Presumptions retained but reframed: the Commission’s final guidelines on exclusionary abuses of dominance

Competition on the merits is defined as “normal competition on the basis of the performance of economic operators”, leading to consumer benefits in the form of lower prices, better quality and improved products. The Commission will consider the relevant conduct in its legal and economic context. Evidence of an intent to restrict competition may be relevant, but a subjective intention to compete on the merits will not preclude a finding of abuse. A finding that the conduct relies on means or resources that are inherent to a company’s dominant position will also be significant.

Little further general guidance is provided. Instead, examples are given from previous cases of: (i) conduct that is likely to deviate from competition on the merits (e.g. providing misleading information to public authorities or misusing regulatory or legal procedures to impede market entry); and (ii) factors that may be relevant depending on the circumstances of the case (e.g. preventing consumers from choosing products on their merits or biased or discriminatory treatment of competing products).

The Guidelines emphasise that the notion of exclusionary effects covers a broad range of hindrances to the competitive process and that the Commission does not need to show that the conduct has actually led to exclusionary effects. The fact that a number of competitors remain on the market therefore does not, by itself, mean that there are no exclusionary effects. Nevertheless, the effects must be more than purely hypothetical, and whilst there is no de minimis threshold, the Guidelines state that “conduct whose actual or potential effects are insignificant” will not be caught. They also clarify that in establishing whether conduct is capable of having exclusionary effects, the Commission does not necessarily have to establish what the counterfactual absent the conduct would have been. Nor is the Commission required to isolate the effects of the relevant conduct from effects that result from non-abusive behaviour of the dominant firm.

Having outlined those principles, the Guidelines explain how the Commission will examine the potential existence of exclusionary effects in respect of the following types of conduct:

• pricing conduct – where the Guidelines acknowledge that the EU Courts have generally required evidence of the capability to exclude a hypothetical as-efficient competition (AEC), typically through the application of a price-cost test²;
• non-pricing conduct, where the AEC concept may or may not be relevant; and
• multi-faceted conduct, which must be assessed on a case-by-case basis.

This requires the application of a price-cost test. Pricing above average total cost is not predatory, pricing below average variable cost will be deemed predatory, and pricing in between is predatory if part of a plan to eliminate competition.

This also requires the application of a price-cost test, with long-run average incremental cost the key benchmark: if this results in a negative downstream margin, the conduct is presumed to distort effective competition.

Obligations or incentives (e.g. rebates) causing a customer or supplier to obtain all or most of its requirements from the dominant firm are presumed to distort effective competition – a presumption which a dominant firm may seek to rebut, e.g. via evidence relating to the coverage and duration of the exclusivity arrangements, or their inability to affect customers’ purchasing decisions. A dominant firm can also submit a price-cost analysis, which the Commission must examine. Rebates not conditional on exclusivity fall outside the specific analytical framework, but in general the Commission will apply a price-cost test to examine whether they are capable of excluding an AEC (the Guidelines note, however, that in certain settings even a less efficient competitor can exercise a genuine constraint, such as in situations where the emergence of an AEC is practically impossible).

For tying to distort effective competition, four conditions must be met: the products must be distinct; the firm must be dominant in the tying market; customers must not be able to obtain the tying product without the tied product; and the tying must be capable of having exclusionary effects.

This is defined as a form of access restriction involving an asset developed or established solely for the dominant firm’s own use. To distort effective competition, the asset must constitute an indispensable input for firms operating in a related market, and the dominant firm’s refusal to provide access must be capable of eliminating all effective competition on the part of the requested undertaking. If the refusal concerns IP, it must in addition limit technical development. This will be the case if it prevents the production of new goods or services (not replicas of what the dominant firm is already offering) for which there is potential consumer demand. However, these strict conditions do not apply to assets that were not developed solely for the dominant firm’s own use or are subject to legal, regulatory or contractual limitations on the firm’s decision-making autonomy. For such access restrictions, only the general two-limb test applies. The same applies to self-preferencing, where a dominant platform’s own products or services benefit from preferential treatment over competing products in a related/adjacent market (in such cases, the importance of the platform as a source of business for competitors, and whether there is an expectation of neutrality on the part of the platform operator, will be relevant factors).

Whereas the draft Guidelines focused on identifying conduct that is “liable to be abusive”, the final version refers instead to conduct that “distorts effective competition”. While no explanation is offered for this change, both concepts capture the same thing: behaviour that will breach Article 102 TFEU unless objectively justified.

The term “naked restrictions” (taken from Intel) has also been replaced by conduct that is “by its very nature harmful to competition”. This may be an attempt to broaden the concept slightly, extending beyond conduct that can only be explained as designed to restrict competition - similar to restrictions “by object” under Article 101 TFEU, which need not be motivated by an anticompetitive intent. This impression is reinforced by the Commission’s suggestion in the Guidelines that such abuses should be subject to the same evidentiary requirements as restrictions by object, with the result that an effects analysis becomes unnecessary - even though this has not yet received judicial confirmation from the CJEU.

Additionally, whereas the draft spoke of specific “legal tests” for certain types of conduct, the final Guidelines instead refer to “analytical frameworks”. This may reflect the fact that the EU Courts have not expressly recognised all of those frameworks as clear legal tests. This is also consistent with the reduced emphasis on the probative value of the presumptions that operate under these frameworks.

As noted above, the Commission maintains that where a specific analytical framework is satisfied, the evidentiary burden shifts to the dominant firm to justify the conduct or show that it does not distort effective competition. However, those presumptions are given much less prominence than in the draft Guidelines. The Commission also acknowledges that the “strength and scope” of these presumptions may vary. And it no longer suggests that they continue to have probative value even once rebutted.

Unlike in the draft, what follows once a presumption has arisen is discussed in detail only in relation to exclusivity rebates. This may be because, in respect of such conduct, the Commission has had to accommodate the 2024 Intel judgment. But whether it has properly done so is debatable. In particular, the CJEU confirmed that a full effects analysis is required whenever a dominant firm submits evidence that its conduct was not exclusionary. But the Guidelines assert that the Commission may reject such evidence as being insufficient to rebut the presumption - potentially allowing it to establish an abuse without positively proving its case on effects.

Overall, the changes made since the draft suggest that the Commission has recognised that presumptions are not a panacea and cannot be applied as mechanistically as may initially have been envisaged in order to short-cut the analysis and establish an abuse.

Unlike the draft, the final version of the Guidelines acknowledges in numerous places that certain types of conduct have, in general, been assessed by means of a price-cost test. Nevertheless, the Commission maintains that an AEC or price-cost test is not necessary or determinative in all cases - even those involving pricing conduct or exclusivity rebates. In relation to the latter in particular, the Commission indicates that a price-cost test will not be relevant in a range of scenarios - a position that is difficult to reconcile with Intel, in which the CJEU held that, in contested cases, the Commission must assess whether there is a strategy aimed at excluding AECs, which “as a general rule” is to be done using a price-cost test.

The final section of the Guidelines, addressing objective necessity and efficiencies, has been greatly expanded since the draft. The concepts, including how a dominant firm might go about establishing such defences, are more fully explained, with a notable shift of tone - potentially signalling a greater willingness on the part of the Commission to engage with such arguments. This is particularly apparent in the case of sustainability benefits, with the recognition that out-of-market efficiencies can be factored into the analysis where they accrue to customers impacted by the conduct. 

That said, efficiency benefits have, to date, never been formally accepted by the Commission (or EU Courts) in an Article 102 case. The Guidelines also caution that efficiencies are less likely to be passed on where a firm is in a particularly strong dominant position, and that claims that restrictive conduct is necessary to preserve investment incentives and recover sunk costs will face close scrutiny where a dominant firm has significant financial resources.

1 This document has now been formally withdrawn and will cease to apply as of 30 days after the publication of the Guidelines in the Official Journal of the EU.

2 This analyses whether a non-dominant competitor with the same cost structure as the dominant firm could, despite relevant conduct, compete for the relevant business without incurring a loss.