On Thin Ice: The Court’s Judgment in Case C-124/21 P, International Skating Union v Commission

Abstract: This judgment provides the foundations for applying art. 101 TFEU to rules or sporting organisations related to competing organisations. At hand was a rule that banned athletes from competing in non-approved skating competitions. The Commission found this rule to be a restriction of competition by object, which was largely upheld by the General Court. This judgment largely confirms the Commission’s assessment and sheds light on what sporting organisations can do within the bounds of competition law. It is a highly relevant judgment not only for those interested in the interaction between sports regulation and competition law, but also for people with a more general focus on competition law as it contains several clarifications and innovations as regards the interpretation of art. 101 that have implications beyond sporting organisations and may also impact the application of art. 102. Finally, the Court sheds some light on the compatibility of arbitration with the EU legal order.

Keywords: competition law – sporting organisations – prior authorisation of competing events – restrictions by object – multisided platform – arbitration.

I.   A December revolution

Just like the “November Revolution” of 1993,[1] we may well have just witnessed a December revolution.[2] In a series of ground-breaking Grand Chamber judgments, the Court has laid down, and to certain extend refined, the foundations for the way EU law applies to sports organisations.[3]

All three cases deal with the powers of sporting organisations to lay down the rules for the sporting competitions they organise and in particular the question whether and to what extent such rules can exclude competing competitions. The three cases all had distinct factual and procedural backgrounds, with ISU being the only direct appeal resulting from what started with a complaint by two professional skaters who considered that the rules adopted by the International Skating Union (ISU) infringed arts 101 and 102 TFEU.

Despite their difference, all three cases share the fact that they are Grand Chamber judgments that are reserved for cases that are difficult, important or involve particular circumstances.[4] Given that neither the judgments nor the opinions of the Advocate’s-General mention any particular circumstances, these cases must be exceptional in view of the difficulty or importance. This makes them very interesting and quite possibly foundational. The importance is further corroborated by the astounding number of members states that intervened in the European Superleague Company (hereinafter ESL) case that deals with the essentially analogous problem at issue in International Skating Union v Commission (hereinafter ISU) [5]. In short, this judicial triplet has the hallmarks of being a series with which the Court wants to set the foundational rules for the regulation by sports organisations. Confining myself to the judgment in ISU, for me the question remains whether it was only ISU that found itself on thin ice, or also the Court. At least one of the three judgments in the 1993 “November Revolution”, Keck and Mithouard[6], was controversial from the beginning and proved to be perhaps a revolution but, but one that is broadly conceived as a mistake that the Court spent significant time correcting or at least limiting. As one of a series of judgments that deal with the general question to what extent and by which doctrines EU law intervenes in national, private or hybrid organisation of markets, ISU stands out as a uniquely detailed guide to the way the Court envisages this revolution within the confines of art. 101 TFEU.[7]

The rules in the dispute related to the prior authorisation of other skating events and the eligibility of athletes in ISU-organised events. The prior authorisation rules are quite predictably named in that they provide for a procedure to obtain the prior approval of ISU for any competition that involves ISU members. The organisers of such a competing competition must submit to ISU financial, technical, commercial and sporting information (including the venue of the planned event, value of the prizes to be awarded, business plan, budget, television coverage), as well as a declaration confirming that it accepts the ISU’s Code of Ethics. The eligibility rules concern the athlete’s eligibility to compete in an ISU organised event. In a nutshell, an athlete can lose this eligibility, for a specific period or for life, when they compete in a non-approved event. Prior to 2016, the eligibility rules mentioned that they are “made for the adequate protection of the economic and other interests of the ISU, which uses its financial revenues for the administration and development of […] sport disciplines and for the support and benefit of [its] Members and their Skaters”.[8] Since 2016, however, the eligibility rules no longer refer to the economic interests and instead mention that “the condition of eligibility is made for adequate protection of the ethical values, jurisdiction objectives and other legitimate respective interests” of that association, “which uses its financial revenues for the administration and development of the ISU sport disciplines and for the support and benefit of [its] Members and their Skaters”.[9]

The bottom line of both rules combined is simple: an athlete will need to choose between participating in ISU events or participating in non-approved third-party events. This in turn makes it rather difficult, if not impossible, to have a skating event without ISU’s prior approval, given that ISU organises the most and most important skating events[10] and athletes – which are the vital input for any sporting competition – are disincentivised from taking part in alternative competitions as they risk not being eligible for ISU events.

This is where competition law kicks in. The effect of such rules is to create an ISU-exclusivity for the athletes, meaning that they are in practice bound to compete only in ISU-organised and ISU-approved third-party events. Doesn’t this amount to a complete exclusion of all competing competitions? In fact, as the complainants pointed out, not a single third-party organised event was actually approved, unless it was ultimately taken over by an ISU national member organisation.[11]

The central issue in ISU, therefore, is whether and under what conditions, a sporting organisation can organise its activity in such a way as to exclude (nearly) all competing competitions.

II.  The competition law of sports competition regulations

Competition law is relevant to the regulation of sports competitions in a number of ways. Focussing on art. 101, we first encounter the question whether competition law applies to such regulations in the first place. This is often presented in connection with art. 165 TFEU. If the competition rules apply, the next question that presents itself is whether the regulations qualify as a restriction of competition by object or by effect. Finally, if there is no restriction by object, the question must be answered whether the regulations are justified by legitimate objectives in the public interest. The means that there are essentially three stages to the application of art. 101 to the sports regulations and in ISU the Court devotes quite a few paragraphs to explaining and structuring these stages and the questions involved, largely following the structure suggested by the Advocate General.[12] These paragraphs contain a detailed and structured exposé of the Court’s understanding of competition law and thus set the scene for the actual consideration of the appeal and cross-appeal.

a) As regards the most fundamental issue, whether the competition rules apply altogether, the Court repeats its earlier and well-established case law in paras 91 to 95 of ISU. The bottom line is that “[o]nly certain specific rules which were adopted solely on non-economic grounds and which relate to questions of interest solely to sport per se must be regarded as being extraneous to any economic activity”.[13] It is obvious that this sporting exclusion is conceived narrowly and can be contrasted with all activities that qualify as “economic”, meaning that the entities involved are “undertakings” and/or “associations of undertakings” as regards these activities.[14] This is pretty orthodox competition law, even when the Court finds itself conflating the nature of the activities with the entity involved, as will be pointed out further below. The so-called “sporting exception” that some find in art. 165 TFEU, is irrelevant at this stage.[15] Reflecting upon this, it seems that the ISU court is reviving, in theory at least, the “purely sporting rule” it had largely condemned to irrelevance in Meca-Medina.[16] However, being conceived as narrowly as it is, it is unlikely to be of much practical significance.

b) The sporting provisions are, however, referred to when the Court holds that the “specific characteristics of an economic sector” may be taken into account when determining whether the regulations at hand restrict competition by object or by effect.[17] We know that deciding whether there is a restriction by object or effect requires what can be called a context-test. This means that the conduct of the sporting associations must be examined in the “economic and legal context of which it forms part”. As part of this context analysis, account may also be taken of, inter alia, the nature, organisation or functioning of the sport involved and how professionalised it is.[18]

The object-effect dichotomy is then operationalised by the Court by first setting out its standard case law on object and effect being alternatives and the need to interpret the “by object” concept narrowly.[19] The threshold for the “by object” classification is that the conduct must be “particularly harmful to competition” which, according to the Court, is the case for types of conduct that are “liable to lead to price increases or fall in production and, therefore, more limited supply, resulting in poor allocation of resources to the detriment of user undertakings and consumers”.[20] In this regard it is interesting to see that the Court construes this threshold for the “particular harmfulness” on the basis of what is essentially a consumer welfare standard that relies ultimately and uniquely on harm to “user undertakings and consumers”. This raises the question what exactly is meant with “user undertakings”? To the best of my knowledge and search abilities, this concept was not used before by the Court. In this context, I assume that it was used by the Court to take account of the fact that ISU (and the national skating organisations) also cater to the needs of professional skaters, broadcasting companies and other undertakings that are in a purchase (and supply) relation with ISU. This seems the clearest indicator that ISU is treated as a platform.[21]

This means that the lens through which the harm is established in ISU is one that relies on consumer welfare effects alone in a significant departure from the Court’s earlier case law. In that regard the Court’s reference to the earlier judgments in BIDS[22] and ING Pensii[23] does not provide authority for this restriction.[24] None of the paragraphs in these judgments referred to in ISU nor the facts of those cases support this restriction. The Court’s reference to para. 51 of CB v Commission[25] in fact rather points at a wider category of harm as a lens where it refers to “poor allocation of resources to the detriment, in particular, of consumers”. The fact that harm to consumers is a particular category of harm to be taken into account shows that, for example, harm to the market structure can also be relevant to find a restriction by object. This connects to the sometimes-voiced criticism that EU competition law values competitors over competition.[26] Whilst this criticism is voice most frequently and vociferously in connection with merger control and art. 102, it is also relevant to art. 101 as ISU shows. The observation that ISU effectively excluded a competing competition thus raises the question whether and to what extent such exclusion in and of itself warrants a finding that the practice is restrictive by object? Following ISU and its focus on consumers and user undertakings, it seems that the exclusion as such is irrelevant in this regard. However, as we will see, the Court has cast the net of the effects for consumers and user undertakings so wide that it will also catch exclusionary effects as such.[27]

The Court then goes on to mention reverse payment practices and the fixing of prices by means of a recommendation by an industry association.[28] Reinforcing the focus on consumer welfare as the onus of harm, these are cases that manifest clear consumer harm. Having thus set the scene, the Court provides us with a structured context-test. This structure comes from the fact that the Court has introduced a specific order for the sub-tests of the context test: first the content of the practice, second the economic and legal context and third, the objectives are to be examined.[29] Of course, given that the outcomes of the sub-tests are not interrelated, this explicit structure does not add anything significant to the analytical framework, but still it is a nice attempt at providing a clearer framework.[30] The Court adds to this – slightly – clearer framework by explaining that the analysis of the economic and legal context is not the same as examining the effects on competition.[31] That is of course completely true and follows from the logic that uses the context-test to determine whether an effects analysis can be omitted. However, the gnomic nature of this statement adds very little to the often-voiced criticism that “taking into consideration the nature of the products or services concerned, as well as the real conditions of the structure and functioning of the sectors or markets in question”, as prescribed by the Court for the context test, looks, barks and smells a lot like an effects-analysis, but apparently it isn’t.[32] It certainly raises the question what it does entail exactly and in that regard, the Court remains silent. What depth of market analysis is required and what kind of rebutting evidence can be put forward at the stage of the context-test? Require too much depth in the analysis of the economic context and counter-evidence and the by object-classification as a short-cut to avoid a fully-fledged counterfactual analysis will become meaningless. Put the bar too low, and we may well risk having false positive by object classifications. The latter, however, should not be a problem given the presence of the third paragraph in the greater scheme of art. 101 TFEU. Whether the accused undertakings need to defend themselves by adducing evidence that the coordination was not particularly harmful for consumers and user undertakings as part of the context test does not meaningfully differ from adducing largely analogous evidence to substantiate that there is economic and technical progress with a fair share for the consumers.

A final point to be noted in this regard is that the Court did not spend any further words on how exactly the “specific characteristics” of the sector at hand are to be taken into account as part of the analysis of the economic and legal context.[33]

The net result of this is that we have a test relying on three subtests, of which the exact content is unclear as regards the economic and legal context, all outcomes of which have to be taken into consideration[34] to meet a threshold that is fundamentally unclear. This does not bode well for legal certainty/predictability and – depending on where the threshold for a sufficient degree of harm is put – competition law altogether.

c) The third stage in the analysis relates to the effects on competition. It comprises all conduct that cannot be qualified as restriction by object and requires actual or potential appreciable effects to be shown by means of a counterfactual analysis.[35] In this regard the Court again refers to settled case law according to which not all agreements that limit the freedom of action also amount to restrictions of competition. This is followed by a quick recap of the case law on restrictions of competition that are inherent in and objectively justified on the basis legitimate objectives in the public interest.[36] I’d like to refer to this line of case law as the Wouters-exception after the first case where it was applied and relying on the fact that it is an exception to the scope of art. 101(1) TFEU.

Whilst this is again perfectly orthodox,[37] the Court does manage to introduce some unclarity as regards the precise place of the Wouters-exception in the structure of art. 101(1). Orthodoxy has it that it is part of the effects analysis, and therefore follows a context-test. This is also what the Court explicitly holds in para. 113. However, in para. 111 the Court introduces the Wouters-exception as part of the “examination of the economic and legal context”. If this refers to the same context-test that is prescribed to determine whether conduct can be qualified as a restriction by object, that would mean that the Wouters-exception would need to be considered twice: first to determine whether the by-object categorisation is warranted and – if this is not the case – then to examine whether there are appreciable restrictive effects on competition. That does not make any sense and confirms the criticism mentioned above that this line of case law conflates the effects-analysis and the context test.[38] Again, this lack of clarity is not conducive to legal certainty and predictability.

Observing the issues at hand from a slightly greater distance, the question of the third paragraph appears at the horizon. It is obvious that there’s a lot of balancing and weighing going on the first paragraph. Whether this is as part of the context-analysis or as part of establishing appreciable effects on competition, balancing pro- and anticompetitive elements and more generally defined public interests is all supposed to happen as part of the administration of art. 101(1). Why, however, not use the third paragraph at least for the Wouters-type of balancing? The benefits of deontology or the proper organisation of a sporting platform can arguably be subsumed under the heading of economic and technical progress or improving the production or distribution of goods and establishing a fair share for consumers should not be impossible either. The proportionality assessment and requirement of sufficient residual competition are part and parcel of the balancing as part of the first paragraph. Not mentioning the third paragraph of art. 101 at all, except as the last and only option to legalise a by object restriction,[39] is puzzling for two reasons. Firstly, why does the Court want to put so much into the first paragraph, even if this flies in the face of the clear legal structure of art. 101? Secondly, what possible benefits could save an agreement that is – to paraphrase the standard for a by object restriction – particularly harmful to consumers, especially if these consumers are to have a “fair share” in these benefits? The answer to the first question lies with the Court. Only the Court knows why, even following the decentralisation of art. 101(3), it has continued to expand the balancing and weighing to be done in the first paragraph. As an answer to the second question, one example I could think of are sustainability agreements that concern the internalisation of environmental costs and include a pass-on of such costs in the price of the products involved.[40] That would amount to horizontal price fixing, which epitomises a by object restriction. Yet, there may be good reasons for such restrictions and, depending on what costs and benefits are included, the net result may even be beneficial for the consumers involved.[41] However, these are all costs and benefits associated with the (over-)consumption of the good involved. In relation to sports competitions, there is no single good to be consumed, if only because there are at least two categories of consumers involved: the athletes and spectators/viewers.

III. ISU: a restriction by object trough two turns

Having thus set the scene, the Court then deals with the actual appeal and cross-appeal. ISU essentially argues that the General Court incorrectly interpreted the concept of a restriction by object by applying the MOTOE[42] and OTOC[43] jurisprudence.[44] This line of case law deals with undertakings and/or associations of undertakings that have been endowed with special or exclusive rights to control competitors’ access to same market on which they are also active themselves. The MOTOE/OTOC jurisprudence thus relies on the need to have equality of opportunity in order to not “prevent the growth of competition therein to the detriment of consumers by limiting production, product or alternative service development or innovation”.[45]

The Court finds that the MOTOE/OTOC jurisprudence does apply and needs only two turns to reach this conclusion.[46] The first is finding that the controls that should apply to the power held by an undertaking to exclude competitors, apply irrespective of whether this power emanates from the grant of a special or exclusive right or comes from the autonomous behaviour of an undertaking in a dominant position.[47] The second is a duty to interpret art. 102 and 101 consistently, meaning that what is an infringement of art. 102 “by its very existence” qualifies as a restriction by object under art. 101.[48]

The applicability of MOTOE/OTOC then translates into a “level playing field rule” according to which the powers of ISU to exclude competitors must be subject to control to avoid amounting to restrictions by object. More specifically, this requires transparent, clear and precise criteria that are to be applied in a non-discriminatory manner subject to proportionate sanctions whilst being capable of being subject to effective review.[49]

I find this one of the most troubling and troublesome parts of ISU. First, the equality of opportunity-line of cases was developed exclusively in relation to art. 102 as it is applied in connection with art. 106 TFEU. RTT v GB-Inno-BM[50] (hereinafter GB-Inno-BM), Raso[51] and MOTOE, the cases that the Court refers to, all deal with so-called public undertakings that benefitted from state-sponsored exclusive rights that enabled them to restrict the market entry of their competitors on downstream markets. A reference to paragraph 25 of GB-Inno-BM suffices in this regard:

“A system of undistorted competition, as laid down in the Treaty, can be guaranteed only if equality of opportunity is secured as between the various economic operators. To entrust an undertaking which markets terminal equipment with the task of drawing up the specifications for such equipment, monitoring their application and granting type-approval in respect thereof is tantamount to conferring upon it the power to determine at will which terminal equipment may be connected to the public network, and thereby placing that undertaking at an obvious advantage over its competitors” [emphasis added]

It is clear that the Court finds fault with the Member State who entrusts the power over market access to the same entity that is also active on the market. This rule is then said in para. 126 of ISU to apply “irrespective” of whether the power of market access results from the state intervention or the autonomous behaviour of the undertaking. The Court thus equates public power over market access with private power over market access. This first turn is based exclusively on GB-Inno-BM and paras 17 to 20 and 24 in particular. The only problem is that RTT, the monopolist in GB-Inno-BM, did not derive the power over market access from its autonomous behaviour. It was entrusted with this power by means of the Ministerial Order of 20 September 1978.[52] Obviously, the mere fact that was no private power in the first place does not exclude that the Court actually conceived a more general rule in GB-Inno-BM. For that we’ll need to analyse that judgment and the specific paragraphs in greater detail. Disconcertingly, however, paras 17-20 do not substantiate the Court’s first turn in any way. Paras 17 to 20 indeed refer to the extension of dominance as a form of abuse. That, however, cannot serve as authority for holding that it doesn’t matter whether that extension results from public or private power.[53] Even if we were to ignore this, the difference still remains between abuse in the form of extending a dominant position to a neighbouring market, as in GB-Inno-BM, and the abuse that exists in the combination of having the power to control access to a market on which the company is also active. This is exactly the point that the Court makes in para. 24 of GB-Inno-BM where it focusses on the extension of dominance, whether by the company or by the member state.[54]

The second turn, consisting of the consistent interpretation of arts 102 and 101 TFEU, is similarly problematic. The Court acknowledges the difference in objectives and scopes, yet infers from the fact that both can apply simultaneously to the same conduct that they must be interpreted consistently. This duty of consistent interpretation is simply not logically related to the fact that the two provisions can apply simultaneously. The main reason for this is that the two provisions address fundamentally different phenomena. Art. 101 TFEU deals with the coordination that affects the competitive constraints that would otherwise have existed between the undertakings involved in the coordination, whereas art. 102 addresses unilateral actions that result from an absence of competitive constraints. Where they apply simultaneously to the same facts, that parallel application can be likened to competition law applicable to the internal functioning of a cartel and that to the external manifestation of the cartel.[55] Whilst it is true that a particular form of harm to competition, say the exclusion of a competitor, could arise from both a 101- and a 102-relevant action, The fundamental perspective is different. In a 101-case, the Court has told us to look at the restricted intra-brand competition in its relation to the increase in inter-brand competition as it manifests itself between the parties.[56] In a 102 case, the very point of departure is the absence of a competitive restraint enshrined in the concept of dominance that enables the exploitation or reinforcement of that lack of competitive pressure.

Moreover, if we accept the consistent interpretation, logic would dictate that, just like there are by object restrictions of art. 101, there must also be by object abuses and that is something that, at least according to AG Rantos in Servizio, doesn’t exist.[57] That being said, the Court does reach the conclusion that the power over market access infringes by its very existence Article 102 TFEU.[58] This does seem to denote exactly that: something that qualifies as abuse just because it exists, not because of any effects it has. Of course, the attentive reader will notice that I have omitted that the Court considers this power by object abuse “when it is conferred on the dominant undertaking” and that is exactly what happened in MOTOE and all 106 in connection with 102 cases that the Court refers to. Perhaps the Court should have been more explicit and have written “when it is conferred by a member state on the dominant undertaking”. To be a bit more explicit: the MOTOE/OTOC jurisprudence makes sense precisely because it is the member state’s duty to ensure the full effectiveness of art. 102 when it defines the scope of the exclusive rights and in particular whether that scope should also include power over market access. This further shows why the transplant is problematic.

It further raises the question what the consequences are of this act of (auto)conferral of the power over market access? We know the consequences of having this power: to avoid infringing competition law the dominant undertaking must ensure that it wields this power subject to transparent, non-discriminatory and proportionate conditions subject to some form of review. When should, for example, Google have realised that it had created for or conferred on itself the power over market access, as it had in Google Shopping[59] or Google Android?[60]

Also, within the legal context of art. 101 TFEU, several questions arise. For one, we know that the concept of a restriction by object is to be interpreted narrowly.[61] How does this then impact the assessment of the clarity, transparency and, perhaps most vexingly, non-discriminatory nature of the criteria for exercising the power over market access held by ISU? In relation to the non-discrimination requirement, the Court adds that the criteria should not make it “impossible or excessively difficult” for the competing organisation to comply. It doesn’t take much imagination to envisage a criterion that is easy to comply with for one organisation and excessively difficult for another.[62] What should be the threshold for a court or competition authority to find a condition discriminatory so that it can be struck down as a by object restriction? A judgment like this is easy if there are no rules or procedures at all, like in ISU.[63] But what if there are rules, but the newcomer considers them unduly burdensome?

As sympathetic as I am to the outcome, I think that ISU is fundamentally flawed, both as regards the doctrinal basis for transplanting the MOTOE/OTOC rule to a context without public law interventions and concerning the implementation of the “level playing field rule”. This is further exacerbated when we consider sports competitions as a platform.

IV. ISU as a platform case and paradigm shift

I’ve already drawn an analogy with Google Shopping and when commenting on the opinion in the roughly analogous ESL case referred to sports organisations as platforms[64] and I’ll continue that. As far as I’m concerned, an organised sports competition is just another platform that connects several markets. Athletes are connected to spectators who are connected to event organisers and broadcasters who are connected to sponsors and the platform to connect all of them is the sporting organisation. Some platforms are bigger than others – ISU considers itself a niche sporting discipline[65] – but the bottom line is identical for all platforms, as are the possibly anticompetitive strategies: leveraging the power over input on one side of the multi-sided market to other sides.[66] Understanding a sporting competition as a platform also allows for an intuitive understanding of why they wish to be monopolists: there is a limited amount of profit to made in that market. For sure, a newcomer offering an innovative format for skating competitions could attract new spectators, athletes and advertisers and thus extra income, but it would also attract income from the existing competition. The cake may get a bit bigger, but it will certainly need to be shared between two or more parties, likely resulting in a smaller slice. That may well be a slice that is so small that the sport cannot be organised profitably anymore. This could in turn trigger increases in the contributions by national sporting organisations that are member of ISU which may ultimately mean that the contributions by individual (amateur) athletes will increase, perhaps even resulting in that talented young skater abandoning the sport. Of course, ISU could also, in attempt to compensate the smaller slice of the cake, try to sell broadcasting rights for a higher price, but that may well mean that the broadcasting rights will not be bought, meaning that our young talent will never be inspired to train as hard as her idols. The point I’m making is that organising sports competitions invariably bears all the hallmarks of platform activities driven by positive network externalities. Whether and to what extent the positive network externalities are also actually implemented in the form of, inter alia, advertising income for international events being used to encourage young talents, I’m not sure, but that is part of the logic often put forward by sporting organisations. At the end of the day, a platform needs all sides to be on board and may and will chose to allocate costs as well as redistribute income between these sides. This means that the negative attention for ISU’s economic interests, that were an integral part of the regulatory framework for approving competing competitions, is unwarranted.[67] The economic interests of ISU (the slice of the cake) cannot be meaningfully separated from the sporting interests involved.[68] If ISU is a platform, the ISU case should be treated as a platform case.

It is striking to see that the Court does not address this platform-nature of the activity at all. The closest we get is para. 146 that warrants integral citation in this regard:

“Those rules are thus able to be used to allow or exclude from that market any competing undertaking, even an equally efficient undertaking, or at least restrict the creation and marketing of alternative or new competitions in terms of their format or content. In so doing, they also completely deprive athletes of the opportunity to participate in those competitions, even where they could be of interest to them, for example on account of an innovative format, while observing all the principles, values and rules underpinning the sporting discipline concerned. Ultimately, they are such as to completely deprive spectators and viewers of any opportunity to attend those competitions or to watch a broadcast thereof.”

This is a singularly thought-provoking paragraph. Firstly, the Court acknowledges that the effects of excluding a competing competition arise in relation to the athletes, but “ultimately” in relation to spectators and viewers. By addressing the two core sides of the platform, athletes and viewers/spectators, the Court at least implicitly addresses this platform nature. However, precisely because of this implicit acknowledgement, the paragraph raises several questions. For one, I’m unsure whether the use of the word ultimately serves to indicate the chronology or is meant to indicate a hierarchy in the effects so that the effects on viewers and spectators are the litmus test. If it is the latter, this shows that the Court did not understand the platform nature of the sporting competition. The fact that the Court commences this paragraph with the mention of an as efficient competitor standard as part of the exclusionary effects in an art. 101-case points to the latter understanding where consumer welfare effects are what clinches a case. However, when reading this paragraph in connection with the level playing field rule, there is a clear coherence in the Court’s vision of what the purpose of EU competition law is: the creation of equality of opportunity so that final consumers have a choice.[69] This leads to the second question: how does this line of reasoning tie in with the more traditional reasoning in relation to exclusion? Traditionally, exclusion turns on (as efficient) competitors and their ability to compete. Para. 146 turns on athletes and viewers being deprived of an opportunity to enjoy the services of a competitor.

Perhaps para. 146 is a paradigm shift where the Court adopts the level playing field rule and no longer cares about the strict distinction between effects on competitors, effects on consumers and effects on competition. A focus on choice enabled by the level playing field rule and perhaps reviving a trend identified by Nihoul in 2012,[70] renders such distinctions superfluous. If ISU (and ESL where we also find this reasoning in para. 176) indeed, entail this paradigm shift, the Court would do good to explicitly say so, rather than to embark on tortuous reasoning to find a shaky foundation.

I have already alluded to what the implications of the “level playing field rule” would be for Google in Google Shopping and Google Android. The Google examples are more generally instructive, also when viewed in light the above-mentioned shaky doctrinal basis. As the competition-afficionado will know, both cases are at the forefront of the discussion about the effects-based approach to exclusionary abuse cases. This is a discussion that boils down to the competition policy question whether and to what extent anti-competitive effects in specific cases must be proven and whether these effects are limited to consumer welfare-aspects only. It is a debate that is epitomised by the concept of the as efficient competitor; a concept that also features in ISU. However, it is also a debate that turns on relinquishing the Michelin-special responsibility to legal history. It is the Michelin-special responsibility according to which a dominant undertaking “has a special responsibility not to allow its conduct to impair genuine undistorted competition on the common market”.[71] The Court could have saved itself one turn, more precisely transplanting the MOTOE/OTOC case law based on the combined reading of Articles 106 and 102 TFEU to the context of Article 102, by simply relying on the Michelin special responsibility. This arguably includes a duty for a dominant company to think twice about ways of wielding private power over market access. The level playing field rule has a far more sensible and intuitive basis in Michelin, but for some reason both the Commission and the Courts all sought to avoid this.

V.  Who is the ultimate arbiter?

A final interesting part of ISU concerns the role of the arbitration by Court of Arbitration for Sport (CAS) and the final review thereof by the Swiss Federal Supreme Court as prescribed by the ISU rules. According to the Commission this reinforced the infringement because it essentially isolated ISUs decisions from review on the basis of EU competition law. The General Court came to a different appraisal and found the arbitration compatible with Union law. The Court, however, dealing with the cross-appeal, sided with the Commission.[72] The Court reaches this outcome by first using the implementation theory to ascertain that the practices fall in the EU competition rules’ jurisdiction. The next step is the observation that arts 101 and 102 have direct effect can thus create rights for individuals that national courts are held to protect, and this doesn’t sit with the mandatory jurisdiction of a Swiss – i.e third state – court.[73] In the words of the Court: “In the absence of such judicial review, the use of an arbitration mechanism is such as to undermine the protection of rights that subjects of the law derive from the direct effect of EU law and the effective compliance with Articles 101 and 102 TFEU, which must be ensured – and would therefore be ensured in the absence of such a mechanism – by the national rules relating to remedies”. [74]

We thus have a rule imposing a requirement of EU-based effective judicial review on arbitration clauses, however, only insofar as matters within the scope of the competition rules are concerned.[75] Purely sporting-related (“non-economic”) disputes can still be subject to the CAS and the Swiss Federal Supreme Court. Interestingly, the Court does not mention the Achmea-based distinction between commercial and investment treaty arbitration whereas both the General Court and the Advocate General devoted attention to this aspect.[76] In Achmea, the Court differentiated between commercial arbitration for which the parties have freely chosen and investment treaty-based arbitration that is imposed by states on private parties, with only the latter possibly impacting the full effectiveness of EU law.[77] It is not difficult to see the analogy between the arbitration scheme in ISU and that laid down in investment treaties. Moreover, the judgment in ISU mentions several of the characteristics that set investment treaty arbitration apart from commercial arbitration and thus warrant imposing a requirement of effective judicial review.[78] If this judgment looks, barks and smells of it, why didn’t the Court mention Achmea? One explanation is that the Court simply wished to create a more generally applicable rule for all sorts of extra-judicial conflict resolution methods that remove disputes within the scope of EU law from the jurisdiction of the EU’s judicature. The take-home message is that the requirement of effective review features very prominently in the Court’s jurisprudence and impacts the autonomy of organisations like ISU just as much as it impacts member state autonomy. In that sense it can also be said that the return to Eco-Swiss’ focus on effective review may well have limited the importance of the Achmea carve-out for investment treaty-based arbitration.[79]

VI. Through the ice?

Following the December Revolution, the Court has spent quite a few cases dealing with the aftermath of the Keck “clarification”.[80] It is my prediction that the Court will also have to devote a few more words to dealing with this new level playing field-rule. The doctrinally problematic legal basis and the substance of the resulting rule are the main reasons for this. The legal basis, essentially the transplant of MOTOE/OTOC from art. 106/102 case law to art. 101-cases, is problematic for a number of reasons that will induce parties to seek to limit it whereas others may well want to stretch it. I can certainly imagine plenty of digital companies that are confronted with a digital giant with power over market access. The reason that I expect to trigger most follow-on cases is that the Court’s new rule relies on an ill-defined procedural requirement to avoid the by object-categorisation. Needless to say, avoiding a by object categorisation is all-important is this allows for far more extensive ways to find a specific way of using the power over market access compatible with EU competition law.[81] The ill-defined procedural requirement, however, makes predicting whether this actually applies in a specific case more difficult. This explains why ISU considers that it is now acting in compliance with EU law, basically continuing business as usual,[82] whereas EU Athletes, the organisation that supported the professional athletes who started the case against ISU, consider this case of “immense significance” and holding that it is about “athletes’ rights to engage in competition and leverage their commercial opportunities”.[83] I’m afraid that ISU will only result in a “right to a procedure”, albeit one that uses objective and transparent criteria to be applied in a non-discriminatory manner. Fair enough, but by no means a guarantee that there will be competing competitions and thus choice.[84]

Whilst the implications of this for sporting organisations are significant, there is no reason why the level playing field rule is confined to these organisations. Any entity with power over the market will now have to think of the implications of its activities for the level playing field and choice on that market, generating cases also beyond the sporting realm. Such follow-on cases may well be the price to pay for paradigm shifts. However uncertain the impact will be on actual choice, the level playing field rule is bound to guarantee more case law, precisely because of the many uncertainties that surround it. As I see it, the Court went on ice that was too thin.[85]