On 3 September 2026, the European Commission adopted its long-awaited Guidelines on the application of Article 102 TFEU to exclusionary abuses by dominant undertakings (the Guidelines). These guidelines are significantly more comprehensive than their predecessor and mark the end of a process that began in March 2023 (see also our earlier post here), followed by draft guidance, a public consultation in 2024 and a stakeholder workshop in February 2025. So, this is a good moment to cast a glance into the crystal ball and ask where the Commission’s abuse of dominance enforcement is heading. To that end, I have attempted to distil all 77 pages of the Guidelines into this short blog post.
Being successful is not the problem
The Guidelines start with a familiar but important principle: Article 102 TFEU does not prevent a company from acquiring a dominant position through its own merits. Also, a dominant company may compete, and compete vigorously. What it may not do is use methods different from those governing regular competition in a way that harms effective competition. However, turning that formula into an operational test has never been easy.
According to the Guidelines, conduct conceptually distorts effective competition where two conditions are met: First, the conduct must depart from competition on the merits. Second, it must be capable of producing exclusionary effects. An objective justification may nevertheless prevent the conduct from being abusive.
The structure sounds reassuringly straightforward. Its application is less so. “Competition on the merits” is not a self-executing concept, and many commercial practices can harm rivals precisely because they make a company’s offering more attractive to customers. The central challenge remains distinguishing legitimate competitive practices from conduct that weakens the competitive process itself.
The dominant position: Market shares ≠ end of the story
The Guidelines provide a broad framework for assessing dominance. First of all, market shares remain an important starting point. A market share of 50% or more over a sustained period is generally evidence of dominance. Based on the Commission’s experience, dominance is generally unlikely below 40%, although the Guidelines make clear that it can still be established below that level where other circumstances support such a finding.
At the same time, the Commission emphasises that market shares must be interpreted in their economic context. Their development over time, the strength of competitors, barriers to entry and expansion, countervailing buyer power and the characteristics of the market may all be relevant.
This is particularly important in innovative and digital markets. In fast-growing markets with short innovation cycles, high market shares may be less informative if they are likely to be temporary. In zero-price markets, metrics such as user numbers, transactions or usage intensity may provide a better indication of market power. In markets characterised by frequent and significant R&D investments, expenditure on R&D, patents or patent citations may also be relevant.
Important in practice: The Commission recognises that market power in an aftermarket may be constrained by competition in the corresponding primary market. Dominance in the aftermarket is therefore unlikely where customers make informed lifecycle-cost decisions and would react to exploitative aftermarket conduct by switching primary suppliers within a reasonable amount of time.
So, dominance cannot always be assessed by putting market shares into a spreadsheet and stopping there.
Departure from competition on the merits and exclusionary effects
With regard to competition on the merits, the Commission stresses that Article 102 TFEU does not protect less efficient competitors, but rather the competitive process itself, and that dominant firms retain a special responsibility not to impair effective competition. In assessing exclusionary effects, the focus is on whether the conduct is capable of restricting competitors’ ability to compete and thereby harming consumers through reduced choice, quality, innovation or competitive pressure. The required effects analysis is context-specific and varies depending on whether the conduct is pricing, non-pricing or multi-faceted behaviour. Tools like the equally efficient competitor test are relevant in some cases but not universally required.
A practical guide to the usual suspects
The Guidelines’ most practical contribution may be their detailed treatment of specific forms of exclusionary conduct. Admittedly, not everything here is new – the Guidelines are both a consolidation of existing case law and a signal that Article 102 enforcement will remain a highly fact-sensitive exercise:
- Predatory pricing: This refers to a below-cost pricing practice. The Commission largely retains the familiar cost-based test. Prices below average variable cost are considered predatory. Prices between average variable cost and average total cost are abusive where they form part of a plan to eliminate or reduce competition. Prices above average total cost cannot be regarded as predatory. Once predatory pricing is established, the Commission does not have to conduct a separate effects analysis, demonstrate that the pricing covered a substantial part of the market, or prove that the dominant undertaking will eventually recoup its losses.
- Margin squeeze: This concerns a vertically integrated undertaking which sells a product, for which it is dominant, as an input to undertakings in a downstream market and competes with those same undertakings on the downstream market. According to the Commission, the question is whether an equally efficient downstream competitor could operate profitably given the dominant undertaking’s upstream and downstream prices. Importantly, neither price must be abusive in isolation. The upstream price does not have to be excessive, and the downstream price does not have to be predatory. It is the spread between them that matters. Where that spread results in a negative margin, the Commission considers exclusionary effects probable and may conclude, absent contrary evidence, that competition has been distorted.
- Conditional rebates: In essence, conditional rebates are incentives offered by a dominant company, in the form of monetary or non-monetary benefits, to encourage customers to purchase in a particular way. Rebates that are not conditional on exclusivity will receive a contextual assessment. The Commission will examine factors such as market coverage, duration, rebate thresholds, transparency and whether the undertaking is an unavoidable trading partner. The design of the rebate is particularly important. Retroactive rebates, which apply to all purchases once a threshold is reached, are generally more concerning than rebates applying only to additional units. Individualised thresholds and long reference periods may increase that effect further. Where the rebate can reliably be quantified, the Commission will generally assess whether an equally efficient competitor could contest the relevant demand. However, a price-cost test is not mandatory in every case. This leaves the Commission considerable flexibility where the exclusionary mechanism cannot readily be reduced to numbers.
- Exclusive dealing: This includes not only express requirements to purchase all or most demand from the dominant undertaking, but also arrangements producing the same result in practice (including exclusive supply obligations). Exclusive dealing is presumed to distort effective competition. It makes no difference whether the exclusivity was imposed by the dominant undertaking or requested by the customer. The presumption is rebuttable, but the dominant undertaking must come forward with evidence grounded in the actual competitive conditions.
- Tying and bundling: Tying consists of offering a specific product only together with another product. The tie may be contractual or technical. Mixed bundling, where separate purchasing remains possible but is, for example, more expensive, falls under the rebate framework. Tying distorts effective competition where four conditions are met: Two separate products, dominance in the tying-product market, an absence of customer choice to obtain the tied product alone (‘coercion’) and a capability to produce exclusionary effects. Whether the products in question are actually distinct depends primarily on whether customers would purchase the tied product independently.
- Access restrictions: One of the Guidelines’ most important distinctions is that between access restrictions and outright refusals to supply. Access restrictions include commercially unviable terms, delays, procedural obstacles and other practices that hinder access to infrastructure, services, intellectual property or data. Such conduct is assessed under the general test: It must depart from competition on the merits and be capable of producing exclusionary effects. Crucially, the input does not have to be indispensable. A dominant undertaking may therefore face intervention without satisfying the strict conditions traditionally associated with the essential-facilities doctrine.
- Refusal to supply: The stricter test is reserved for a genuine refusal to supply an input that the dominant undertaking owns and developed solely for its own use. An obligation to supply in such circumstances directly affects property rights, freedom of contract and investment incentives. Against this background, a finding that such refusals to supply distort effective competition requires that two strict conditions be satisfied: The input must be indispensable, meaning that it cannot realistically be duplicated and no viable substitute is available. The refusal must also be capable of eliminating all effective competition from the requesting undertaking. Where intellectual property is concerned, the refusal must additionally limit technical development.
- Self-preferencing: This refers to the practice of favouring a company’s own products or services over those of competing undertakings. The Guidelines emphasise that there is no general prohibition on dominant companies favouring their own products. Problems may arise where an undertaking uses its dominance in one market to favour its products in another through means that depart from competition on the merits and can exclude rivals. Possible examples include preferential ranking, steering users and manipulating selection mechanisms.
- Conduct harmful by its very nature: Finally, the Guidelines identify conduct that is considered harmful by its very nature. This includes practices with no apparent economic purpose other than restricting competition, such as paying customers not to sell a competitor’s products or deliberately dismantling infrastructure on which a rival depends. Once conduct falls into this category, it is deemed to distort effective competition. The Commission considers that an effects-based rebuttal could succeed only exceptionally, although the EU Courts have not yet conclusively resolved that question. This is the closest the Guidelines come to an Article 102 equivalent of a restriction by object.
Conclusion
The Guidelines do not reinvent Article 102 TFEU, but they do make the Commission’s enforcement priorities more explicit and more structured. For businesses, the key takeaway is that legal certainty increases in some areas, while the Commission also preserves considerable room for case-by-case assessment. Market shares still matter, but context matters more; effects still matter, but not always in the same way. The Guidelines are expressly restricted to exclusionary conduct, so exploitative abuses remain a blind spot.
Photo by AbsolutVision on Unsplash
Antitrustpolitics.com 