in an expansive manner) and therefore the decision violated the proportionality principle as far as that third party’s rights were concerned.

The reasoning is quite interesting. First, the Court essentially says that a commitments decision, for the period of the commitments’ validity, allows national courts to adopt a stricter approach and find an infringement but does not allow a finding of non-infringement. However, this means additional duties for the Commission in terms of safeguarding third parties’ rights. We see here a certain “qualification” of the (bad) Alrosa judgment (remember Frederic Jenny’s 2014 Fordham article “Worst Decision of the EU Court of Justice: The Alrosa Judgment in Context and the Future of Commitment Decisions”?). The risk, in my view, is that this may reduce the Commission’s appetite to accept commitments whenever there are pre-existing contractual relations with third parties. So I fear that we may have gone from one extreme, the Alrosa carte blanche to adopt commitments decisions without any regard to proportionality, to another extreme, the Canal+ preference for infringement as opposed to commitments decisions. Let’s see how the Commission reacts in future cases….

Second, in the case at hand, the Court seems to imply that the specific clauses in the contracts are incompatible with Art 101. So the national courts would probably consider them as null and void under Art 101(2). At the same time, the US studios must refuse to respect such clauses, otherwise, they might be liable to fines by the Commission. And, after the annulment of the commitments decision, the Commission will be tempted to proceed to the adoption of an infringement decision….. So I am not sure this is really a “victory” for Canal+.

Third, I think the judgment may have an impact also on infringement decisions. What happens if the Commission orders a remedy (through an Art. 7 decision) that has an impact on existing contracts with suppliers or customers (third parties) and these contracts are not themselves void under Art 101(2) – e.g. follow-on contracts (Folgeverträge in German)? These third parties could seek the annulment of the Commission decision. Imagine for example an imposed remedy that changes a company’s business model and thousands of existing contracts are affected. The Commission will need to submit its remedy to an additional proportionality check. Or imagine an order to supply a customer (in a refusal to supply case), where the quantities in question are already committed to third parties.

By the way, Canal+ also challenged the commitments proposed by NBCUniversal, Sony Pictures, Warner Bros, and Sky Hollywood Studios in the same investigation. I don’t see how the Commission can avoid the annulment of that decision, too.

  1. Wikingerhof (private enforcement & bases of jurisdiction)

Another important ECJ judgment is Wikingerhof. The case concerns private enforcement. The issue at stake was whether an abuse of dominance civil claim made by the customer (hotel) of a digital platform ( is subject to the special basis of jurisdiction for contracts (Art. 7(1) Reg. 1215/2012) or to the special basis of jurisdiction for torts (Art. 7(2) Reg. Reg. 1215/2012). The facts are a bit complicated. The hotel sued in Germany against, although there was a Dutch forum clause. The first instance court said that the choice of forum prevails but the Court of Appeal complicated things by saying that the choice of forum clause was not validly concluded but, at the same time, it had no jurisdiction based on any of the special bases of jurisdiction. Then, on further appeal, the German Supreme Court sent the question of which special basis of jurisdiction applies to the ECJ. So the ECJ was not deciding on the choice of forum but only on which special basis of jurisdiction the claim should be subject to – i.e. whether this was a “contractual” or “tortious” claim.

After referring to previous case law on jurisdictional bases, the Grand Chamber of the ECJ, following the Opinion of AG Saugmandsgaard Øe, distinguished between “contractual” and “tortious” claims by using an “indispensability” criterion, i.e. whether the substance of the dispute makes it “indispensable” to analyse the contract in question or not. In particular, two requirements were set for a claim to be deemed of a “tortious” nature: (a) the claimant relies on the special jurisdiction basis for tort and (b) the assessment of the disputed conduct does not make it indispensable to interpret the contract that may be in existence between the parties. Given that in the case at hand the claimant alleged that committed an abuse of dominance within the meaning of German competition law, its claim was not “contractual” but rather of a “tortious” nature.

The commentary of Wikingerhof so far has emphasized that the ECJ took a very claimant-friendly approach and that from now on it’s clear that a claim based on an abuse of a dominant position is of a “tortious” nature and can be brought where the harmful event took place. However, this does not mean that such a (tortious) claim may not fall within a choice of forum clause that has been validly concluded (under Art. 25 Reg. 1215/2012) and is broad enough to cover such a claim. In that sense, Wikingerhof does not reverse or otherwise affect Apple. However, now that we know that such claims are not “contractual” but “tortious” in nature, it is very important how to phrase choice of forum clauses. If the latter are broad enough to cover such tortious claims, defendants can continue relying on them. I think that is what we are going to see more and more after these judgments: contracts with wide choice of forum clauses that cover tortious claims, too.

  1. SABAM (excessive pricing)

Finally, let’s not forget the SABAM case, where the ECJ ruled that the Belgian collecting society’s methodology for setting fees charged for music played at festivals is not necessarily abusive. I do not think there is anything ground-breaking in this ruling, which adds to the rather rich case law on excessive pricing and collecting societies. The Court simply said that, in principle, a model under which fees are calculated based on the total value of ticket sales of a festival is reasonable as long as it corresponds to the economic value provided. It also offered some concrete advice to the national court on how to analyse ad hoc whether the system was reasonable or not. So quite an open-ended ruling – in fact, both parties claimed victory, which means that the national court will have to do the dirty job.

II. General Court

  1. Lithuanian Railways (abuse of dominance)

Now, it’s the General Court’s turn. Let’s start with Lithuanian Railways. This was the case of the intentional removal of 19km of rail track by Lithuanian Railways (following a dispute with a customer) so that its customer could not use it in order to switch to the Latvian railway operator. The case is considered as one of those extremely rare cases where the dominant company destroys an asset, merely in order to make it impossible for customers to use it while switching to a competitor. If there still remain any “by object” abuses post-Intel, this is certainly one such case (together with pricing below AVC, and abuses like AB InBev).

[Small excursus: By the way, I have always believed that Article 102 has no space for “by object” infringements, as we have in Article 101, simply because the Treaty has adopted different systems of enforcement for the two provisions: Article 101 is based on a system of prohibition where the law prohibits the existence itself of an agreement or concerted practice, while conduct, as such, is in principle immaterial, whereas Article 102 is based on a system of abuse, where not the existence of a dominant position but only its abuse is prohibited. Therefore, an Article 102 violation based on likely or actual effects is the only way forward. What we sometimes call “by object” abuses, in reality, are practices that are prima facie anti-competitive but I am not sure this means much more than simply a small facilitation in terms of the evidential burden that the competition authority bears. And, after Intel, I doubt that references in older case law (Michelin II) to “by object” abuses are good law any longer.]

The Court upheld the Commission’s decision (which, by the way, has no reference whatsoever to a “by object” abuse 🙂). The Court, like the Commission, follows an effects-based analysis (which reinforces the above point), and in a way its analysis reminds us of a “no economic sense” standard. In other words, the dismantling of the 19km track can be explained by no other rational reason than by harming competition.

Lithuanian Railways tried to rely on Bronner and on the essential facilities line of case law. It argued that a smaller part of the track needed repairs anyway and it just decided to dismantle the whole track, in order to rebuild it at some point in the future. So, its conduct should be examined only as a refusal to rebuild the track (hence the refusal to supply angle). The Court was unimpressed and stressed the first part of the conduct in question: the dominant company had no reason to remove the track in dispute in great haste and the Commission made no error in determining that the removal of the track before the renovation works had even begun constituted highly unusual conduct in the rail sector. Then, the Court rejected the Bronner-based arguments and held that it was not necessary to apply here the essential facilities legal test, which includes the indispensability condition. In the Court’s view, Lithuanian Railways, in its capacity as Lithuania’s railway infrastructure manager, is responsible under EU and national law for granting access to public railway infrastructure and for ensuring the good technical condition of that infrastructure, as well as for ensuring safe and uninterrupted rail traffic and, in the event of disturbance to rail traffic, for taking all necessary measures to restore the normal situation.

Furthermore, Lithuanian Railways holds a dominant position on the market for the management of railway infrastructure which derives from a former statutory monopoly, and has not invested in the railway network, which belongs to the Lithuanian State. According to the Court, “[w]here there is a legal duty to supply, the necessary balancing of the economic incentives, the protection of which justifies the application of the exceptional circumstances developed in […] Bronner […], has already been carried out by the legislature at the point when such a duty was imposed” (para. 92). I am not shocked by this reasoning, which is nothing but a confirmation of para. 82 of the Article 102 Guidance Paper, although commentators have raised some persuasive arguments against. In any event, the teleology behind this argument has nothing to do with cases where the asset is the product of a private undertaking’s own innovation and investment.

Interestingly, the Court decided to use its “unlimited jurisdiction” to reduce the fine considerably. That does not happen that often. The reasoning behind that reduction is entirely cryptic. For those attentive to detail, it is also interesting to note that the Court did not award costs to the Commission as would normally be expected. So there is clearly “something” that attracted the Court’s sympathy. Could it be the fact that this is a public undertaking? Or that Lithuanian Railways in the end rebuilt the rail track in question, which was completed a few months before the délibéré? Can it be that the fine reduction reflects the rebuilding of the track?

  1. International Skating Union (eligibility & arbitration rules)

Then, we have the recent International Skating Union judgment. The General Court confirmed parts of the Commission’s infringement decision against the eligibility rules of the International Skating Union (ISU). This was a case brought by two skaters alleging that these rules did not allow them to compete in international events organized by third parties. The Commission found that the rules foreclosed such third parties. In essence, an exclusionary theory of harm and an Article 101 violation.

The Court confirmed the “conflict of interests” reasoning of the Commission’s decision, according to which, where a sports association is active in the organisation and commercial exploitation of sports events, but at the same time, through its regulatory function, has the power to authorise such events organised and commercially exploited by other, independent service providers, this may lead to a conflict of interest. This is a reasoning that has been used in Article 106 cases but the Court confirmed that it can be relied upon also in “pure” Article 101 (or 102) cases. Then, the Court stressed that the ISU retained too broad a discretion to refuse to authorise competitions proposed by third parties and could also impose severe penalties on skaters that were clearly disproportionate. While the protection of the integrity of a specific sport constitutes a legitimate objective recognised in Article 165 of the Treaty and it was legitimate for the ISU to establish relevant rules, the restrictive rules in question went beyond what is necessary to achieve that objective. Bearing in mind the above, the Commission was therefore right to conclude that the eligibility rules of the ISU revealed a sufficient degree of harm, to be regarded as restricting competition “by object”.

However, in its decision, the Commission (totally gratuitously) had also said that the ISU’s arbitration rules, which conferred exclusive jurisdiction to hear eligibility disputes on the Court of Arbitration for Sport in Switzerland, “reinforced the restrictions of competition”. At the hearing, the Commission was particularly aggressive against arbitration, although there was no finding that the arbitration rules themselves amounted to a violation. The Decision’s injunction also included an order to change the arbitration rules, which (again) were not found to amount themselves to an infringement! In my view, this was clearly an ultra vires order. The Court partially annulled the decision on that point. Paras 154-164 have a lot of good things to say about arbitration and clearly distinguish Achmea. In short, a welcome development on that particular point and a sigh of relief for arbitration.

III. European Commission

  1. Teva/Cephalon (pay-for-delay)

On the Commission front, we must certainly mention the Teva/Cephalon decision. This is one of the longest investigations ever and, no-doubt, it will spoil the Commission’s average length of investigations record. The formal opening of proceedings goes back to April 2011 and the investigation must have started even earlier! It’s unclear to me why it took so long. True, there were a number of important ECJ and General Court rulings that set the ground for the application of Article 101 to “pay-for-delay” agreements, but still…. The finding is that Teva and Cephalon violated Article 101 by agreeing to delay for several years the market entry of a cheaper generic version of Cephalon’s drug for sleep disorders, modafinil, after Cephalon’s main patents had expired. The agreement was concluded well before Cephalon became a subsidiary of Teva. The fine amounts to €60.5 million. According to the Commission, the decision “completes the cycle of pay-for-delay investigations launched with the Commission’s 2009 sector inquiry into the pharmaceutical sector”.

  1. Google/Fitbit (merger)

Finally, there is the Google/Fitbit merger clearance! There is no published text of the decision yet, so there is not much one can say. The Commission has just published the non-confidential text of the commitments, which are quite detailed. To all those that are criticising the Commission for not prohibiting this acquisition, I can only say that the Commission has to act within the confines of the law and the “law” here is the Merger Regulation and the Court’s case-law. While a possible prohibition may have made headlines now, an almost certain annulment would have followed three years later….

I have dealt with almost everything of importance but have left out the “elephant in the room”, the publication on December 15 of the Commission’s DSA and DMA proposals. The DMA, in particular, is certainly a ground-breaking competition law development (even if the Commission stresses that it does not amount to “competition law” 🙂) that will change our enforcement paradigm. But for that analysis, I need to ask Thibault to offer me exceptionally the chance of a separate guest article….


Citation: Makis Komninos, Competition Stories: November & December 2020, CONCURRENTIALISTE (December 22, 2020)

Read all the competition stories over here: link

Related Posts