First of its kind: The EU Guidelines on exclusionary abuses of dominance

On 3 September 2026, the European Commission adopted Guidelines on the application of Article 102 TFEU to abusive exclusionary conduct by dominant undertakings (Guidelines). The Guidelines replace the 2008 Guidance Paper on enforcement priorities related to abusive exclusionary conduct by dominant undertakings (2008 Guidance Paper) and represent the most significant overhaul of the European Commission's approach to abuse of dominance in nearly two decades.

What you need to know

• The Guidelines are the European Commission's first comprehensive guidance on Article 102 TFEU. They set out binding principles and an operational analytical framework for assessing exclusionary abuses, going significantly beyond the 2008 Guidance Paper, which was limited to enforcement priorities. They withdraw and replace the 2008 Guidance Paper in its entirety.
• Six areas mark major change: a soft safe harbour at 40% market share (below which dominance is generally unlikely), with market shares of 50% or above constituting evidence of dominance save in exceptional circumstances, but no absolute threshold shields against a dominance finding; a two-limb test for abuse with a sliding evidential burden and presumptions; a dedicated section on causation; a reduced role for the as-efficient competitor (AEC) test in non-pricing cases (though it remains an available defence); a broader toolkit for access restrictions and self-preferencing; and an expanded framework for objective justifications including sustainability and EU resilience.
• The European Commission retains significant flexibility under the new framework, with context playing a central role at every stage of the analysis. Dominant companies should proactively develop robust evidence to demonstrate the absence of exclusionary effects, including qualitative evidence, quantitative analysis (such as the AEC test where appropriate), and counterfactual analysis. Presenting such evidence would cause the competition authority to consider and rebut the arguments submitted.

From enforcement priorities to full guidelines

On 3 December 2008, the European Commission published a Guidance Paper setting out its enforcement priorities for exclusionary conduct under what was then Article 82 of the Treaty establishing the European Community, now Article 102 of the Treaty on the Functioning of the European Union (TFEU). The European Commission (Commission) did not intend the 2008 Guidance Paper to be a statement of law. The 2008 Guidance Paper departed from the formalistic approach prevailing for decades in the case law and decisional practice and promoted an effects-based approach to Article 102 TFEU (including through the introduction of economic analysis for the assessment of certain abuses).

While the EU Courts have endorsed the main tenets of an effects-based approach, the framework in the Guidance Paper was overtaken by nearly two decades of case law developments. Moreover, the Commission significantly refined its case practice, notably to take account of rapidly evolving markets.

The Guidelines set out for the first time principles and operational guidance for assessing whether conduct by dominant undertakings constitutes an exclusionary abuse under Article 102 TFEU, drawing on both the case law of the EU courts and the Commission's own decisional practice. The Guidelines purport to promote a workable effect-based approach balancing rigorous effects analysis with legal certainty and administrability. By doing so, the Guidelines seek: to modernise EU competition policy, taking into account market realities and ensuring that effective enforcement is timely; to increase legal certainty and help undertakings self-assess whether their conduct could amount to an exclusionary abuse; and to enhance consistency in the application of Article 102 TFEU, including by national courts and national competition authorities.

The Guidelines cover a much broader range of topics than the 2008 Guidance Paper, including single and collective dominance (including collective dominance through tacit coordination facilitated by algorithms), multiple forms of exclusionary conduct (predatory pricing, margin squeeze, rebates, exclusive dealing, tying and bundling, refusal to supply and, for the first time, access restrictions and self-preferencing), conduct that is by its very nature harmful to competition, and a significantly expanded section on objective justifications.

Assessment of dominance: reinstatement of the soft safe harbour for market shares below 40%

The Guidelines confirm that a market share of 50% or more is, save in exceptional circumstances, evidence of dominance. Dominance may also be found below 50%, and the Commission considers that dominance is generally unlikely below 40%. However, the Guidelines make explicit that even below 40%, dominance can still be found where, for example, customers are generally dependent on the undertaking concerned or where competing undertakings face serious capacity limitations.

The 40% ‘soft safe harbour’ was nearly lost during the consultation process. The draft Guidelines published for consultation in August 2024 had removed the 40% reference entirely and indicated only that market shares below 10% would generally exclude dominance. Stakeholders criticised this significantly, arguing that the absence of any meaningful safe harbour would chill pro-competitive behaviour and reduce legal certainty. In response, the Guidelines restate that dominance is generally unlikely below 40%. In practice, the 40% ‘soft safe harbour’ remains a useful indicator.

However, companies cannot rely on market share alone to shield themselves from a finding of dominance. The assessment must always take the specific circumstances into account, including factors such as barriers to entry or expansion (including data-driven advantages and network effects), countervailing buyer power, and the dynamics of the relevant market. In particular, the Guidelines place increased emphasis on barriers to entry in digital markets, including data accumulation, data-driven network effects, ecosystem dynamics, and the role of artificial intelligence.

How the Commission assesses abuse

The two limb test

To find that a dominant undertaking has committed an exclusionary abuse, the Commission must show that the conduct in question distorts effective competition. The Guidelines codify a two-limb test from case law: the Commission must show (1) that the conduct departs from competition on the merits, and (2) that the conduct is capable of having exclusionary effects. In line with settled case law, direct harm to consumers need not be shown to establish that conduct distorts effective competition.

• The first limb asks whether the dominant undertaking has gone beyond the scope of normal competitive behaviour. Competition on the merits covers, in principle, a competitive situation in which consumers benefit from lower prices, better quality and a wider choice of new or improved products, both in the short and long term. A dominant company departs from that standard where, for example, it provides misleading information or misuses regulatory procedures. The use of resources or means inherent to its dominant position to foreclose competitors will generally also be an indicator of departure from competition on the merits.
• The second limb asks whether the conduct can harm competition. The Commission must demonstrate that the conduct is capable of producing exclusionary effects, meaning it can hinder or eliminate competitors' effective market access, raise barriers to entry, or reduce competitors' ability or incentive to compete. This assessment must be grounded in specific, concrete evidence and take into account all relevant facts and circumstances – including factual elements that may not themselves constitute abusive conduct but are nonetheless relevant as context. The effects must be more than hypothetical, but actual harm need not have occurred.

Critically, the Commission need not always prove both limbs separately. This applies where specific analytical frameworks recognised by the EU courts apply; the conduct is capable of excluding an equally efficient competitor; or the conduct is by its very nature harmful to competition.

A dedicated section on causation

The Guidelines introduce a dedicated section on causation. The Commission must establish a causal link between the dominant undertaking’s conduct and the exclusionary effects identified, though it is sufficient to show that the conduct increases the likelihood of the exclusionary effects materialising – effects can be actual or potential. The conduct need not be the sole cause of exclusionary effects.

Conceptually, the analysis entails comparing the situation with and without the conduct; in practice, this often involves comparing the market situation before and after implementation. In certain cases, a counterfactual scenario accounting for developments that would have occurred independently may be appropriate, though developing reliable counterfactual assumptions can be challenging – establishing one credible alternative scenario may suffice.

However, the Commission may rely on a range of evidence to prove causation without being required to use any single methodology systematically, including counterfactual analysis. This is particularly relevant where the conduct has made it difficult to ascertain objective causes of market developments. The assessment is based on facts and circumstances existing at the time of the conduct, not the actual reaction of third parties or the ability of competitors to mitigate effects. Critically, where users’ behaviour has been distorted by the conduct, it may be impossible to attribute users’ choices to the superior performance of the dominant undertaking’s product – a principle with significant implications for digital markets where conduct may shape user behaviour over extended periods.

A sliding scale of evidentiary requirements

The Guidelines rely on different evidential standards depending on the type of conduct, operating on a sliding scale. The general principle is that the more likely the conduct is to distort effective competition, the less case-specific evidence the Commission needs to prove its case.

• Conduct that is by its very nature harmful to competition (so-called naked restrictions – a term which refers to conduct with no economic interest for the dominant undertaking, other than that of restricting competition, e.g. paying customers not to sell a competitor's product or dismantling infrastructure a competitor relies on) triggers the strongest presumption: the Commission deems such conduct to distort effective competition as such, and the dominant company faces a very difficult – if not impossible – burden to challenge that finding.
• For predation, satisfying the relevant price-cost test is in principle sufficient for a finding that the conduct distorts effective competition.
• For exclusive dealing and exclusivity rebates, there is a rebuttable presumption that the conduct is abusive, given exclusive dealing’s ‘high potential’ to generate exclusionary effects. However, the dominant undertaking can submit evidence showing the conduct was not capable of restricting competition – for example, by demonstrating that customers were aware no competitor could have supplied the volumes covered by the exclusivity arrangement.
• For margin squeeze, tying and refusal to deal, established legal tests apply (the price-cost test, the Microsoft test and the Bronner test respectively).
• Other conduct, such as conditional rebates, financial bundles, access restrictions and self-preferencing, is subject to the general two-step analysis.

The final Guidelines represent an improvement over the draft, which placed different types of conduct in rigid boxes and sought to apply numerous presumptions of exclusionary effects – including for classic tying – that were not supported by the case law.

However, some aspects of the final Guidelines may still be considered at odds with the case law. In particular, on exclusivity rebates and exclusive dealing, the principles on the application and rebuttal of the presumptions are arguably inconsistent with the Intel I judgment (C-413/14) which concerned exclusive rebates. According to Intel, if during the administrative procedure the dominant company submits evidence that its conduct was not capable of restricting competition, the Commission is required to examine that evidence and prove foreclosure by reference to at least the extent of the dominant position, the market coverage, the rebate conditions, duration and potential existence of an exclusionary strategy. The Commission has defended the presumption approach based on its reading of the case law, including the Intel II judgment, and notes that economic theory also supports a rebuttable presumption in exclusive dealing cases.

When economic analysis matters: the as-efficient competitor test

The AEC test asks whether a hypothetical competitor with the same cost structure as the dominant undertaking could profitably match the dominant undertaking's pricing conduct. If not, the conduct is capable of excluding equally efficient rivals and may amount to an abuse. The 2008 Guidance Paper applied the AEC test primarily to pricing conduct, including predatory pricing, conditional rebates, and multi-product rebates, but did not apply it uniformly across all types of exclusionary conduct.

The Guidelines take a more nuanced position. They recognise that the AEC test remains a central analytical tool for pricing conduct, but significantly limit its role for non-pricing abuses.

The Guidelines distinguish between three categories of conduct:

• Pricing conduct (predatory pricing, margin squeeze, conditional rebates, multi-product rebates): the Commission typically assesses whether the conduct can exclude a hypothetical equally efficient competitor, using a price-cost test. This remains the primary analytical tool for these types of abuse.
• Non-pricing conduct (tying, refusal to supply, self-preferencing, access restrictions): the AEC test is generally not required and in many cases not relevant. The EU courts have confirmed that not every finding of an exclusionary abuse requires proof that the conduct can exclude an equally efficient competitor. The Commission considers that applying a price-cost test to non-pricing behaviour could only arise in exceptional circumstances, given the general difficulty in reliably quantifying non-pricing elements. Instead, the Commission will rely on qualitative evidence and actual market developments to assess exclusionary effects.
• Multi-faceted conduct (combinations of pricing and non-pricing behaviour): the Commission determines the type of analysis on a case-by-case basis. Where the conduct predominantly consists of non-pricing behaviour, it may not need to apply a price-cost test to the pricing component at all.

One important safeguard remains for dominant companies: if a dominant undertaking submits its own AEC analysis, the Commission must examine its probative value. However, the Commission is not required to carry out an AEC analysis of its own for non-pricing conduct or for exclusivity rebates where a presumption of abuse can apply. The practical consequence is that, for the growing category of non-pricing abuses, dominant companies cannot rely on passing a price-cost test as a defence – the Commission can establish an abuse on qualitative evidence alone. That said, a workable effects-based approach cannot mean that if a dominant firm does not provide cost data, the Commission has to stop its assessment – alternative evidence, including rebate conditions, coverage, duration, internal documents, and impact on competitors, can be used.

The Guidelines provide that the concept of a hypothetical AEC will generally not be relevant in digital markets and ecosystems that are characterised by significant barriers to entry and network effects. Interestingly, the Guidelines address this only in their section on non-pricing conduct, which suggests that the AEC test remains an appropriate defence for digital companies (and other sectors characterised by network effects) at least for pricing conduct.

First guidance on access restrictions and self-preferencing

Under the 2008 Guidance Paper, the Commission indicated it would consider refusal to supply practices as an enforcement priority if all the following circumstances were present: the input had to be indispensable (objectively necessary to compete effectively downstream), the refusal had to be likely to eliminate effective competition, and it had to be likely to lead to consumer harm.
The Guidelines significantly extend the guidance on input-related abuses, distinguishing three categories: access restrictions, refusal to supply, and self-preferencing.

• Access restrictions now cover a much broader range of scenarios than outright refusal to supply. The Guidelines define them as situations where a dominant undertaking restricts access to an input by denying access under commercially viable conditions, or otherwise hindering or delaying it. Crucially, the Commission can find abuse even if the input at issue is not indispensable – a significantly lower threshold than the Bronner standard for refusal to supply. However, the importance of the input remains a key factor: the more important the input for the party seeking access, the greater the likelihood that restrictions will lead to exclusionary effects. The Guidelines identify specific scenarios that may constitute abusive access restrictions, including: disrupting supply to existing customers competing with the dominant undertaking downstream; refusing access to an input developed to enable third-party use; failing to comply with regulatory or contractual obligations to supply; and applying unfair or unreasonable access conditions, such as failing to set transparent terms or delaying access negotiations. Given the breadth of conduct potentially captured, it remains to be seen how this framework will be applied in practice.
• Refusal to supply remains subject to the stricter Bronner analytical framework, but the Guidelines confine this to a narrower scenario: where a dominant undertaking owns an input it developed solely for its own use and refuses to grant access. In those cases, the Commission must still show that the input is indispensable and that the refusal is capable of eliminating all effective competition.
• Self-preferencing appears as a distinct category of abuse for the first time in formal guidelines, reflecting recent case law. The Commission recognises that no general rule makes self-preferencing problematic, even for dominant undertakings. However, the Guidelines identify specific situations where it may distort effective competition – particularly where a dominant undertaking controls conditions of access in one market (the leveraging market) and uses that position to give preferential treatment to its own products in another market (the leveraged market). The Bronner conditions do not apply: the Commission does not need to show that the leveraging market is indispensable, only that it constitutes an important source of business for competitors that they cannot effectively replace through other means. It remains to be seen how this framework will be applied outside digital markets.

Expanded guidance on objective justifications

Even where the Commission establishes that conduct distorts effective competition, the dominant undertaking can seek to justify it. The 2008 Guidance Paper set out the basic framework for objective necessity and efficiency defences, including the conditions for efficiencies. The Guidelines significantly expand on this foundation, devoting an entire section to objective justifications with detailed guidance on each element and introducing new categories of legitimate aims.

The Guidelines recognise two types of defence:

• The objective necessity defence requires the dominant undertaking to demonstrate that its conduct is objectively necessary to achieve a legitimate aim and that the means used are proportionate. The Guidelines give a broader range of examples of legitimate aims than the 2008 Guidance Paper, including protection against unfair competition, technical justifications, compliance with regulatory obligations, product and network safety, and – notably – the European Union's resilience, covering conduct necessary to reduce supply chain dependencies or ensure sufficient production capacity for the EU's security and defence industry. However, the Commission will not accept this defence where the dominant undertaking could have achieved the same aim through less restrictive means.
• The efficiency defence requires the dominant undertaking to satisfy four cumulative conditions: (i) the conduct achieves verifiable efficiencies, (ii) those efficiencies counteract the negative effects on competition and consumers, (iii) the conduct is necessary (indispensable) to achieve those efficiencies, and (iv) the conduct does not eliminate effective competition. The Guidelines apply a sliding scale: the greater the potential of the conduct to harm competition, the harder it becomes to satisfy these conditions. For conduct that the Commission considers by its very nature harmful to competition, an efficiency defence is very unlikely to succeed.

The Guidelines break new ground in two areas. First, they recognise sustainability benefits as a form of qualitative efficiency – for example, where the dominant undertaking's conduct enables the use of less raw materials or less polluting production technologies, increases the recyclability of products, enables reliance on more resilient infrastructure, reduces the risk of supply chain disruptions, enables the production of new more sustainable products, or reduces the time it takes to bring products to the market. Second, in line with the approach under the guidelines on horizontal agreements, they introduce the concept of out-of-market efficiencies, which the Commission can take into account where the consumers harmed by the conduct and those who benefit from the efficiencies substantially overlap.

The burden of proof for both defences lies squarely with the dominant undertaking. Vague, general, or theoretical claims will not suffice – the dominant undertaking must provide convincing arguments and verifiable evidence. In the Commission's own experience, efficiency defences have rarely succeeded in practice, particularly for conduct it considers by its very nature harmful to competition.

Impact for businesses: what this means in practice

• Reassess dominance exposure. Businesses with significant market positions should not assume that a market share below 40% shields them from a finding of dominance. The Guidelines make clear that barriers to entry, network effects, data-driven advantages, and customer dependency can all support a dominance finding at lower market shares. Digital market and ecosystem operators face particular scrutiny.
• Prepare for a context-driven assessment. The Commission retains significant flexibility under these Guidelines, and context plays a central role at every stage – from dominance through to the assessment of abuse and objective justifications. Businesses should be able to demonstrate the competitive rationale for their conduct based on facts at the time, not with hindsight.
• Understand the limits of economic evidence. For pricing conduct, the AEC test remains the primary analytical tool, and passing it still provides meaningful protection. For non-pricing conduct, the Commission can establish an abuse on the basis of qualitative evidence alone. While this may suggest that economic evidence plays a reduced role, the AEC test remains an available defence even for non-pricing conduct. Dominant companies should proactively develop robust evidence to demonstrate the absence of exclusionary effects, regardless of the type of conduct at issue. This includes qualitative evidence (such as short duration of the conduct, or genuine possibilities for early termination), quantitative analysis (such as low market coverage or an AEC test where appropriate), and counterfactual analysis showing that competitors would not have been better off absent the conduct. Presenting such evidence would cause the competition authority to consider and rebut the arguments and economic analysis submitted, or to provide alternative evidence of foreclosure effects and causation – even where the authority was not initially required to conduct such analysis or could otherwise rely on presumptions of foreclosure. This applies to all conduct except conduct specifically designed to exclude competitors (or 'naked restrictions').
• Review access and self-preferencing practices. The Guidelines recognise access restrictions other than refusal to deal and self-preferencing as distinct categories of abuse for the first time and codify the existing case law and decisional practice in this respect. Companies that control inputs, platforms, or ecosystems on which competitors depend should review their terms of access, interoperability practices, and any preferential treatment of their own products.
• Build a justification file. The burden of proof for objective justifications lies with the dominant undertaking, and the Commission has made clear that vague or theoretical claims will not suffice. Businesses should document efficiency gains, sustainability benefits, and any objective necessity at the time they adopt conduct that could face scrutiny. The Guidelines offer new arguments – including sustainability and EU resilience – but the evidential bar remains high.
• Monitor the broader trend. The Guidelines reflect a wider shift in EU competition law towards a more context-sensitive, effects-based approach, mirroring recent developments under Article 101 TFEU. Businesses should expect national competition authorities and courts across the EU to follow this direction.

Authors: Sergej Bräuer, Partner; Donald Slater, Partner; Jessica Bracker, Senior Associate; Dimitra Karakioulaki, Associate and Sarah Schaible, Transaction Lawyer.