Following its 2023 consultation, the European Commission published draft guidelines on unilateral exclusionary conduct early August ahead of the Summer break (the Guidelines). Comments are invited by 31 October 2024. The Guidelines aim to reflect the developments in the case-law and the Commission’s ‘extensive experience’ enforcing Article 102 TFEU.

The Guidelines will replace the Commission’s 2009 Guidance on Enforcement Priorities on exclusionary abuses (the Enforcement Priorities). The Enforcement Priorities were the cornerstone of the so-called ‘more economic approach’  which were intended to shift Article 102 enforcement away from a form-based approach to focus on whether unilateral conduct had anti-competitive effects. This aim was underpinned by the principle that Article 102 sought to protect consumers (or the competitive process), not competitors. The Union’s Courts have since enshrined the effects-based approach in a slew of seminal cases from Post Danmark through to Intel. 

The backdrop to the Guidelines has been regulatory concerns that the Enforcement Priorities may have been too far-reaching in shifting Article 102’s enforcement framework, making it unduly difficult for the Commission to adopt prohibition decisions within a reasonable time. The Commission’s 2023 Policy Brief on Article 102 questioned whether the ‘heightened substantive legal standard’ under effects-based enforcement may have ‘inadvertently lead to undesirable outcomes’. Olivier Guersent more pithily suggested that the Commission may have ‘made its own misery’. 

This blogpost focuses on how the Guidelines propose to treat the assessment of exclusionary effects. While the Guidelines raise a host of issues over which much ink will be rightly spilt, effects remain at the heart of Article 102: we explore how the Guidelines approach the assessment of effects; the degree to which they seek to lower the threshold for proving effects and why some parts of the Guidelines risk undermining the effects-based approach which protect customers (and ultimately consumers) against anti-competitive effects (including higher prices, less quality, less innovation and less variety or choice).

The draft guidelines and their approach to effects

The structure of the Guidelines follows the conditions for establishing whether a dominant firm’s conduct is abusive, namely whether such conduct: (A) deviates from competition on the merits (B) is ‘capable’ of exclusionary effects and (C) cannot be justified on objective grounds or because of overriding efficiencies. 

The Guidelines summarise three main rules for assessing effects:

• The need to prove exclusionary effects depends on the type of conduct: the Guidelines delineate between: (A) naked restrictions where a firm can only ‘exceptionally’ rebut a presumption of exclusionary effects; (B) exclusive dealing, predatory pricing, margin squeezing and ‘certain forms’ of tying which are subject to a rebuttable presumption of exclusionary effects; and (C) all other conduct, for which the Commission must prove the ‘capability to produce’ such effects. Although the presumptions are in principle rebuttable, conduct presumed to have exclusionary effects is deemed to have ‘a high potential’ to be anti-competitive. 

• The relevant effects are ‘exclusionary’ effects: exclusionary abuses concern conduct that ‘can harm consumers’ by hindering the existing degree of competition or the growth of competition on a market. The Guidelines omit, in particular, any statement that Article 102 focuses on anti-competitive foreclosure (i.e. foreclosure of competitors as efficient as the dominant firm which is likely to result in consumer harm). The Guidelines instead stress that it is not necessary to prove ‘direct consumer harm’ nor that the competitors on the market ‘are as efficient as the dominant undertaking’. The key enforcement goals and evidentiary standards set out in paragraphs 19 and 20 of Enforcement Priorities, which articulated the need for anti-competitive foreclosure, have no place in the Guidelines.
• Where effects must be proven, such effects must be more than hypothetical on the basis of specific and tangible points of evidence: conduct must be ‘at least capable’ of producing ‘potential’ exclusionary effects - there is no need to prove ‘actual exclusionary effects’. Such effects must be ‘more than hypothetical’ but there is no requirement that they must be likely, as was the case in the Enforcement Priorities. The Guidelines recognise that, when the Commission needs to prove effects, it needs to do so ‘on the basis of specific, tangible points of analysis and evidence’.

Have the Guidelines moved the legal bar on the effects-based approach? 

While the Guidelines ostensibly seek to codify the case law on Article 102, they side-step a number of (key) elements in the jurisprudence concerning the assessment of effects. 

The Guidelines first overstate the circumstances and degree to which the Commission need not prove exclusionary effects. The Court of Justice held in Intel that the Commission must prove effects for exclusive dealing where the dominant firm submits that its conduct was not capable of restricting competition on the basis of supporting evidence. However, neither Intel nor any other judgment finds that the Commission faces a lower legal standard or evidentiary burden for demonstrating effects where the dominant firm has successfully rebutted the presumption of exclusionary effects. Indeed, Intel indicates the opposite. 

The Guidelines also significantly broaden the notion of ‘effects’. The Courts have repeatedly held that ‘not every exclusionary effect is necessarily detrimental to competition’: competition may by its very nature foreclose less efficient rivals. When it comes to pricing practices, the Courts have stressed the ‘as efficient competitor’ principle because it is foreclosure of as-efficient rivals that matters. Yet the Guidelines notably fail to offer a definition of what must be understood by actionable exclusionary effects and reduce the appropriateness of an ‘as efficient competitor’ test in the analysis. In doing so, the Guidelines move away from the spirit of the Enforcement Priorities and ignore significant parts of the case law.

Finally, the position that effects need only be ‘more than hypothetical’ ignores that the Union’s courts have repeatedly held that effects must be ‘probable’ or ‘likely’ to be actionable (i.e. exclusionary on the balance of probabilities). The Guidelines also manage to impose an asymmetric evidentiary burden for anti- and pro-competitive effects: while the Commission need only advance ‘specific, tangible points of analysis and evidence’ to demonstrate exclusionary effects, dominant firms must provide a ‘cogent and consistent body of evidence’ to substantiate objective necessity or countervailing efficiencies. This was again different under the Enforcement Priorities where the Commission would only prioritise enforcement if there was cogent and convincing evidence that the alleged abusive conduct was likely to lead to anti-competitive foreclosure.

A ‘new’ effects-based approach or a return to form-based assessment? 

That the Guidelines seek to mitigate perceived weaknesses of the effects-based approach makes sense from an enforcement policy perspective. However, the ‘new’ approach risks undermining effective effects-based enforcement. While the Enforcement Priorities were clear in what they stood for, the Guidelines run the risk of abandoning the key building blocks for robustly assessing anti-competitive effects in what could be viewed as a pursuit of greater administrative expediency. Therefore, we would, in particular, recommend – at a minimum – that the Commission consider at least the following changes. 

Using properly calibrated presumptions for conduct highly likely to have anti-competitive effects makes sense (and indeed the Enforcement Priorities had wording to this effect for naked restrictions): aiding legal certainty and reducing the Commission’s administrative burden whilst minimising any chilling effect on pro-competitive conduct. The Guidelines ignore, however, two critical safeguards. First, the categories of such conduct should be interpreted narrowly: otherwise conduct unlikely to have exclusionary effects will too easily become presumptively abusive. Second, there is no basis for a lower evidentiary standard if a dominant firm has successfully rebutted the presumption of exclusionary effects (for example by pointing out that the conduct does not prevent as-efficient rivals from competing profitably or showing that exclusionary effects are unlikely because of low market coverage, to name just the most straightforward defences). Furthermore, lowering the standard for the Commission in those circumstances risks being unworkable: where on the spectrum between proving effects and not proving effects would it lie? 

Discarding the notion of anti-competitive effects also risks undermining the key principle of rigorous effects-based enforcement: the distinction between ‘aggressive, yet healthy and permissible, competition’ (foreclosure) and conduct that harms consumer welfare (anti-competitive foreclosure). It incorporates the need to prove what is sometimes called antitrust injury and is also part of the non-horizontal merger guidelines, which also deal with the risk of (anti-competitive) foreclosure. While the 2023 Policy Brief observed that protecting less efficient rivals may be warranted in some circumstances (e.g., a new entrant at the early stage of market entry), this does not justify dispensing with the need to prove anti-competitive effects, beyond mere foreclosure of rivals. Absent such clarification, the Commission would be able to prohibit any conduct that it deems not to be competition on the merits (which is a broad test) and may exclude one or more rivals, irrespective of whether the effects of such conduct are in fact pro-competitive.

Finally, there is no basis for lowering the standard for establishing exclusionary effects below the balance of probabilities or adopting different evidential burdens for anti- and pro-competitive effects. The Court of Justice only recently endorsed the balance of probabilities for EU merger control in CK Hutchison. There is no plausible reason why the standard should be different under Article 102.