European Commission softens its presumption-based approach in its final guidelines on exclusionary abuses

The European Commission (EC) has adopted its final Guidelines on exclusionary conduct by dominant companies (final Guidelines). The final Guidelines mark the culmination of a two-year process that began with a 2024 draft that had proposed a significantly more presumption-based approach and had drawn considerable criticism from stakeholders. In their final form, the EC has recalibrated several of the most contentious elements of the draft, but ultimately steers a middle course by not reversing the fundamental departure from the 2008 Guidance Paper’s more economic approach.

Background 

In 2008, the EC published its Guidance Paper on Enforcement Priorities for Exclusionary Abuses (2008 Guidance Paper), which introduced a more economic approach based on effects rather than form. The 2008 Guidance Paper’s central focus was on "anti-competitive foreclosure" (as opposed to exclusion which is the result of competition on the merits) and on conduct that produces "consumer harm". While the 2008 Guidance Paper was only intended to set out enforcement priorities, it proved influential and was later endorsed by the EU Courts for the most part. 

In 2024, the EC issued draft Guidelines on exclusionary abuses in which it moved away from that more economic approach and the fundamental concept of "anti-competitive foreclosure". It essentially proposed a more legalistic approach with the introduction of categorisations and presumptions based on form and also departed from the so-called "as efficient competitor" principle that transcended the 2008 Guidance Paper. Under this proposed approach, for five categories of practices (exclusive supply or purchasing agreements, exclusivity rebates, predatory pricing, margin squeeze and certain forms of tying) exclusionary effects could be presumed and the burden of proof shifted to the dominant company to rebut the presumption by proving that the conduct is not capable of having exclusionary effects. These draft Guidelines were heavily criticised in the ensuing consultation and after a two-year period of reflection, the EC has published final Guidelines with a certain shift in approach. The final Guidelines will apply as of 30 days after their publication in the Official Journal on 9 September 2026. At that time, i.e. on 10 October 2026, the 2008 Guidance Paper will be withdrawn. 

Assessment of dominance

In contrast to the draft Guidelines, the EC has reintroduced a "soft safe harbor" in its final Guidelines, meaning companies with market shares of below 40% will usually not be seen as dominant. The final Guidelines state that, based on the EC's experience, dominance is generally unlikely if the company concerned holds a market share below 40%, although dominance can still be found below that threshold, for example where customers are generally dependent on the dominant company or where competing undertakings face serious capacity limitations.

The final Guidelines have an expanded chapter on collective dominance, in particular on tacit collusion based on collective dominance. Collective dominance refers to a situation in which two or more legally independent undertakings present themselves or act together from an economic point of view on a particular market and are able to behave to an appreciable extent independently of competitors, customers and ultimately of consumers. The final Guidelines note that Article 102 has been enforced only rarely against companies that hold a collective dominant position based on tacit coordination, but warns that "enforcement may still be warranted in certain situations" (for instance through the use of algorithms).

The final Guidelines provide additional details on ecosystems. With respect to assessing dominance, they discuss network effects occurring in the context of digital ecosystems of interlinked products, services or platforms, and note that, in addition to network effects, such ecosystems can reinforce a dominant position. With respect to non-pricing conduct, they note that the concept of a hypothetical equally efficient competitor may not be relevant in digital markets and ecosystems, where features such as innovation, access to data, multi-sidedness, user behaviour or network effects play a decisive role.

There is a new section on aftermarkets. To conclude that the primary market and aftermarket are interdependent - such that competition on the primary market disciplines power on the aftermarket and dominance there is excluded - four cumulative conditions must be met: (i) customers can make an informed choice between suppliers on the primary market, taking lifecycle pricing into account; (ii) customers are likely to actually make such an informed choice; (iii) if the dominant company pursued an exploitative policy in the aftermarket (e.g. a significant price increase or deterioration in supply conditions), a sufficient number of customers would adapt their purchasing behaviour on the primary market; and (iv) customers would adapt that behaviour within a reasonable time.

When does conduct constitute an abuse

The general framework of analysis

Generally speaking, the final Guidelines step away from the broad, multi-category presumption architecture set out in the draft Guidelines. While the final Guidelines have retained presumptions, these have been repackaged as soft presumptions. The EC has also introduced the sliding scale approach, meaning that the enforcer’s evidentiary burden depends on the type of conduct at issue. The more a given conduct is considered generally likely to distort effective competition, the less case-specific evidence is required to prove that this is the case, and the other way around. In some circumstances, the demonstration of certain factual elements may lead to the application of a presumption that the conduct distorts effective competition. Such presumptions, whose strength and scope may vary depending on the specific analytical framework and the specific evidence adduced in its support, shift the evidentiary burden from the EC onto the dominant company. The dominant company can rebut the application of the presumption by showing with sufficient evidence, during the administrative procedure, that the conduct does not distort effective competition in light of the factual circumstances of the case.

The final Guidelines retain the draft Guidelines’ general framework of analysis that is based on the fulfilment of two cumulative conditions. For conduct to be in breach of Article 102, it must:

• depart from competition on the merits; and
• must be capable of having exclusionary effects. 

Having exclusionary effects requires the effects in question to be more than hypothetical. However, it is not necessary to show that actual exclusionary effects have been produced or that the conduct would be profitable.

However, the final Guidelines elaborate on how the above two conditions relate to each other and in essence qualify their importance by clarifying that, in certain circumstances, it may be sufficient to satisfy only one of them in order to establish an abuse. Scenarios where it is unnecessary to analyse the two limbs separately are: 

• types of conduct for which the EU courts have developed a specific recognised analytical framework, which in reality is the case for most exclusionary practices; 
• where it is established that the conduct can exclude a hypothetical equally efficient competitor; and
• where the conduct is "by its very nature" harmful to competition because it has no economic rationale other than restricting competition.

There is no legal requirement to demonstrate direct harm to consumers to establish that conduct distorts effective competition. Absent from the draft Guidelines, the final Guidelines describe in a short paragraph "for explanatory purposes" the role of a theory of harm in the identification of factors that are relevant for the assessment of exclusionary abuses.

Where it is demonstrated that conduct distorts effective competition, it remains possible for the dominant undertaking to show that the conduct is either objectively necessary to achieve a legitimate aim and proportionate to that aim, or that the distortion of effective competition is counterbalanced or even outweighed by advantages in terms of efficiency that also benefit consumers. 

The fundamental distinction between pricing and non-pricing conduct

The type of conduct and its legal and economic context determine the type of effects analysis to be applied in each specific case. The final Guidelines, in line with the case law, make a fundamental distinction between pricing and non-pricing conduct.

For pricing conduct, the final Guidelines explain that the assessment of exclusionary effects is typically based on whether such conduct is capable of excluding a hypothetical "as efficient competitor" (AEC). A price cost-test is normally applied to determine whether this is the case. The price-cost test is typically carried out using the dominant undertaking's own price and cost data at the time of the alleged abuse. However, the final Guidelines recognise that dynamic considerations may also be relevant - for example, the need for equally efficient competitors in innovation-driven industries to recover product-related development costs. 

For non-pricing conduct, the assessment of non-pricing conduct may be based either on the conduct’s capability to exclude a hypothetical AEC or on other types of exclusionary effects in relation to which the concept of a hypothetical AEC is not relevant. In scenarios where the focus is on the exclusion of a hypothetical AEC, the final Guidelines expand the notion of "efficiency" underlying the AEC concept beyond price, to encompass other parameters of competition, such as quality, innovation and choice. A price-cost test will generally not be informative in these scenarios, since it struggles to capture non-price parameters. Instead, other types of evidence become relevant, including qualitative evidence and actual market developments.

The final Guidelines go further, identifying situations where the AEC concept is set aside altogether. In digital markets and ecosystems shaped by innovation, data access, multi-sidedness or network effects, the concept may not be relevant at all, since conduct can impede entry by potential competitors at an earlier stage. In addition, the document recognises that in some situations the emergence of as efficient competitor may be practically impossible - for example due to very large market shares, significant barriers to entry, or regulatory constraints. In such settings, even a less efficient competitor may exert a genuine competitive constraint on the dominant undertaking.

Specific legal tests for types of conduct

The final Guidelines have chapters explaining how the general framework is applicable to specific exclusionary practices including access restrictions, rebates not conditional on exclusivity and self-preferencing. The Guidelines also provide guidance on types of conduct for which the EU Courts have developed a specific analytical framework and legal tests to determine whether there is distortion of competition. This guidance concerns predatory pricing, margin squeeze, exclusive dealing, tying and bundling and refusal to supply. 

Noteworthy is that exclusive dealing (including exclusivity rebates) is presumed to distort competition but subject to rebuttal by the dominant company. The final Guidelines provide examples of the type of evidence that the dominant company can submit to seek to rebut this (rather soft) presumption.

The final Guidelines also contain a section of conduct that is by its very nature harmful to competition. This is conduct that is manifestly outside of competition on merits. This exceptional category encompasses the so-called "naked restrictions" of the draft Guidelines and also brings in the latest case law on "by object" abuses (mostly in the area of sport bodies) and the Lithuanian Railways types of conduct that make no economic sense and are explained only by the aim to harm competitors. A dominant company may submit evidence during proceedings to argue its conduct is not by its very nature harmful to competition, but it remains unsettled whether it can rebut such a finding by showing the conduct could not produce exclusionary effects in the specific circumstances. Pending clarification by the EU Courts, the EC considers such a challenge could only very exceptionally succeed, and it will adapt its approach if the courts later align the evidentiary standard with that for object restrictions under Article 101 TFEU. The dominant company can also try to justify such inherently harmful conduct on objective necessity or efficiency grounds, but the EC's own experience shows these attempts are very unlikely to succeed given the nature of the conduct involved. 

Efficiencies

The final Guidelines provide significantly more detail on arguments that companies can submit to claim efficiencies from any conduct that harms competition. Eligible efficiencies include all objective economic efficiencies, whether cost efficiencies or qualitative efficiencies, and can be short-term or longer-term. To establish the defence, the dominant undertaking must show four cumulative conditions: (i) the conduct allows efficiencies to be achieved, considering their actual existence and extent; (ii) those efficiencies counteract the conduct's negative effects on competition and consumers in the relevant market(s); (iii) the conduct is necessary (indispensable) to achieve those efficiencies; and (iv) the conduct does not eliminate effective competition by removing all or most existing sources of actual or potential competition. The burden of proof lies with the dominant company, and the assessment operates on a "sliding scale" - the greater the conduct's potential to harm competition, the less likely the four conditions will be met, meaning conduct found to be by its very nature harmful to competition is very unlikely to qualify.

Claimed efficiencies must be objective, concrete and verifiable, substantiated where possible by contemporaneous documents, financial data, expert studies or economic models, and quantified as precisely as reasonably possible, with a direct causal link between the conduct and the efficiencies (indirect or speculative links, such as higher profits merely enabling more R&D spending, are not sufficient). The dominant company must also show a "fair share" is passed on to consumers - sufficient to at least compensate those harmed by the conduct - and the EC treats pass-on as less likely the greater the company's market power or the more inelastic demand is, though efficiencies arising in a related market can count if the same group of consumers is substantially affected and benefited. On indispensability, the conduct must be necessary because the claimed efficiencies could not be achieved to a similar extent by less anticompetitive, realistically practical alternatives. Finally, the conduct must not eliminate effective competition by removing all or most remaining sources of actual or potential competition, since rivalry itself is viewed as an essential driver of longer-term efficiency and consumer welfare.

Sustainability is specifically recognised as a category of qualitative efficiency. The final Guidelines even allow for certain collective benefits, i.e., benefits accruing to a wider section of society to be taken into account¹. However, the requirement that the consumers negatively affected by the conduct in the relevant market substantially overlap with the beneficiaries of the benefit seriously limits the practical significance of this timid opening. 

Objective necessity defence

An objective necessity defence requires the dominant company to show its conduct is objectively necessary to achieve a legitimate aim. Necessity may rest on legitimate commercial considerations (e.g., protection against unfair competition, abnormal orders, or threats to service integrity/security, or below-cost sales to clear obsolete stock, though not mere price-matching with non-dominant rivals) or on technical justifications, such as maintaining or improving product performance. Where interoperability is refused on technical grounds, the mere absence or difficulty of a solution is not enough unless interoperability would compromise product integrity/security or be technically impossible. An objective necessity defence "may also relate to the pursuit of a genuine and objectively defined public interest objective," listing examples including "contribution to European Union's resilience."

Comments

It remains to be seen how the EC will apply the final Guidelines in practice. In a speech² on the day after publication of the final Guidelines, the EU Competition Commissioner promised antitrust enforcement to be "relevant and sharp", allow companies to flourish, and apply equally to companies across the world. Despite the softening of the presumption-based approach contained in the draft Guidelines, the final Guidelines still retain an expansive view of the special responsibility borne by dominant companies, a relatively low bar for finding exclusionary conduct, particularly for non-pricing practices, and an efficiency defence that remains difficult to satisfy.

_{1 For example, when the use of cleaner technologies or the production of more sustainable products bring about sustainability benefits to a wider section of society, independently of consumers’ individual appreciation of the technology or product.
2 Keynote, International Bar Association, Florence, Italy, Sept. 4, 2026}

_{White & Case means the international legal practice comprising White & Case LLP, a New York State registered limited liability partnership, White & Case LLP, a limited liability partnership incorporated under English law and all other affiliated partnerships, companies and entities.}

_{This article is prepared for the general information of interested persons. It is not, and does not attempt to be, comprehensive in nature. Due to the general nature of its content, it should not be regarded as legal advice.}

_{© 2026 White & Case LLP}