Competition Stories: October 2022 – June 2023

Welcome to the Competition Stories – an exploration of recent courts and competition law agencies’ decisions. Authored by Makis Komninos, a renowned expert in the field. This column aims to go through the latest and most important developments in competition law in recent 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!


Μy catching up with the backlog continues. This time my “stories” cover the period October 2022 to June 2023 – so nine months in total! The current survey has to be rather summary for some cases – I focus mostly on those cases that I consider of importance. So we had a number of ground-breaking Court of Justice rulings in the area of Article 102 TFEU (Unilever Italia, Lithuanian Railways, Towercast) and some other notable rulings in private enforcement (DB Station & Service, Paccar, RegioJet, Tráficos Manuel Ferrer and Repsol). There were also a few interesting AG Opinions, in particular, in CK Telecoms, Towercast, ISU, European Superleague and EDP. And of course, we also had the Commission’s Amending Communication that introduced changes in its 2008 Guidance Paper on exclusionary abuses and the new Horizontal Agreement Guidelines. 

I. Court of Justice Judgements

1.1. Lithuanian Railways

This relates to the appeal from the General Court’s judgment upholding the Commission’s Decision in the Baltic Rail case (for the rather extraordinary facts, see my stories of November-December 2020). In its appeal, Lithuanian Railways first argued that the General Court failed to consider its conduct as a refusal to supply case and apply the Bronner criteria. I am sure the judges have heard this argument many times – I have actually been involved in two other cases where the “B” word (Bronner) figured prominently (more on that when the time comes). As a reminder, according to Bronner, for a refusal to supply to be abusive, (i) an undertaking in a dominant position must be refusing to grant access to an indispensable service, (ii) the refusal is likely to eliminate all competition in the market, and (iii) there is no objective justification. So was the General Court right or wrong? 

In short, the Court followed entirely the approach proposed by AG Rantos in his Opinion (see my stories of July-September 2022). Perhaps the most important element is the Court’s espousal of a “no economic sense” test or approach that the Advocate General had proposed. And the Court’s distinguishing of the present case from Bronner offers a good idea as to how it sees refusal to supply and the Bronner criteria these days, when a number of pending cases relate to that question. 

Before going to the legal test applied here, we are quite lucky to have a very didactic restatement and confirmation of the Bronner conditions by the Court in para. 79: 

As regards practices consisting of a refusal to grant access to infrastructure developed by a dominant undertaking for the purposes of its own business and owned by it, it is apparent from the case-law of the Court that such a refusal may constitute an abuse of a dominant position provided not only that that refusal were likely to eliminate all competition in the market in question on the part of the entity applying for access and that such refusal were incapable of being objectively justified but also that the infrastructure, in itself, were indispensable to carrying on that undertaking’s business, inasmuch as there was no actual or potential substitute in existence for that infrastructure.” 

Note that the Court speaks of “elimination of all competition”, not just a restriction or deterioration of competition, and of indispensability to “carry on business”, not indispensability to compete effectively. I have had this battle so many times with competition authorities… They have always insisted that not “all” competition must be eliminated and that indispensability relates to safeguarding an effective competitive source. Well, the above passage is clear! 

Another important element is that, in distinguishing the present case from Bronner, the Court reminds us that “[t]he examination of the abusive nature of a dominant undertaking’s practice pursuant to Article 102 TFEU must be carried out by taking into consideration all the specific circumstances of the case” (para. 78). The reference to “all circumstances”, a test that has long existed in pricing abuses that are analysed under an effects-based approach, means that we should avoid speaking of Lithuanian Railways as an exceptional “by object” case (see here where I analyse this issue). So, although it is tempting to think in terms of “object”, the Court does not go there, and to be fair, not even the Commission in its Decision treated this as a “by object” case. 

Then, the Court explains why the Bronner legal test should not apply to the conduct at stake. The removal of track should not be analysed as a refusal of access but as “an independent form of abuse” (para. 91) for three main reasons. Here, the Court follows entirely AG Rantos’s Opinion: 

First, “the destruction, by a dominant undertaking, of infrastructure must be distinguished from a refusal of access” (para. 81). The Bronner case law concerns “a refusal of access to infrastructure, whereby, ultimately, the dominant undertaking reserves the infrastructure which it has developed for its own use” (para. 82). In other words, it is normal for the law to protect the right of ownership of a dominant company over “infrastructure, the use of which the dominant undertaking reserves for itself in pursuit of an immediate benefit”. By contrast, “destroying infrastructure entails sacrificing an asset […] As a result of the destruction, the infrastructure inevitably becomes unusable by competitors but also by the dominant undertaking itself” (para. 83). This is the “no economic sense” approach that the Advocate General had advocated for this case in his Opinion. The case thus did not involve a refusal to provide access, but a situation where the dominant company incurs a cost to dismantle an asset, apparently with a view to preventing market entry. 

Second, the Bronner criteria are “intended to strike a fair balance between, on the one hand, the requirements of undistorted competition and, on the other hand, the freedom of contract and the right to property of the dominant undertaking. In that sense, those criteria are intended to be applicable in the event of refusal of access to infrastructure which the dominant undertaking owns and which it has developed for the needs of its own business by means of its own investments” (para. 86, emphasis added). However, the Bronner criteria could not apply in this case “where the infrastructure in question was financed by means not of investments specific to the dominant undertaking, but by means of public funds and that undertaking is not the owner of that infrastructure”. By the way, this is very much in line with para. 82 of the Commission’s Article 102 Guidance Paper. 

Third, referring to its recent Deutsche Telekom / Slovak Telekom judgments, the Court noted that the Bronner criteria would not be applicable in any event, given that Lithuanian Railways was subject to a regulatory obligation to give access to its infrastructure. The text actually used by the Court is telling: 

“[I]t is important to recall that the Court has already held that a regulatory obligation can be relevant for the assessment of abusive conduct, for the purposes of Article 102 TFEU. While the existence of a regulatory obligation on the dominant undertaking to grant access to the infrastructure in question cannot relieve the Commission of the requirement to establish the existence of an ‘abuse’ within the meaning of Article 102 TFEU, the fact remains that the imposition of such an obligation has the consequence that the dominant undertaking cannot actually refuse to give access to that infrastructure, without prejudice, as appropriate, to its decision-making autonomy in relation to the conditions for such access.” (para. 88, emphasis added) 

In other words, the violation of a regulatory duty in itself cannot amount to an abuse of dominance. At the same time, the existence of regulation is relevant to the legal assessment. If there is a regulatory duty to deal, it means that the dominant company has no legal right to refuse access, therefore this is not a refusal to deal case. It may though be a case about the “conditions for such access”, which is a slightly different story. That is clearly reminiscent of paras 50-51 of Slovak Telekom. 

Finally, the Court rejected the appeal on the fine. One of the mysteries of this case was that the General Court had decided to use its “unlimited jurisdiction” to reduce the fine considerably, although it entirely upheld the legality of the Commission’s Decision. In addition, the General Court had not awarded costs to the Commission as would normally be expected. In my stories of November-December 2020, I had speculated that the judges maybe saw with some sympathy the fact that Lithuanian Railways had rebuilt the track prior to the délibéré. Another mystery is that the Commission did not cross-appeal the point on the fine reduction, although there was no reasoning. Indeed, AG Rantos stressed this (with a degree of surprise) in his Opinion (para. 161). The third mystery is that Lithuanian Railways failed to argue the lack of reasoning behind the fine adjustment, although it was challenging the proportionality of the fine, as determined by the General Court. Again, AG Rantos stressed this in his Opinion. In the end, the Court of Justice decided that the reasoning was sufficient and found that, contrary to what the appellant argued, the General Court had not set the new fine in consideration of its anticompetitive intent and that, in any event, the appellant “failed to adduce evidence capable of showing that the level of the fine, as reduced by the General Court, is not only inappropriate, but also excessive to the point of being disproportionate” (para. 159). And, this time, it awarded costs to the Commission (and the supportive intervener)… 

1.2. Unilever Italia

This is a fascinating case and although I was not personally involved, I did actually travel to Luxembourg to attend the oral hearing. The case stems from an infringement decision of the Italian Competition Authority (ICA) against Unilever on abuse of dominance grounds (exclusive dealing). In reality, the main questions at stake were two. One is about the concept of “single economic unit”, which the ICA used in a rather curious way, considering that Unilever and its distributors were a single economic unit, because Unilever exercised tight control over them. So effectively all contracts signed by them (with exclusivity clauses, among others) were attributed to Unilever. The second question was about the legal test in exclusive dealing cases.  

On the second and more important question, it was never doubted that the ICA had to look at “all circumstances” and at the parameters mentioned in para. 139 of Intel. This was undisputed and actually the ICA argues that it had run such an analysis. What was disputed is whether the ICA ought to have taken into account certain economic studies prepared by Unilever, which were inspired by the AEC test. In particular, Unilever had relied on an “efficient breach of contract” analysis done by their economists, which sought to attribute a value to each contract and hence check whether the competitors could match that value.  

Essentially, the question was: does Intel apply only to rebates or also to exclusive dealing? The Court unequivocally says that Intel is clear: it covers exclusive dealing, too! Read paras 50-51: 

50. It is true that, in providing that second clarification [means the Intel ‘clarification’ 😀], the Court referred only to rebate schemes. However, since both rebate practices and exclusivity clauses are capable of being objectively justified or of having the disadvantages which they generate counterbalanced, or even outweighed, by advantages in terms of efficiency which also benefit the consumer, such a clarification must be understood as applying to both of those practices.

51. Moreover, in addition to the fact that such an interpretation appears to be consistent with the first clarification provided by the Court in that judgment of 6 September 2017, Intel v Commission (C-413/14 P, EU:C:2017:632, paragraph 139), it must be held that, although, by reason of their nature, exclusivity clauses give rise to legitimate concerns of competition, their ability to exclude competitors is not automatic, as, moreover, is illustrated by the Communication from the Commission entitled ‘Guidance on the Commission’s enforcement priorities in applying Article [102 TFEU] to abusive exclusionary conduct by dominant undertakings (OJ 2009 C 45, p. 7, paragraph 36).

There you go. Another major endorsement of the Guidance Paper. Still, the Commission had other plans, as we will see in a moment…. 

Very importantly, the Court clarifies that competition authorities are under a duty to take into account economic analyses produced by dominant companies. This is based on the right to be heard, a general principle of EU law. This is clear from paras 54 and 55. In other words, we have a principle of both substantive and procedural basis. The Court’s emphasis on the procedural nature of this duty of competition authorities goes as far as obliging them not just to discredit the defendant’s economic evidence but also to put forward their own (economic?) evidence that should substitute the former: 

55. It follows that, where the undertaking in a dominant position has produced an economic study in order to demonstrate that the practice of which it is accused was not capable of excluding competitors, the competent competition authority cannot exclude the relevance of that study without setting out the reasons why it considers that the study does not contribute to demonstrating that the practices in question were incapable of undermining effective competition on the relevant market and, consequently, without giving that undertaking the opportunity to determine the evidence which could be substituted for that study” (emphasis added). 

And what about the AEC test? The Court first repeats that a competition authority is not obliged to use the AEC test (that we have known for some time) (para. 58). Para. 56 includes an exemplary definition and description of the AEC test (there have been errors when the EU Courts attempted to define it in the past – see e.g. Google Shopping para. 539). That being said, the Court does recognise the possibility that in exceptional cases, the AEC test may have to be run on the basis of the cost structure of a less efficient competitor (see para. 56 in fine: “[the ability of a practice to produce anti-competitive exclusionary effects] is generally determined in the light of the cost structure of the undertaking in a dominant position itself” (emphasis added). However, the word “generally” seems to imply that this must be an exceptional circumstance and the departure from the rule must be duly reasoned.   

Can the AEC test be used in exclusive dealing (which is – not to forget – a non-pricing practice)? Surprisingly, the Court goes quite far and says YES!!!  

59. Nevertheless, even in the case of non-pricing practices, the relevance of such a test cannot be ruled out. A test of that type may prove useful since the consequences of the practice in question can be quantified. In particular, in the case of exclusivity clauses, such a test may theoretically serve to determine whether a hypothetical competitor with a cost structure similar to that of the undertaking in a dominant position would be able to offer its products or services otherwise than at a loss or with an insufficient margin if it had to bear the compensation which the distributors would have to pay in order to switch supplier, or the losses which they would suffer after such a change following the withdrawal of previously agreed discounts. 

Not even AG Rantos in his Opinion had gone as far as that! 

Before I go to the first question on the attribution of liability to the dominant company, let me deal with one particular question that relates to the general part of the Court’s reasoning on the concept of abuse. In para. 39 (in my view by mistake), the Court says that in order to have an abuse you need to have anti-competitive effects on as efficient competitors or conduct against competition on the merits. The Court cites Servizio Elettrico Nazionale for that proposition, but that is clearly wrong. That judgment makes clear that the correct word should have been and, not or. The very recent Grand Chamber judgment in European Superleague (para. 129) corrects that mistake and makes clear that the conditions are not alternative, i.e. you need conduct against competition on the merits plus anti-competitive effects. More on that in the next stories. 

With regard to the first question that the Court had to deal with, the Court distanced itself from AG Rantos’s Opinion (see my stories of July-September 2022) and I think it was right to do so. The Court avoids approaching the question from the angle of the “single economic unit” and opts for a more direct approach: according to the Court, nothing precludes the possibility of finding a dominant company liable under Article 102 TFEU for the conduct of its independent distributors, if certain circumstances are present. In the present case, the dominant company was giving specific instructions as part of the implementation of a policy that was decided unilaterally without involvement of the distributors, and with which the distributors were required to comply. For the Court, 

In such a situation, the distributors and, consequently, the distribution network which they form with that undertaking, must be regarded as merely an instrument of territorial implementation of the commercial policy of that undertaking and, on that basis, as being the instrument by which, as the case may be, the exclusionary practice at issue was implemented. That is the case, in particular, where such conduct takes the form of standard contracts, drawn up entirely by a producer in a dominant position and containing exclusivity clauses for the benefit of its products which the distributors of that producer are required to have signed by the operators of sales outlets without being able to amend them, unless that producer expressly agrees. In such circumstances, that producer cannot reasonably be unaware that, in view of the legal and economic links which it has with those distributors, the latter will implement its instructions and, thereby, its commercial policy.” (paras 30-31) 

This is preferable to relying on the “single economic unit” concept, since otherwise we would have serious issues for the application of Article 101 TFEU…

1.3. Towercast

This is a preliminary reference from France. The French Autorité de la concurrence had rejected a complaint, holding that it cannot apply Article 102 TFEU to a merger because of Article 21 of the Merger Regulation. I had serious doubts as to that view. Article 102 TFEU cannot be affected by secondary EU law. The judgment was eagerly awaited, especially after AG Kokott’s Opinion. The Court generally followed the Advocate General and held that Article 102 TFEU can be applied by NCAs (and by implication by national courts) to mergers that fall below both EU and national thresholds.  

According to AG Kokott’s Opinion, national courts and NCAs can apply Article 102 TFEU to consummated mergers that were never notified, because they were below the notification thresholds. The Advocate General did not go into the extreme of proposing that notified and cleared mergers should be subject to Article 102 TFEU. Her view is that Article 102 TFEU can be applied only to unnotified mergers below national thresholds. 

First, AG Kokott concludes that provisions contained in the Merger Regulation (or equivalent national merger control legislation) cannot preclude or limit the applicability of Article 102 TFEU to transactions, because Article 102 TFEU, as a provision of primary and directly applicable law, is hierarchically superior to the secondary rules contained in the Merger Regulation. And national ex ante merger control regimes do not have the status of a lex specialis derogating from the application of general antitrust rules (para. 42). She further opines that the referral mechanism under Article 22 of the Merger Regulation, according to which Members States can ask the Commission to review non-reportable mergers that do not meet the EU or national thresholds, is irrelevant to the interpretation of the relationship between the Merger Regulation and Article 102 TFEU, and therefore cannot justify the exclusion of Article 102 TFEU (para. 47). Second, AG Kokott argues that an acquisition by a dominant company in itself may amount to abusive conduct and thus fall within Article 102 TFEU, bearing in mind that the latter provision in effect is a general clause and has a wide scope (para. 45). Third, AG Kokott finds that Article 102 TFEU fulfils a complementary role as it helps to address an enforcement gap in capturing “problematic” mergers that were not notified because they did not trigger thresholds at national or EU level but nonetheless merit review. AG Kokott refers in particular to the so-called “killer acquisitions” or acquisitions occurring in highly concentrated markets, where the aim is to eliminate competitive pressure from an emerging competitor (para. 48). Interestingly, AG Kokott also makes a statement (wish? proposal?) that if a below-threshold transaction is subsequently found to be in breach of Article 102 TFEU, it most likely would not be unwound but rather the dominant company would be imposed a fine (para. 63). In addition, relying on the principle of legal certainty, the Advocate General considers that completed transactions that were approved as part of a merger control regime cannot be subsequently reviewed under Article 102 TFEU (para. 60). All in all, a very sensible Opinion. 

The Court a few months later followed the Advocate General. For the Court, transactions that (i) fall below EU merger control thresholds or (ii) fall below national thresholds and at the same time were not referred to the Commission pursuant to Article 22 of the Merger Regulation, may be subject to Article 102 TFEU (paras 52-53). This reference to Article 22 of the Merger Regulation and the referral mechanism is intriguing. Does the Court imply anything about the cause célèbre of Illumina/Grail that is currently pending before it (Cases C-611/22 P and C-625/22 P)? Probably not. Equally intriguing in that sense is para. 34 which refers to a “situation […] where the concentration concerned, first, has not met the thresholds for control laid down by EU and national law and, second, has not been referred to the Commission under Article 22 of that regulation” (emphasis added). Not clear. We shall soon know, anyway. 

That being said, the Court does not give a “carte blanche” to NCAs. The Court emphasizes that strengthening a dominant position through an acquisition is not in itself sufficient for a finding of abuse. It specifies that “it must be established that the degree of dominance reached through the acquisition would substantially impede competition, that is to say, that only undertakings whose behaviour depends on the dominant undertaking would remain in the market” (para. 52). It is not entirely clear how this test is to be applied. It does seem, however, to indicate that the merger must have in reality resulted in near-total elimination of competition. If what remains is competitors/ customers/ suppliers that can only subsist by their conduct being entirely dependent on the dominant company, this is not real competition. Hence, we must have a total or near-total elimination of effective competition. 

Unlike AG Kokott,  the Court does not explicitly refer to mergers that were reviewed and approved as part of a merger control regime. Here, a distinction can be made between mergers that were reviewed and approved by the Commission and mergers that were nationally reviewed and cleared. For the former mergers, the Court does not even consider this as a hypothetical scenario. Instead, it conditions the possibility for a competition authority capturing a concentration under Article 102 TFEU on “certain conditions” being present (para. 40). And when it sets out its conclusions in para. 52, it only refers to the application of Article 102 TFEU to mergers that fell below the thresholds. This may be interpreted as meaning that for the Court of Justice the first scenario is impossible to contemplate. For the latter mergers, the Court’s position is not clear. Although the operative part mentions only mergers that fell below national notification thresholds and were never referred to the Commission, a degree of uncertainty remains because parts of the reasoning may suggest otherwise. So, it cannot be excluded that NCAs (and – not to forget – national courts) could conceivably apply Article 102 TFEU even to mergers that were nationally reviewed and cleared.  

Clearly, an important case both for Article 102 TFEU and merger control. However, the importance is more academic than practical, because I don’t see NCAs running to open such cases in the future. We will also see what Towercast means for Illumina/Grail. From a purely policy point of view (not necessarily legal), if anything, Towercast shows that there is no enforcement gap that can justify the Commission’s change of approach with the publication of its new Guidance on Article 22 of the Merger Regulation. True, there may be mergers below dominance, but sorry: this can’t be a serious problem/gap.

1.4. Rulings in the area of private enforcement

Let me now go to the private enforcement cases. For sure, the most interesting one was DB Station & Service. I think this is one of the most controversial judgments of the last years. Essentially, the question at stake was: Should the existence of EU secondary regulation impose a rule of precedence in favour of the sectoral regulator, thus resulting in limiting private enforcement of EU/national competition law? Surprisingly for many (but not for me), the Court chose not to follow AG Ćapeta’s Opinion (see my stories of March-April 2022) and said yes!  

Here is the context: there was a private action in Germany brought against a subsidiary of Deutsche Bahn (DB), a railway infrastructure undertaking which maintains 5,400 stations (traffic hubs), by a rail transport undertaking which uses the defendant’s traffic hubs for passenger railway services. The civil action alleged that the charges levied for that purpose were excessive and thus constituted an abuse of a dominant position. The complication here was, however, that the sector is regulated by EU secondary law (Directive 2001/14/EC, later repealed by Directive 2012/34/EU) and the Federal Network Agency (the Bundesnetzagentur), acting as the competent regulatory body, declared the DB price system to be invalid, albeit with effect from a later date than their adoption. The referring court entertained doubts as to whether national civil courts are entitled and obliged to review the charges levied based on Article 102 TFEU and national competition law, independently of the monitoring carried out by the regulatory authority. In other words, shouldn’t the Federal Network Agency take precedence? 

The German court’s doubts were fuelled by a previous ruling of the Court of Justice, not related to competition law, which had imposed limitations on national courts. In CTL Logistics, the Court held that the German courts’ reliance on the German Civil Code (BGB) to perform a review of the equity of the charges levied by railway infrastructure undertakings was incompatible with the provisions of Directive 2001/14/EC. In essence, the Court thought that this constituted “excessive protection” incompatible with the requirements and objectives of EU sectoral legislation. In the meantime, there was another ruling of the Court of Justice, in the Koleje Mazowieckie case, which followed the CTL Logistics logic. So the issue was how to reconcile these judgments, which allowed the national sectoral regulators a degree of precedence, with the direct effect of Article 102 TFEU, bearing in mind that, unlike those cases, the present case related to a rule of primary EU law!  

The Court’s reasoning is rather cryptic and sounds at times self-inconsistent. So, on the one hand, the Court repeats that Article 102 TFEU “produces direct effects in relations between individuals and creates rights for the persons concerned, which the national courts must safeguard” (para. 46). Therefore, the direct effect of Article 102 TFEU cannot be limited. But, on the other hand, the Court clearly does not wish to overrule or even distinguish CTL Logistics and Koleje Mazowieckie and places emphasis on the “technical” competence of national regulatory bodies (references to the word “technical” abound: paras 57, 61, 78, 83). So, the Court proceeds to the adoption of a somewhat concealed rule of precedence, which makes more practical than theoretical sense. The critical paragraph is 81: “Thus, where a railway undertaking intends, on the basis of Article 102 TFEU, to obtain reimbursement of an alleged overpayment of infrastructure charges, it must, prior to any action being brought before the national courts having jurisdiction, refer the question of their lawfulness to the national regulatory body.” My own translation: yes to follow-on but no to stand-alone private litigation. 

Clearly, a very controversial ruling, although it makes practical sense. It also shows how difficult it is for the Court of Justice to overrule its own rulings. It usually tries to find a way to reconcile conflicts in diverging strands of case law, sometimes not very successfully. In any event, I have to say, I feel somewhat vindicated. I was closely watching this case (see here). And what the Court says is exactly what I had advocated for the Digital Markets Act (DMA), i.e. that the legislator should allow only follow-on litigation and exclude stand-alone claims. See my article here. In the end, the EU legislator decided otherwise, so this is just a theoretical debate. My proposal had been heavily attacked: how is it possible for the legislator to restrict legal rights and to introduce a rule of precedence for the Commission, the argument went. Well, now the Court of Justice has gone even further. It has introduced a rule of precedence and has restricted legal rights, indeed even rights based on primary EU law! My DMA-related proposal was much less radical… The DB Station & Service ruling proves that it would have been perfectly possible…  

There were also a number of other rulings in the area of private enforcement, but I am not going to spend too much time on them. 

First, we had Paccar, a preliminary reference from – let me guess – Spain again! The question at stake related to the interpretation of Article 5 of the Damages Directive, which contains rules on evidence. Does Article 5 go as far as ordering a defendant to disclose documents that it must create ad hoc for the first time, by compiling or classifying information, knowledge or data? The case goes back to the 2016 Commission Decision in the trucks cartel case. The plaintiffs here asked the referring court to order the defendants to disclose documents to allow them to compare recommended prices for trucks before, during and after the cartel period, to calculate production costs and to distinguish between gross and net prices in order to prepare an expert report. The problem was that such documents had to be created ad hoc. However, that was not a problem, according to the Court of Justice. This is a plaintiff-friendly judgment but with some limitations. It follows AG Szpunar’s Opinion and holds that Article 5 of the Damages Directive can go as far as ordering a defendant to disclose documents that it must create ad hoc for the first time, by compiling or classifying information, knowledge or data. Still, according to the Court, disclosure must be “subject to strict review by national courts”. In particular, national courts are required “to restrict the disclosure of evidence to that which is relevant, proportionate and necessary, taking into account the legitimate interests and fundamental rights of [the defendant]”. One intriguing point appears in para. 66: “That reasoning applies, a fortiori, to proceedings in which no penalty has been previously imposed by the Commission or by a national competition authority for unlawful conduct.” My own translation: national courts have to be more restrictive in allowing an expansive disclosure in stand-alone as opposed to follow-on cases! 

The judgment has some very interesting dicta, in my view, on the aims of private antitrust enforcement and how it complements public enforcement. If you read paras 55 and 56, it is clear that the Court indicates that the primary objective of private enforcement is functional and has to do with the effectiveness of enforcement and increasing deterrence – the secondary objective is compensatory and has to do with effective judicial protection. The judgment includes a lot of economic parlance (NB: juge-rapporteur: Nils Wahl). As in Sumal, the Court clearly mentions again economic efficiency (again the FR is better than the EN text) and consumer benefits. And, like in Sumal, the Court makes a distinction between “direct damage” (in the case of a cartel, that would be the overcharge) and “indirect harm” which is not personalised – in reality the Court here refers to the famous “deadweight loss”. Of course, we all know that it’s almost impossible to concretize and compensate that harm, but that’s a different story. So another very economically minded judgment! 

Second, RegioJet was a case about disclosure of documents, in the aftermath of Paccar. But there are also some nice statements on the relationship of independence and at the same time complementarity between public and private enforcement (see e.g. paras 63-66). It’s a sort of an incremental development and was decided by the same Chamber with the same juge-rapporteur (Nils Wahl). At stake was Article 6 of the Damages Directive and, in particular, Article 6(5), which deals with documents prepared expressly for the purpose of public enforcement proceedings (so-called “grey list documents”). There were many questions to the Court, which to me were sort of obvious, but anyhow… The Court essentially held that the mere fact that there are simultaneous proceedings before the Commission does not mean that the national court cannot rely on Article 5(1) of the Damages Directive and order disclosure of documents (subject to the relevant conditions being fulfilled – including proportionality). For the rest, the Court interpreted Article 6(5) of the Damages Directive in a strict way: documents “prepared” specifically for the public enforcement proceedings cannot be equated with documents “submitted” to the public authority. Documents that were pre-existing to the investigation cannot fall within the “grey list” (para. 105). The fact that they have been submitted to the authority is irrelevant. The national court can defer the examination of what has been “prepared” specifically for the public enforcement proceeding, as long as there are safeguards in place that ensure such documents are not prematurely communicable. 

Third, we had Tráficos Manuel Ferrer. This was another preliminary reference from Spain in a follow-on damages case again further to the Commission’s trucks cartel Decision. The main question was the compatibility with the Damages Directive and the EU principle of effectiveness of a national procedural rule, which provides that, in the event the claim is upheld in part, costs are to be borne by each party and each party bears half of the common costs, except in cases of wrongful conduct. On this particular point, the Court’s judgment follows to a degree AG Kokott’s Opinion but not entirely. The Court makes a fundamental distinction between (i) the provisions of the Damages Directive that, in a way, codify and are derived from primary law (Article 101 TFEU) and the principle of effectiveness and (ii) other provisions of the Directive that do not relate to the EU principle of effectiveness (para. 33 et seq.). This distinction makes total sense. In other words, the Court means that the Damages Directive is the conscious choice of the EU legislator to create a new legal reality. But that does not mean that everything that existed prior to the Directive fails the EU principle of effectiveness. So, one thing is a particular provision in the Damages Directive, quite another thing is the corrective function of the EU principle of effectiveness. In the case at hand, the Court so no issue with the national rule in question, unlike AG Kokott who had advocated a more interventionist outcome (see my stories of July-September 2022). 

Fourth, there was Repsol. The Court here was dealing with the legal effect of NCA infringement decisions on actions for nullity. The Court explained first that the Damages Directive applies only to actions for damages, therefore actions for nullity are outside its scope (paras 30-32). In addition, the Damages Directive did not apply ratione temporis. Article 9(1) of the Damages Directive, which provides that final infringement decisions of NCAs constitute a non-rebuttable presumption of illegality, is substantive in nature, within the meaning of Article 22(1) of the Directive. In the case at hand, even if the action were an action for damages, the presumption rule would not have applied (para. 46). So the issue was whether as a matter of effectiveness NCA infringement decisions had to be given such a presumption power. The Court says yes, albeit not going as far as Article 9(1) of the Directive. In effect, the Court introduces a rebuttable presumption of illegality, thus shifting the burden of proof to the defendant, “provided that the nature of the alleged infringement that is the subject of those actions and its material, personal, temporal and territorial scope coincide with those of the infringement found in the said decision” (para. 67). Sensible solution. 

Finally, we had a couple of cases where the Court responded by means of an order (ordonnance) and not a full judgment (arrêt). This happens when the answer to the national court’s questions is obvious or there has been recently another judgment that has made the matter clear. In Dalerjo, the Court in effect repeated what it had already decided in Landkreis Northeim: that the Commission’s trucks cartel Decision must be interpreted such that its scope includes specialised trucks, including articulated dumpers. In Deutsche Bank, the Court thought that its Volvo/DAF judgment had given the answer as to the temporal applicability of the Damages Directive and the date the limitation period starts running.

1.5. Cases of relevance to public enforcement

In November 2022, we had a ruling in a preliminary reference from Romania: Zenith Media. The facts of the case are a bit complicated because of the somewhat complicated legal regime at the time applicable in Romania. It is interesting, however, because the Court mentions proportionality as a limitation to (but also as a guide for the exercise of) an NCA’s fining powers, when applying EU competition law. That was clearly not a ground-breaking judgment and this may explain the absence of an AG Opinion. And I think the judgment is not necessarily only defendant-friendly but rather neutral. It can also work to the benefit of the NCA. Para. 39 is clear in that respect: “A national competition authority must therefore have the possibility to examine the legal and factual merits of any evidence that could credibly demonstrate that the amount of turnover shown in the profit and loss account does not reflect the real economic situation of the undertaking concerned.” Essentially, the Court says that an NCA should be guided by substance and not form and that the two paramount guiding principles should be deterrence and proportionality. 

Far more important was a December 2022 judgment (of the Grand Chamber) in a non-competition case, which, however, has profound repercussions for public antitrust enforcement: Orde van Vlaamse Balies. Essentially, the Court has redesigned the scope of legal professional privilege (LPP) by rejecting a narrow scope and confirming that correspondence between lawyers (only EEA-admitted external lawyers or does it go further?  – that’s an open question) and clients should be protected not only under Article 47 of the Charter of Fundamental Rights of the EU (right to an effective remedy and to a fair trial) but also under Article 7 (respect for private life). In the previous state of affairs, LPP protected only those communications that were linked to the exercise of an undertaking’s rights of defence in the particular case at stake, at least that was the Commission’s position (to be fair to the Commission, in non-merger proceedings the Commission was always open-minded). Now, however, LPP is protected in all cases, irrespective of the matter or case at stake. Simply, the Commission (and NCAs) cannot touch lawyer-client correspondence and advice. So, the judgment is vastly consequential and certainly makes a difference in the everyday life of public enforcement. 

In the first six months of 2023, we also had interesting judgments. First, it’s worth mentioning a cartel case: HSBC. Nothing major. This is part of a long saga. It relates to the Commission’s 2016 Euribor cartel Decision, which was largely upheld by the General Court. However, the General Court had annulled the fine imposed on the basis that the statement of reasons was insufficient. The Court of Justice set aside the General Court’s judgment, in so far as it dismissed HSBC’s appeal. The Court adopted a strict approach regarding the presumption of innocence and found errors in the General Court’s reasoning. There were also other errors in the General Court’s assessment of what constituted a restriction “by object”. Then, the Court decided to proceed itself to the assessment of the applicants’ pleas at first instance. The rest was not as good for HSBC. The Court dismissed the original action and confirmed the applicants’ and appellants’ participation in the cartel at issue. 

Second, in French Supermarkets, the Court dealt with three cases: C-682/20 P, C-690/20 P and C-693/20 P (full disclosure: my firm was representing the appellants in one of the three cases). The Court’s judgments show how far the Court’s case law has gone in the area of due process. In all three cases, the Court found that the Commission is required to record any interview it conducts in order to collect information relating to the subject matter of an investigation. That obligation applies irrespective of whether the interview in question was conducted before the formal opening of an investigation, in order to collect indicia of an infringement, or afterwards, for the purpose of collecting evidence of an infringement. The Court added that the Commission may record the interviews in any form, including oral, thereby ensuring the effectiveness and speed of the investigation. In the cases at hand, the General Court had committed a legal error by holding that the obligation to record did not apply to the specific interviews in question, since they took place before the dawn raids, on the ground that no investigation had yet been formally opened in respect of those undertakings. The Court added that, in order to determine whether those interviews came within the scope of the Commission’s obligation to record, the General Court ought to have considered whether they were aimed at collecting information relating to the subject matter of an investigation, having regard to their content and context. Such an examination would have led to the conclusion that those interviews had to be recorded. Consequently, the Court set aside, in part, the judgments of the General Court and decided to give final judgment itself. Because of the above errors, the information derived from the interviews at stake was inadmissible. Since this information constituted the bulk of the evidence on which the Commission’s inspection decisions were based, the Court considered that the Commission did not possess sufficiently serious evidence to make the inspection decisions. 

This is a remarkable development! Clearly, the Court’s case law on due process has gone a long way. The new judgments also confirm that companies targeted by an inspection have the right to appeal any measure adopted during the raid that substantially affects their interests (C-693/20 P, para. 52). This aligns the case law of the Court of Justice to that of the European Court of Human Rights and is a welcome development for due process. 

Finally, we had the Amazon case in April 2023. The Court of Justice dismissed the appeal against the General Court’s dismission by order in 2021 of Amazon’s challenge of the carving out of Italy, when the Commission opened formal proceedings in the BuyBox case. Essentially, the question at stake was: Can the Commission open an investigation for the whole of the EU/EEA, while carving out a particular Member State, in order to allow the local NCA to continue its own investigation in the same case? Amazon thought this was a violation of Article 11(6) of Regulation 1/2003. And did this act produce effects adverse to Amazon and hence was it challengeable or not? The General Court thought otherwise and found the application inadmissible and the Court of Justice agreed. Full disclosure: I was representing Amazon in this case, so don’t expect too much critique here 😀.

The Court thought that the exclusion of Italy was a mere procedural step, which didn’t per se negatively affect Amazon’s legal position and defence rights; further, protection against parallel proceedings afforded by Regulation 1/2003 does not apply if the Commission has not initiated proceedings in respect of a given territory, like here, where it had carved Italy out. In addition, the Court held that Regulation 1/2003 does not imply any right for an undertaking to have a case dealt with in its entirety by the Commission and that the Commission enjoys broad discretion when it adopts a decision to open proceedings and defines the geographical scope of its investigation. All in all, the Court took a policy-oriented decision. It wanted to safeguard the Commission’s discretion when opening investigations. This is clear from para. 36 of the judgment. In fact, the reasoning of the Court is stronger on the point of the interpretation of Article 11(6) of Regulation 1/2003 (substance) than it is on the point of admissibility and production of effects. So Amazon lost, because to win would have meant that the Commission would be seriously constrained in exercising its competence to open cases and frame the scope of its investigation. Still, in my personal view, from a policy point of view and leaving the law aside, what happened here with the two investigations and ultimately decisions is not a high for EU competition law enforcement… ☹.

1.6. Other judgments

I would mention here American Airlines, where the Court dismissed an appeal against a 2020 ruling of the General Court that had dismissed American Airlines’ action for annulment against the 2018 Commission Decision to grant Delta Air Lines grandfathering rights over slots at Heathrow Airport in the aftermath of the 2013 American Airlines / US Airways merger clearance Decision. Not much to say here. We also had Super Bock, which, in my view, is not a ground-breaking judgment or, at least, it must not have been seen like that by the Court, otherwise we would have had an AG Opinion… Others, however, have viewed this in a different way. Anyway, it’s an interesting incremental development in the rich case law on what constitutes a “by object” infringement of Article 101 TFEU. The case was about resale price maintenance (RPM). While confirming that the concept of restriction of competition “by object” must be interpreted restrictively and that vertical restrictions are generally less harmful but can also restrict competition “by object”, the Court restated the Cartes Bancaires and Generics (UK) line of case law (paras 34-36). Interestingly, and this is probably where an element of novelty exists, the Court stresses that the concepts of “hardcore restrictions” in Block Exemption Regulations and “by object” restriction of competition are distinct. I refer and defer here to a leading commentator in this area, who draws the consequences from Super Bock and is certainly much more enthusiastic than me on the judgment’s importance.

II. Advocates General Opinions

2.1. AG Kokott’s CK Telecoms Opinion

In October 2022, AG Kokott delivered her Opinion in CK Telecoms. The case is possibly one of the most consequential in EU competition law. Certainly, the Commission was uncharacteristically outspoken against the General Court’s ruling in the same case. Essentially, the argument was that if the Court of Justice confirmed the General Court, that would be the end of EU merger control (I am exaggerating a bit)… This is an appeal brought by the Commission against the General Court’s landmark 2020 judgment overturning the Commission’s prohibition of the Hutchinson 3G UK / Telefónica UK mobile telecoms merger. Well, AG Kokott made the Commission quite happy. Her Opinion, which, as we will see in the near future, was followed by the Court, advised the Court to uphold the Commission’s appeal on mostly all main grounds and refer the case back to the General Court for reconsideration. 

AG Kokott first points out that this is the first case in which the Court has been given the opportunity to provide a ruling on the concept of “significant impediment to effective competition” (SIEC), in so far as it is based on non-coordinated (unilateral) effects, and to provide clarifications both concerning the standard of proof which the Commission is required to meet for the purposes of applying that concept and concerning the scope of the review of legality which the EU Courts must carry out.  

As to the judicial review standard, AG Kokott thought that this must be the same for the application of the SIEC concept, irrespective of the type of concentration concerned. In that regard, the Commission has a margin of discretion with regard to economic matters for the purposes of applying the substantive rules of the Merger Regulation. It follows that the review by the EU Courts of a Commission decision is confined to ascertaining that the facts have been accurately stated and that there has been no manifest error of assessment. 

As to the standard of proof that the Commission must meet to challenge a merger, the Merger Regulation does not impose different standards of proof with respect to decisions authorising a concentration and decisions prohibiting a concentration, since those standards of proof are perfectly symmetrical. In any event, the standard of proof for prohibiting a merger must be a “balance of probabilities”, as opposed to the General Court’s “strong probability” test. According to AG Kokott, the “strong probability” test would be too strict. In the light of various “conceivable” market scenarios, the Commission may prohibit a merger if it is “more likely than not” anti-competitive (paras 56, 57).  The same standard of proof should apply regardless of whether the merger results in the creation or strengthening of a dominant position, whether it is challenged based on unilateral effects in an oligopolistic market, or if the Commission advances a theory that is complex or uncertain. In addition, according to AG Kokott, contrary to the General Court’s findings, merging parties do not have to be “particularly close” competitors for the Commission to infer anti-competitive effects (para. 121), even if all competitors in the market are relatively “close”.  Similarly, while the term “important” competitive force implies that the company’s impact on competition should be “substantial” (beyond what is suggested by market shares or similar measures), contrary to the General Court’s position, it should not apply only to companies “competing particularly aggressively in terms of prices” or “mavericks” (para. 110).

2.2. AG Rantos’s Opinions in European Superleague and ISU

We are now in the tricky area of the application of competition law to sport. These Opinions were eagerly awaited. European Superleague is a preliminary reference case from Spain, where we have a very intense legal battle, while ISU is an appeal from the General Court’s homonymous judgment upholding the Commission’s Decision of 2017. Since the Opinions were not followed in the end, as we will see in the next edition, I am only staying at a high level. Both Opinions should be read together. 

In the first case, the facts have been extremely well-publicised and there is no need to analyse them. AG Rantos’s Opinion concluded that the FIFA-UEFA rules, under which any new competition is subject to prior approval, are compatible with EU competition law. Any restrictive effects are inherent in, and proportionate for achieving, the legitimate objectives related to the specific nature of sport pursued by UEFA and FIFA. For the Advocate General, the EU competition rules do not prohibit FIFA, UEFA, their member federations or their national leagues from issuing threats of sanctions against clubs affiliated to those federations, when those clubs participate in a project to set up a new competition which would risk undermining the objectives legitimately pursued by those federations. Likewise, the restrictions in the FIFA Statute concerning the exclusive marketing of the rights relating to the competitions organised by FIFA and UEFA are inherent in and proportionate to the pursuit of the legitimate objectives related to the specific nature of sport. Finally, EU law does not preclude the FIFA and UEFA Statutes which provide that the setting up of a new pan-European interclub football competition is to be subject to a prior approval scheme, since that requirement is appropriate and necessary for that purpose, taking into account the particular characteristics of the planned competition. 

The second case goes back to the Commission’s infringement Decision against the eligibility rules of the International Skating Union (ISU). The General Court upheld in part the Decision and annulled only the part that related to ISU’s arbitration rules (see my stories of November-December 2020). There was an appeal by ISU and two cross-appeals by the two skaters involved. AG Rantos proposed a somewhat dramatic turnaround, opining that the General Court’s judgment should be set aside and the case referred back. Essentially, AG Rantos took issue with the finding of “by object” infringement. He noted that the approach of the General Court was somewhat confused. The General Court initially followed the traditional approach of identifying a restriction of competition “by object”, by first analysing the content of the eligibility rules. However, when subsequently examining the objectives of those rules, the General Court appears to assess them in the light of the criteria laid down in Meca-Medina, which concerns the objectively justified nature of the restrictions of competition identified. That was not well-founded, since it would unduly extend the concept of “by object” restriction. Still, it had to be analysed whether the rules at issue have as their “effect” the restriction of competition within the meaning of Article 101(1) TFEU. Procedurally, this assessment could only take place before the General Court, so he proposed that the case be referred back. On the cross-appeal, he examined whether the Commission was entitled to classify the ISU binding arbitration mechanism as an element which “reinforced” the restriction of competition. In his view, since the Commission had not considered that the arbitration clause could constitute an infringement in itself, the General Court was fully entitled to recognise that agreeing to an arbitration clause as such did not constitute a restriction of competition. He also pointed out that recourse to arbitration can, in the present case, be justified by legitimate interests linked to the requirement that sporting disputes be submitted to a specialised dispute-resolution body. Such sport arbitration proceedings were not comparable to investment arbitration that took place between Member States and private parties and were at issue in the judgments in Achmea (in my humble view, one of the most unfortunate judgments ever) and PL Holdings. So the cross-appeal had to be dismissed.

2.3. Other Opinions

We also had a few other Opinions but I am not going to analyse them in any detail. In January 2023, AG Pitruzzella delivered his Opinion in the Lithuanian Notaries case, a preliminary reference from Lithuania (footnote: I had an indirect involvement before the Lithuanian Competition Council for the Chamber of Notaries, prior to the reference). The Advocate General considered whether notaries belonging to the Board of the Chamber of Notaries can be considered to have infringed Article 101 TFEU and be imposed fines. He sensibly concluded that Article 101 TFEU does not prevent notaries who are members of the Board being sanctioned with individual fines, in addition to the sanctions imposed on the Chamber, provided however that the notaries acted as undertakings, meaning that they made a specific and concrete contribution as undertakings, different and additional to that of the other notaries belonging to the Chamber. 

In March 2023, we had AG Rantos’s EDP Opinion. This is a Portuguese preliminary reference. It has to do with the application of Article 101(1) TFEU to a non-competition clause in a co-promotion agreement between a food retailer and an electricity supplier. Can this be a “by object” restriction of competition and could the two parties be described as “potential competitors”? The case gave AG Rantos an excellent opportunity to analyse both concepts and how the Court’s case law has developed over the last few years. We will see in the next edition that the Court followed the Advocate General. 

For the rest, it’s worth looking at AG Collins’s Altice Opinion of April 2023, which proposes the dismissal of the appeal against the General Court’s ruling that had confirmed the Commission’s Decision imposing a gun-jumping fine on Altice. The Advocate General considered that the entering into certain types of pre-closing covenants by an acquirer may constitute gun-jumping, regardless of the absence of the transfer of shares. AG Collins also emphasized the importance of deterrence when fixing the level of fines for gun-jumping. 

III. General Court 

We had a decent share of cartel cases before the General Court. In November 2022, the General Court rejected an appeal by steel producer Westfälische Drahtindustrie against a Commission letter requesting payment of interest stemming from the 2010 Decision in prestressing steel (the Commission later corrected some of the fine amounts). The Court confirmed that a 2015 judgment on appeal did not impose a new penalty and therefore interest was due since the imposition of the original fine.  

In December 2022, the General Court also rejected an appeal by Consorzio Cooperative di Produzione e Lavoro and its subsidiaries against the Commission’s 2020 Decision in retail food packaging. This was a re-adoption of an infringement decision, imposing a much lower fine than the original 2015 Decision, which was annulled by the General Court in 2015, due to lack of sufficient reasoning concerning the fine reduction granted to the applicants in the scope of the assessment of their application for inability to pay under para. 35 of the 2006 Fining Guidelines. This time, the applicants were not as lucky.  

Then, again in December 2022, the General Court rejected by order as inadmissible an action for annulment and damages filed by British Airways, which was seeking default interest on an annulled cartel fine. This was in the aftermath of the 2010 air cargo cartel Decision of the Commission that was annulled in 2015 on procedural grounds (in 2017, the Commission reimposed almost identical fines and further appeals are currently pending before the Court of Justice). This is a long saga… 

In January 2023, we had the GEA judgment, where the General Court rejected the appeal against a decision readopted by the Commission in one of the parts of the heat stabiliser cartel case. Another long saga and nothing major. 

Outside the area of cartels, we have (i) the Meta judgments (Case T-451/20 and Case T-452/20), where the General Court essentially confirmed the legality of “virtual data rooms” in antitrust investigations, (ii) RWE, where the General Court dismissed the action brought by the German electricity producer EVH against the Commission’s approval of the acquisition of E.ON’s generation assets by RWE, and (iii) Polwax, where the General Court dismissed the appeal against the Commission’s approval of Grupa Lotos/PKN Orlen.

IV. European Commission

4.1. Commission Decisions

In the period under review, I single out two Commission decisions: the Amazon commitments Decision in December 2022 and the Microsoft/Activision merger clearance Decision in May 2023. 

The Amazon commitments Decision relates to two separate antitrust investigations. The first case involved Amazon’s use of non-public data of its marketplace sellers (AT.40462 – Amazon Marketplace). The second case related to whether the criteria that Amazon sets to select the winner of the “Buy Box” and to enable sellers to offer products under its Prime Programme, lead to preferential treatment of Amazon’s own retail business or of the sellers that use Amazon’s logistics and delivery services (Case AT.40703 – Amazon Buy Box). 

To address the data use concern, Amazon proposed to commit: 

  • not to use non-public data relating to, or derived from, the independent sellers’ activities on its marketplace, for its retail business. This applies to both Amazon’s automated tools and employees that could cross-use the data from Amazon Marketplace, for retail decisions; 
  • not to use such data for the purposes of selling branded goods as well as its private label products. 

To address the “Buy Box” concerns, Amazon proposed to commit to: 

  • treat all sellers equally when ranking the offers for the purposes of the selection of the Buy Box winner; 
  • display a second competing offer to the Buy Box winner if there is a second offer from a different seller that is sufficiently differentiated from the first one on price and/or delivery. Both offers will display the same descriptive information and provide the same purchasing experience. 

To address the Prime concerns Amazon proposed to commit to: 

  • set non-discriminatory conditions and criteria for the qualification of marketplace sellers and offers to Prime; 
  • allow Prime sellers to freely choose any carrier for their logistics and delivery services and negotiate terms directly with the carrier of their choice; 
  • not use any information obtained through Prime about the terms and performance of third-party carriers, for its own logistics services. 

After the market test, Amazon amended the initial proposal and committed to: 

  • Improve the presentation of the second competing Buy Box offer by making it more prominent and to include a review mechanism in case the presentation is not attracting adequate consumer attention; 
  • Increase the transparency and early information flows to sellers and carriers about the commitments and their newly acquired rights, enabling, amongst others, early switching of sellers to independent carriers; 
  • Lay out the means for independent carriers to directly contact their Amazon customers, in line with data-protection rules, enabling them to provide equivalent delivery services to those offered by Amazon; 
  • Improve carrier data protection from use by Amazon’s competing logistics services, in particular concerning cargo profile information; 
  • Increase the powers of the monitoring trustee by introducing further notification obligations; 
  • Introduce a centralised complaint mechanism, open to all sellers and carriers in case of suspected non-compliance with the commitments. 
  • Increase to seven years, instead of the initially proposed five years, the duration of the commitments relating to Prime and the second competing Buy Box offer. 

The Commission accepted Amazon’s final commitments and proceeded to making them binding. If anything, the case is very interesting, since it can be described as a DMA case before the DMA’s application … 

The Microsoft/Activision conditional clearance Decision is another important development. Unlike the original decision of the UK Competition and Markets Authority (CMA) to prohibit the deal, the Commission proved its more business-like attitude by conditionally clearing the acquisition. The Commission’s initial concerns were that Microsoft could harm competition (i) in the distribution of console and PC video games, including multi-game subscription services and cloud game streaming services; and (ii) in the supply of PC operating systems. In Phase II, the Commission dropped its concerns about harm to rival consoles and rival multi-game subscription services but confirmed that Microsoft could harm competition in the distribution of games via cloud game streaming services, an innovative market segment that could transform the way many gamers play video games, and that its position in the market for PC operating systems would be strengthened. 

To address the competition concerns identified by the Commission in the market for the distribution of PC and console games via cloud game streaming services, Microsoft offered the following comprehensive licensing commitments, with a 10-year duration: 

  • A free licence to consumers in the EEA that would allow them to stream, via any cloud game streaming services of their choice, all current and future Activision Blizzard PC and console games for which they have a license. 
  • A corresponding free licence to cloud game streaming service providers to allow EEA-based gamers to stream any Activision Blizzard’s PC and console games. 

According to the Commission, these licences will ensure that gamers that have purchased one or more Activision games on a PC or console store, or that have subscribed to a multi-game subscription service that includes Activision games, have the right to stream those games with any cloud game streaming service of their choice and play them on any device using any operating system. The remedies also ensure that Activision’s games available for streaming will have the same quality and content as games available for traditional download. For the Commission, the remedies would empower millions of EEA consumers to stream Activision’s games using any cloud gaming services operating in the EEA, provided they are purchased in an online store or included in an active multi-game subscription in the EEA. In addition, the availability of Activision’s popular games for streaming via all cloud game streaming services will boost the development of this dynamic technology in the EEA. 

This was clearly the merger control global drama of the year! The repercussions of the UK prohibition decision and – I would add – the overall semantics were not good for global convergence, notwithstanding the CMA’s final u-turn. Convergence, of course, is not an aim in itself but is important for the legitimacy of action of competition authorities. We have always had a degree divergence in merger control outcomes, but what made this case particularly bitter was the semantics accompanying it.

4.2. Policy Developments

I mention en passant the adoption in October 2022 of the revised Informal Guidance Notice that allows businesses to seek informal guidance on the application of EU competition rules to novel or unresolved questions. The new Notice replaces the 2004 one and could have been more ambitious, since the changes are very marginal. 

The more ground-breaking development came in March 2023, when the Commission decided to revisit its 2008 Guidance Paper on exclusionary abuses and kick-started a process of consultation with a view to adopting formal Guidelines by 2025. I was hearing rumours about DG COMP “cooking” something in this area, presumably because the Guidance Paper’s enemies within the Commission have grown over the last years – not counting the Legal Service, which has been negative from the very beginning. Times have changed… 

The Commission proceeded to a surgical amendment of the Guidance Paper and withdrew certain parts of it. In my view, as far as the so-called Amending Communication is concerned, in a nutshell, the changes are more important in terms of symbolism than real substance. Starting with symbolism, the first paragraph of the Communication sounds like a departure from “consumer welfare” as guiding principle and talks of a competition enforcement that “serves Europe well by contributing to a level playing field where markets serve consumers [and that] can moreover contribute to achieving objectives that go beyond consumer welfare, such as plurality in a democratic society”. In terms of substance, if one reads the Annex accompanying the Communication, one will see that the changes are minor: apart from some specific parts where it was already clear that the Guidance Paper was not in conformity with the case law (e.g. the treatment of margin squeeze), the changes are mainly three: 

  1. Less efficient competitors may also be important for the competitive process (we knew as much – the original Guidance Paper was also clear about that – see para. 24). 
  2. The Commission does not have to but simply may run an AEC test in pricing abuses (again, we knew as much from the recent case law – but we also know that if a party puts it forward the Commission must rebut it). 
  3. The Commission introduces a separate category of abuse, “constructive refusal to supply”, which should not be analysed as normal refusals. 

The Commission also published a Policy Brief with the title “A dynamic and workable effects-based approach to Article 102 TFEU” which essentially includes its new manifesto. The Policy Brief makes it clear that the Commission’s aim is to recede from too much economics in the operational tests of the abusive practices at stake. In a much-discussed LinkedIn post, I prepared a critical “annotation” of the Policy Brief with the points where I think it is not in conformity with the most recent case law. I refer the readers to my “annotation”. 

I am sure that the Commission accepts in the broad sense the so-called “effects-based approach” in Article 102 TFEU (how could it not? – the case law is very clear), but I also expect more presumptions and less numerical/quantitative economic tests. It does not matter so much, but the explanation behind the change is probably a number of defeats for the Commission before the EU Courts in pricing abuse cases, starting from the Intel case and going to Qualcomm and the partial annulment in Google Android. Still, if that is the explanation, it strikes me as a wrong incentive. The Commission, notwithstanding these defeats, continues having an enviable record in judicial review proceedings. Besides, what counts should not be the Commission’s success rate in Luxembourg but rather whether its enforcement relies on sound legal and economic bases and is effective and beneficial to consumers. By mid-2024, we should have the Commission’s draft Guidelines, so we will know more… 

Finally, in June 2023, we had the long-awaited revised Horizontal BERs on Research and Development and Specialisation agreements, accompanied by revised Horizontal Guidelines. The revised rules introduce a number of changes. 

  • First, the scope of the Specialisation BER is expanded to cover more types of production agreements concluded by more than two parties. In addition, the revised rules introduce a more flexible approach for the calculation of the market shares for the purpose of applying the block exemption. They also include specific guidance on how to apply the latter.  
  • Second, there is more clarity and flexibility as regards the calculation of market shares for the purpose of applying the R&D BER. The revised rules give more prominence to the protection of innovation competition, especially in cases where it is not possible to calculate market shares, and in this context emphasises the powers of the Commission and NCAs to withdraw the benefit of the exemption in individual problematic cases.  
  • Third, the introductory chapter of the Guidelines has been updated with the latest case law on key concepts such as “concerted practices”, “potential competition”, restrictions “by object” and “by effect”, and ancillary restraints. This chapter also contains new guidance on the application of Article 101 TFEU to agreements between joint ventures and their parent companies, as well as expanded guidance on how to apply the Guidelines  to agreements that involve cooperation on more than one type of activity (e.g., production and commercialisation). 
  • Fourth, there is a new section on Mobile Telecommunications Infrastructure Sharing Agreements reflecting recent enforcement practice in the chapter on Production Agreements. This new guidance sets out factors relevant for the assessment of these agreements and includes a list of minimum conditions that companies must comply with to reduce the risk of infringing the competition rules. 
  • Fifth, there is an updated and expanded chapter on Purchasing Agreements to reflect recent case practice. The revised chapter explains the distinction between joint purchasing and buyer cartels. It also clarifies that joint purchasing extends to arrangements whereby buyers negotiate purchase conditions jointly, but each buyer makes its purchases independently. It also gives more prominence to possible anti-competitive effects on the upstream supply side and provides guidance on certain joint negotiating tactics, including the use of temporary order stops. 
  • Sixth, there is a new section on bidding consortia and guidance on the distinction with bid rigging. 
  • Seventh, there is a substantially reworked chapter on Information Exchange including additional guidance on: (i) the concept of commercially sensitive information; (ii) the types of information exchange that may constitute restrictions of competition “by object”; (iii) potential pro-competitive effects of data pools; (iv) indirect forms of information exchange, including hub-and-spoke arrangements; (v) anti-competitive signalling via public announcements; and (vi) practical measures that companies can take to avoid infringements, such as limiting the scope of the exchange, using clean teams or independent trustees and public distancing. 
  • Eighth, the new chapter on Standardisation Agreements offers greater flexibility regarding the requirement of open participation in the standard-setting process. The revised chapter also clarifies (i) that the disclosure by parties to a standardisation agreement of a maximum cumulated royalty rate is not anti-competitive; and (ii) the requirement for participants to disclose relevant intellectual property rights. 
  • Finally and perhaps more interestingly, there is a new chapter covering Sustainability Agreements to clarify that the antitrust rules do not stand in the way of agreements between competitors that pursue a sustainability objective. The new guidance contains a broad definition of sustainability objectives and lists various examples of sustainability agreements that generally fall outside the scope of Article 101(1) TFEU. The new rules also provide a soft safe harbour for sustainability standardisation agreements that meet certain conditions. They also clarify how a sustainability agreement can be exempted by describing types of benefits that may be taken into account. The new chapter further contains hypothetical examples illustrating the application of Article 101 TFEU. It also reminds companies wishing to enter into a sustainability agreement that they can request informal guidance from the Commission. The provision of such guidance may complement the general framework of analysis set out in the new sustainability chapter. 

Makis Komninos


Citation: Makis Komninos, Competition Stories: October 2022 – June 2023, Network Law Review, Winter 2023.

Read all the competition stories over here: link

Related Posts