Reflections on the draft Article 102 guidelines: Bringing clarity or creating room to manoeuvre?

Overview

On 1 August 2024, the European Commission published draft guidelines on exclusionary abuses of dominance. Their stated purpose is to increase legal certainty for firms, national courts and national competition authorities by setting out how the Commission interprets Article 102 TFEU and the EU courts’ case law on exclusionary abuses. 

Since then – for reasons both related and unrelated – the Commission’s approach to Article 102 has rarely been out of the limelight. In September, the European General Court (GC) overturned the Commission’s decision in Google AdSense, while Mario Draghi caused a splash in Brussels with the publication of his report on “The future of European competitiveness”, which Commission President von der Leyen has specifically directed the new Competition Commissioner, Teresa Ribera, to draw on (as reported in our recent newsletter). Draghi is highly critical of the Commission’s draft guidelines, which he views as leaving “excessive discretion” to the Commission and lacking clarity.

In some ways, the draft guidelines represent a useful summation of the case law from the EU courts.  However, they can also be viewed as an attempted land grab by the Commission, pushing against and beyond the boundaries of settled case law in a number of important respects, and carving out more room to manoeuvre when it comes to enforcement. 

Most notably, the draft guidelines downplay the fundamental principle that Article 102 is not intended to protect competitors that are less efficient than the dominant firm. They also seek to tip the balance of the burden of proof away from the Commission, towards the allegedly dominant firm, including by identifying a wide range of types of conduct where the capability to produce exclusionary effects may be presumed. The draft guidelines also seek to address novel categories of abuse – such as self-preferencing – that are primarily associated with the rise of the digital economy. But in doing so they fail to provide the necessary clarity for firms to be able to self-assess compliance. 

The draft guidelines are the subject of a public consultation, running until 30 October 2024.

Background

In its last major review of Article 102 policy – the publication of guidance on its enforcement priorities in 2008 – the Commission set out its stall for an “effects-based” approach. This was endorsed by the European Court of Justice (CJ) in 2017 in Intel (see our previous client briefing) and has broadly been followed since (for example, in Slovak Telekom and Unilever Italia). The effects-based approach was premised on the idea that “in principle” only competitors that were “as efficient” as the dominant undertaking were worthy of protection under Article 102, offering dominant undertakings scope to justify their conduct on the basis of efficiencies and raising the threshold for the Commission to establish an abuse.

In the years since, and in the face of a number of enforcement challenges (including most recently in the Google AdSense appeal), the Commission’s attitude has shifted. The direction of travel was evident from tweaks made last year to the 2008 guidance, which paved the way for greater consideration of competitors that are less efficient than the dominant firm (see our previous newsletter). The accompanying policy brief referred to the “growing importance of digital markets and services” and their association with “'winner-takes-all’ dynamics”. In such markets, the Commission claimed, challengers may not be expected to achieve similar cost structures to a dominant incumbent, but could still play a significant role. In these cases, mandating evidence of foreclosure of as-efficient competitors would have the potential to lead to under-enforcement. The Commission therefore sought to avoid an “unduly strict and dogmatic application” of the as-efficient competitor (AEC) standard.

The policy brief also highlighted that developments in the case law went “significantly beyond the topics” covered in the amended 2008 guidance, setting the stage for the introduction of more detailed guidelines.

Two-stage test

The draft guidelines published in August propose a two-stage test for determining whether conduct by a dominant undertaking is liable to constitute an exclusionary abuse: first, does the conduct depart from competition on the merits; second, is it capable of having exclusionary effects? For these purposes, the draft guidelines distinguish three categories of conduct: naked restrictions, conduct with specific legal tests, and other types of conduct.

Category one: Naked restrictions

According to the draft guidelines, naked restrictions – conduct that has no economic interest for an undertaking beyond restricting competition – are deemed to fall outside the scope of competition on the merits and can be presumed to be capable of exclusionary effects.^[1] Moreover, the dominant undertaking will be able to rebut this presumption “[o]nly in very exceptional cases”. This represents a return to the more formalistic assessment of abuse seen prior to the 2008 guidance.

The approach of the draft guidelines to this category of conduct arguably represents a fair reflection of recent case law. For example, in 2022 the CJ ruled in SEN that “any practice the implementation of which holds no economic interest for a dominant undertaking, except that of eliminating competitors so as to enable it subsequently to raise its prices by taking advantage of its monopolistic position, must be regarded as a means other than those which come within the scope of competition on the merits.” 

Category Two: Conduct with a specific legal test

Self-preferencing

The draft guidelines acknowledge that self-preferencing “can be widespread in certain sectors of the economy and the question whether a given self-preferencing conduct contravenes Article 102 TFEU depends on an analysis of all relevant circumstances”. This begs the question what those circumstances might be. The draft guidelines are vague in this regard. Citing the GC’s judgment in Google Shopping, they propose three non-cumulative “elements”, or circumstances, in addition to the six general factors referred to above.   

The first circumstance is that the preferential treatment takes place on a leveraging market that constitutes “an important source of business” for competitors in the leveraged market that they cannot effectively replace through other means. This is a lower standard than that set by the CJ’s recent judgment in Google Shopping, which made clear that it is only the subset of traffic (or business) from the leveraging market that is “diverted” by the conduct at issue that is relevant to the assessment of replaceability – not all traffic. Whilst, according to Google Shopping, the conduct causing the diversion must also be discriminatory – a mere difference in treatment will not depart from competition on the merits – the draft guidelines do not reflect this: discrimination is conspicuously absent (see below). 

The second circumstance in the draft guidelines is that the preferential treatment is likely to influence the behaviour of users, irrespective of the intrinsic qualities of the leveraged product. The draft guidelines stretch the case law by presenting this as an optional, not cumulative, condition. The final circumstance in the draft guidelines – that the self-preferencing is likely to be contrary to the firm’s “business rationale” in the leveraging market – draws on an aspect of the GC judgment in Google Shopping that neither appeared in the Commission’s decision nor was considered by the CJ as necessary to find an abuse.

Another issue is that the draft guidelines do not present a clear explanation for why its Google Shopping-based tests for self-preferencing conduct should not be used to assess self-preferencing conduct involving access restrictions and margin squeezes (for which different tests are set). This could cause confusion – for instance, is the unfair positioning and display of products an example of “self-preferencing” of the Shopping variety or is it an “access restriction”?

Access restrictions

The draft guidelines suggest that access restrictions can be abusive even if the input is not indispensable, as the need to protect the undertaking’s freedom of contract and incentives to invest does not apply to the same extent as in a refusal to supply setting. Examples are provided of access restrictions which may be considered to be contrary to Article 102, including where the conduct may disrupt the supply of existing customers, where the dominant undertaking fails to comply with a regulatory obligation to give access, where the dominant undertaking degrades or delays the existing supply of an input by imposing unfair access conditions, and (following the GC in Google Shopping) where the dominant undertaking develops an input for the declared purpose of sharing it widely with third parties but later does not provide or restricts access.

The approach to access restrictions set out in the draft guidelines is expansive, given the potentially significant implications for freedom of contract, right to property and incentives to invest.

Concluding remarks

The draft guidelines emphasise the importance of applying the rules on abuse of dominance in a “predictable and transparent manner”. But if this is their objective, the draft guidelines currently fall short. The somewhat selective approach to summarising the case law, and the attempt to airbrush the AEC principle out of enforcement history, is likely to increase rather than reduce uncertainty. More generally, building in room to manoeuvre when it comes to running cases – by referring to “non-exhaustive examples” and placing weight on broad concepts such as “self-preferencing” – while understandable, is clearly unhelpful for businesses anxious to stay the right side of the rules.

The draft guidelines also contain some notable quirks and omissions. Given the length of time that has elapsed since the 2008 guidance, it is regrettable that the draft guidelines provide a patchwork of specific legal tests and factors, rather than a more coherent summation of the Commission’s approach to abuse of dominance cases. There is almost no reference to the principle of discrimination, despite the fact that it is embedded in Article 102(2)(c), and that certain of the proposed “new” types of conduct involving a difference in treatment would be natural candidates for some form of discrimination requirement. Exploitative abuses are not covered at all, despite recent expressions of interest by the Commission in bringing enforcement cases in this area.

It will be interesting to follow the development of the draft guidelines under the new leadership of Teresa Ribera. Most known for her work supporting the green energy transition, Ribera’s stance on antitrust policy is less clear, but President von der Leyen has directed Ms. Ribera and other nominated Commissioners to draw on Mario Draghi’s report, which emphasises that the “speed” and “predictability of decisions” needs to be improved, noting that “[d]ecade-long cases like the Intel case” are “not isolated episodes”. As already noted, the report is highly critical of the draft guidelines. For Draghi, the Commission’s approach to exclusionary abuses under Article 102 is an area that needs to be “urgently streamlined”.