Contents

Introduction. 1

General Principles Applicable to the Assessment of Dominance 2

General Principles to Determine If Conduct By a Dominant Undertaking Is Liable to Be Abusive. 3

Principles to Determine Whether Specific Categories of Conduct Are Liable to Be Abusive. 5

General Principles Applicable to the Assessment of Objective Justifications 6

Recommendations 7

Conclusion. 8

Endnotes 9

Introduction

On July 31, 2024, the European Commission (Commission) issued draft guidelines on the application of Article 102 on the Treaty of the Functioning of the European Union to abusive exclusionary conduct by dominant undertakings (Guidelines).[1] The Guidelines replace the Commission’s earlier Guidance on the Commission’s enforcement priorities in applying Article 82 of the EC Treaty to abusive exclusionary conduct by dominant undertakings (Enforcement Priorities).[2]

The Information Technology and Innovation Foundation (ITIF), the world’s top-ranked science and technology policy think tank, greatly appreciates the opportunity to respond to the Commission’s public consultation and comment on the Guidelines from the standpoint of promoting sound and pro-innovation competition policy in Europe. While ITIF commends the Commission for attempting to “enhance legal certainty and help undertakings self-assess whether their conduct constitutes an exclusionary abuse under Article 102 TFEU,” this comment highlights issues with the analytical framework adopted by the Guidelines, and especially their general untethering of Article 102 liability from harm to competition and consumers.[3]

This comment proceeds in six parts that mirror the structure of the Guidelines. The first analyzes the Guidelines’ standards for determining whether an undertaking holds a dominant position, particularly its market share presumption and safe harbor threshold. The second part expounds upon and critiques the Guidelines’ general principles for demonstrating exclusionary abuses under Article 102. Part three critically evaluates the specific legal tests provided by the Guidelines for particular types of exclusionary conduct. The next part assesses the Guidelines’ criteria for showing that conduct which is capable of producing exclusionary effects may nonetheless constitute competition on the merits by virtue of having an objective justification. A brief conclusion follows.

General Principles Applicable to the Assessment of Dominance

The Guidelines make clear that a market share above 50 percent will “save in exceptional circumstances” be sufficient to presume the existence of a dominant position.[4] In so doing, the Guidelines reflect a marked shift from the Enforcement Priorities, which in lieu of any similar presumption, stated that “the Commission will interpret market shares in light of the relevant market conditions, and in particular of the dynamics of the market and of the extent to which products are differentiated.”[5] In support of this change, the Guidelines cite no intervening case law. Instead, they refer back to Akzo v. Commission, which is itself in tension with subsequent Commission decisions which not only found that dominance did not exist despite a 50 percent share but declined to apply the Akzo presumption despite the existence of a market share in excess of 50 percent and expressly considered additional relevant evidence, such as whether there were significant barriers to entry.[6]

The lack of any standalone market share presumption of dominance is consistent with both the U.S. approach and sound policy. Under U.S. law, courts have been clear that to make out a circumstantial case for monopoly power, significant barriers to entry must also be demonstrated.[7] This ensures that unilateral conduct requirements only apply to firms with durable market or monopoly power that enables them to charge supra-competitive prices or otherwise exploit consumers. Indeed, a standalone market share presumption for Article 102 is particularly concerning given that a 50 percent dominance threshold is significantly lower than the approximately 70 percent market share U.S. courts typically require to show monopoly power.[8] With market share presumptions using such a low threshold, the Guidelines thus risk fostering an environment where firms are subject to Article 102 despite not having any real ability to harm consumers—a recipe for false positives and the stifling of competition.

This concern about over-enforcement is exacerbated by the Guidelines’ statement that, save in exceptional circumstances, a safe harbor from a finding of dominance only exists when market shares fall below 10 percent.[9] And here again, without citing any intervening case law, the Guidelines substantially depart from the Enforcement Priorities, which emphasized that “dominance is not likely if the undertaking’s market share is below 40% in the relevant market.”[10] Indeed, in support of this new threshold, the Guidelines cite only one nearly 40 year-old case, and in so doing, appear to conflate a sufficient condition for lack of dominance with a necessary one: the Commission itself continues to maintain that under EU law “[i]f a company has a market share of less than 40%, it is unlikely to be dominant.”[11]

Opening the door to generally imposing Article 102’s requirements on firms that have market shares below 40 percent would likely chill innovation and other pro-consumer behavior. For example, as ITIF has explained:

Market power that does not take the form of durable monopoly power very often falls well short of the point on the inverted-U curve where concentration is likely to harm innovation. Indeed, many studies in the business strategy literature suggested that healthy dynamic competition is optimized with approximately three equal sized firms, each of whom may have some degree of market power that helps to facilitate innovation competition.[12]

The Guidelines also contain a discussion regarding “collective dominance”—yet another novelty relative to the Enforcement Priorities, which did not conduct any analysis of this issue. As the Guidelines explain, collective dominance can first be shown directly through the existence of an agreement either express or implied by conduct in the form of “structural or other links (e.g., personal ties).”[13] And yet, imposing Article 102 restrictions for these forms of collective dominance would seem to cover no behavior that is not already prohibited under Article 101, which already encompasses concerted restraints that result in exclusionary effects.[14] As such, expanding liability for collective dominance in this way risks creating not only unnecessary redundancy within EU competition law, but potential confusion about the standards that will be applied to evaluate concerted restraints that have exclusionary effects.

The Guidelines also state that collective dominance can be established if there is sufficient evidence of firms engaging in tacit coordination divorced from any express or implied agreement.[15] To be sure, while tacit coordination is not in itself unlawful under U.S. law, for example, in certain circumstances courts will inquire as to whether unilateral facilitating practices can violate the antitrust laws.[16] But here the Guidelines appear to make a category mistake: while it is true that unilateral practices which facilitate tacit coordination can harm competition, the mechanism by which they do so is collusion, not exclusion. Indeed, exclusionary behavior can actually make coordination less likely by undermining cooperation or other collusive dynamics between firms. As such, attaching Article 102 liability to collective dominance achieved through tacit coordination may have the counterproductive result of encouraging the very collusion that is the primary means by which consumers are harmed in these contexts.

General Principles to Determine If Conduct By a Dominant Undertaking Is Liable to Be Abusive

The Guidelines state that a necessary, but not sufficient, condition for a firm to violate Article 102 is engaging in conduct that departs from competition on the merits. The Guidelines state that conduct does not depart from competition on the merits simply because it marginalizes “competitors that are less efficient that than the dominant undertaking.”[17] In addition to this question of whether “a hypothetical competitor as efficient as the dominant undertaking would be unable to adopt the same conduct,” the Guidelines list other factors that Union courts have analyzed to determine whether behavior departs from competition on the merits, such as whether the conduct “is considered as abnormal or unreasonable in light of the market circumstances at stake.”[18] However, the Guidelines elsewhere suggest that failing any sort of no-economic sense or profit sacrifice test is not necessary to show a lack of competition on the merits. For example, in the case of margin squeezes, the Guidelines note that the “price-cost test is applied from the perspective of a hypothetical as efficient downstream competitor.”[19]

The Guidelines further declare that an additional necessary—and, in conjunction with a departure from competition on the merits, jointly sufficient—condition for conduct to be deemed an exclusionary abuse under 102 is “capable of having exclusionary effects.”[20] The Guidelines underscore that the capability of producing exclusionary effects “must be more than hypothetical” but can nonetheless be satisfied without showing either “actual harm to competition” or “direct consumer harm.”[21] That is, instead of competition and consumer welfare, the Guidelines put forward an analysis that is structural and focused on “the importance of actual or potential competitors for the maintenance of effective competition,” the protection of which the Guidelines elsewhere describe as the central purpose of Article 102.[22]

The Guidelines also elaborate upon the presumptions that apply to many specific forms of conduct as presumptively either departing from competition on the merits, being capable of producing exclusionary effects, or both. For example, naked restrictions are presumptively both not competition and the merits and capable of producing exclusionary effects.[23] Moreover, while conduct like predatory pricing is not presumed to depart from competition on the merits, it is presumed to be capable of producing exclusionary effects.[24] By contrast, other forms of conduct, like self-preferencing, are subject to no presumption either that they depart from competition on the merits or are capable of producing exclusionary effects.[25]

The Guidelines thus appear to view Article 102 as condemning behavior that departs from competition on the merits by virtue of excluding a hypothetically equally efficient rival and which is capable of producing exclusionary effects because it harms competitors in a way that “impairs effective competition.”[26] Indeed, this view appears to be consistent with the recent decision by the European Court of Justice in Intel,which made clear that “in order to find, in a given case, that conduct must be categorised as ‘abuse of a dominant position’, it is necessary, as a rule, to demonstrate, through the use of methods other than those which are part of competition on the merits between undertakings, that that conduct has the actual or potential effect of restricting that competition by excluding equally efficient competing undertakings from the market or markets concerned or by hindering their growth on those markets.”[27]

In so doing, the Guidelines diverge substantially from the framework of the Enforcement Priorities, which did not treat a lack of competition on the merits and a capability of producing exclusionary effects as two distinct elements of an Article 102 violation. Instead, there the Commission distinguished between “anti-competitive foreclosure” and “price-based exclusionary conduct.”[28] And, as concerns the former, the Commission explicitly defined “anti-competitive foreclosure” in terms of conduct allowed a dominant firm to “profitably increase prices to the detriment of consumers.”[29] As such, the Guidelines reflect a problematic shift away from requiring harm to consumers and competition to show an exclusionary abuse under Article 102, a change which is aggravated by the presumptions in the Guidelines noted supra whereby several forms of conduct are deemed presumptively capable of producing exclusionary effects. The Guidelines thus not only radically deviate from U.S. legal standards, which require that exclusionary conduct harm consumers and competition, but risk formalizing a unilateral conduct enforcement regime that proscribes practices that may benefit consumers and competition simply because they harm competitors in a way that threatens effective competition—a decentralized competitive ideal inapposite to the dynamic Schumpeterian competition that typifies myriad markets in the modern high-tech economy.

While it is true that Union courts have suggested that competition on the merits is not inconsistent with harm to a less efficient competitor, the Guidelines’ apparent preference to make an equally efficient competitor standard sufficient to show that conduct departs from competition on the merits is also troubling as a general rule for analyzing exclusionary behavior, especially in dynamic and digital contexts.[30] For example, with respect to “predatory innovation,” it borders on the nonsensical to inquire whether a product improvement harmed a hypothetically as-efficient rival, as the answer is effectively always both “yes” and “no”: the fact of the innovation itself suggests that the firm had some unique efficiency or dynamic capabilities. As such, U.S. courts often condone such behavior if there is a legitimate procompetitive justification, regardless of whether it is ultimately outweighed by anticompetitive harms, which can be difficult to balance both in theory and practice.[31] Moreover, the as-efficient competitor test can also result in administrability issues, especially concerning multi-product or non-price conduct, that stem from difficulties associated with artificially constructing a hypothetically as efficient competitor to be used as a benchmark.

Principles to Determine Whether Specific Categories of Conduct Are Liable to Be Abusive

The Guidelines’ treatment of exclusive dealing varies from naked restrictions like “payments by the dominant undertaking to customers that are conditional on the customers postponing or cancelling the launch of products that are based on products” and non-naked conduct like “exclusive supply or purchasing agreements” and “rebates conditional upon exclusivity” that are only treated as presumptively capable of producing exclusionary effects.[32] Notwithstanding the highly problematic nature of the common presumption that exclusive dealing is capable of resulting in exclusionary effects—low amounts of foreclosure may not have any meaningful effect on competitors, let alone consumers or competition—the Guidelines are less than clear as to what the Commission must show to prove that non-naked exclusive dealing prima facie departs from competition on the merits—an analytical gap that limits the Guidelines’ efficacy in articulating the legal standards that apply to this behavior.

The Guidelines’ discussion of the standards that apply to tying and bundling are simply diverse. While not sounding in any naked restriction, the Guidelines assert that some forms of tying can, like exclusive dealing, be presumed capable of resulting in exclusionary effects, whereas other forms of tying and bundling may not.[33] However, here again, it is unclear what the Commission must generally show to prove that tying or bundling departs from competition on the merits. And, to the extent the coercion element serves this purpose, the test risks chilling procompetitive conduct: for behavior like technical tying, products may be thoroughly integrated in a way that not just benefits consumers but also makes economic sense but for any capability of resulting in exclusionary effects.

Unlike exclusive dealing or tying, refusals to supply are not presumed to either depart from competition on the merits or be capable of producing exclusionary effects. Rather, to show the former, the Guidelines suggest that the Commission must demonstrate indispensability; for the latter, the “capability to eliminate all competition on the part of the requesting undertaking.”[34] But once again, notwithstanding the general problem in condemning unilateral behavior without proving harm to competition or consumers, the indispensability requirement is not sufficient to show that the refusal departed from competition on the merits. For example, in U.S. law, not only has the Supreme Court never recognized an essential facilities doctrine, but refusals to deal more generally are only condemned if there is a termination of a prior course of dealing, which serves to avoid chilling innovation by ensuring that the refusal did not make economic sense.[35]

Despite its differences from exclusive dealing, the Guidelines presume that predatory pricing is capable of restricting competition, such that all that remains for the Commission to prove in its prima facie case is that the pricing was below-cost and thus departed from competition on the merits.[36] To be sure, while the price-cost test can also be de facto relevant to assessing whether the pricing was capable of producing exclusionary effects, below-cost pricing is neither a necessary nor sufficient condition for foreclosure: not only can above cost pricing exclude rivals, but below-cost pricing may not be sufficient to undercut competitors if they are more efficient. Indeed, given the lack of any recoupment requirement to show actual harm to consumers, such a presumption is all the more likely to condemn conduct which benefits consumers with low prices.

Closely related to predatory pricing are margin squeezes, which the Guidelines also treat as presumptively capable of producing exclusionary effects in the case of a negative spread. Accordingly, they here again condone the use of a price-cost test to determine whether the conduct would exclude an equally efficient competitor and thus depart from competition on the merits.[37] However, treating a price-cost test as a de facto sufficient condition to show a prima facie lack of competition on the merits is even more concerning for margin squeezes than predatory pricing. This is because, as the U.S. Supreme Court has made clear, unlike predatory pricing, margin squeezes are premised upon an underlying refusal to deal, which should only be condemned, as discussed supra, in the event of the termination of a prior course of dealing so as to avoid condemning procompetitive behavior that makes economic sense.[38]

In addition to behavior that the Guidelines describe as having a “specific legal test,” the Guidelines discuss several other types of conduct categories that lack a specific legal test “but for which the Union Courts have provided guidance as to how to apply the general legal principles set out in section 3.”[39] For these behaviors, there are accordingly no presumptions either that the conduct departs from competition on the merits or is capable of producing exclusionary conduct. Nonetheless, the lack of showing either harm to consumers—in all cases—or also that the conduct fails the no-economic sense test—especially in the cases of self-preferencing and access restrictions—raises concerns similar to those discussed supra about false positives and chilling innovation.

General Principles Applicable to the Assessment of Objective Justifications

As the Guidelines explain, conduct that is capable of producing exclusionary effects and prima facie not competition on the merits may ultimately be deemed competition on the merits if objective justifications can be presented.[40] To make out such an objective justification, the Guidelines describe how the undertaking must first show “that the efficiency gains likely to result from the conduct under consideration counteract any likely negative effects on competition and on the interests of consumers in the affected markets.”[41] In so doing, to show efficiencies, the Guidelines appear to introduce a “likelihood” burden of proof that is higher than the “capability” burden needed for the Commission show exclusionary effects—a double standard that will undoubtedly and unfairly tip the scales in favor of the Commission for Article 102 cases.[42]

Next, the Guidelines state that efficiency gains presented as part of an objective justification must be shown to “have been, or are likely to be, brough about as a result of the conduct.”[43] Through this criterion, the Guidelines appear to introduce a sufficient causation requirement that seems to stack the deck in the Commission’s its favor. That is, the Guidelines nowhere appear to expressly impose any sort of similar causation requirement on the Commission to prove by a likelihood that the anticompetitive conduct resulted in a capability to result in exclusionary effects. And, to the extent such a causation standard inheres in the “capability of producing exclusionary effects” element,” here again there would be a double standard: a lower reasonable capability causation standard to show harm, and a higher likelihood standard to show an objective justification.

Third, the Guidelines also include an additional necessary causation requirement for firms to prove an objective justification whereby undertakings must show “that the conduct is necessary for the achievement of those efficiency gains.”[44] To be sure, while necessary causation requirements are well established in U.S. law in the context of agreements—the least restrictive alternative test—and mergers—a lower “unlikely to be accomplished” or reasonably necessary test—courts do not typically inquire into necessary causation for unilateral conduct offenses.[45] And the reason is both sound and straightforward: as the U.S. Supreme Court has explained, antitrust law should not be a license “carte blanche to insist that a monopolist alter its way of doing business whenever some other approach might yield greater competition.”[46]

Fourth, the Guidelines make clear that they will only credit an objective justification that “does not eliminate effective competition, by removing all or most existing sources of actual or potential competition.”[47] However, limiting the admissibility of objective justifications only to those where the underlying conduct does not come at the expense of effective competition is likely to chill procompetitive behavior. This is in large part because, as noted supra, conduct that allows firms to achieve concentration that is inconsistent with effective competition may nonetheless spur dynamic efficiencies, Schumpeterian competition, and long-run consumer benefits that greatly outweigh any short-run static harms.[48]

Recommendations

For these reasons, ITIF has significant concerns about the Guidelines and offers the following recommendations:

▪ Dominance thresholds. The Commission should reconsider the Guidelines’ 50 percent presumption of dominance as well as the 10 percent safe harbor. Both are likely to result in Article 102’s requirements being imposed on undertakings which do not have any durable market power that would allow them to engage in behavior that harms consumers or competition.

▪ General principles. The Guidelines’ putative two-pronged framework for understanding exclusionary abuses under Article 102 as conduct that departs from competition on the merits by excluding an equally efficient rival, and which harms competitors in a way that is capable of producing exclusionary effects, will chill procompetitive behavior that both benefits consumers and, in certain cases, also makes economic sense but for any exclusionary effects. Rather than push forward with the Guidelines, the Commission should continue to utilize the framework of its earlier Guidance and attendant focus on competition and consumer welfare.

▪ Specific legal tests. Not only do the Guidelines fail to explain how the Commission must show that non-naked exclusive dealing and tying prima facie depart from competition on the merits, but presumptions that behavior like refusals to supply, predatory pricing, and margin squeezes with negative spreads are capable of resulting in exclusionary effects will result in false positives. Moreover, demonstrating that behavior involving refusals to deal departed from competition on the merits should require some showing that the refusal did not make economic sense and not just that an equally efficient competitor would be excluded.

▪ Objective justifications. The Guidelines’ criteria for demonstrating an objective justification risk, among other things, creating a double standard between what the Commission has to prove to make out its case and what an undertaking needs to show to present an objective justification for its behavior. 

Conclusion

At bottom, the Guidelines represent the continuation of reactionary effort by the Commission to move away from the more economic approach reflected in the Enforcement Priorities that tethered exclusionary abuses to “likely negative effects on competition and consumer welfare” and return to the ordoliberal model of condemning behavior that is deemed to unfairly harm rivals in a way that threatens effective competition.[49] The Guidelines suggest that this change is a response to “growing market concentration in various industries and the digitization of the Union economy, which makes strong network effects and ‘winner-takes-all’ dynamics increasingly widespread.”[50] Not only are those concerns unfounded—claims about increased concentration are routinely overblown and leading dominant firms are often not the first-movers in their markets, which in any case regularly display multi-homing—but they are inapposite with the very nature of the Schumpeterian competition that typifies the modern high-tech economy.[51]

Recent history leaves little doubt about the consequences these Guidelines may have. Beginning in the 1970s, the U.S. and EU began to take two very different approaches to competition policy. While the former generally condemned exclusionary behavior only if it resulted in harm to consumers, the latter pushed forward with an ordoliberal model that is largely consistent with the Guidelines. And the results of this experiment have long been clear. Whereas the U.S. led a digital revolution driven by Schumpeterian and leapfrog competition by large technology firms that helped America maintain its technological and economic edge amidst a rising China, Europe is now faced with a productivity gap due to its “failure to capitalise on the first digital revolution led by the internet—both in terms of generating new tech companies and diffusing tech into the economy.”[52] Indeed, by the time this competition policy error was evident and the EU was in the process of shifting toward the more economic approach reflected in the Enforcement Priorities, it was already too late:

“…by the end of [2004], Google had already gone public, Amazon had beta tested its first cloud offering, Apple had begun work on the iPhone, and Facebook had been founded.”[53]

In the words of French President Emmanuel Macron, Europe must “take responsibility for the evolution of our competition policy” and specifically the “ordo-liberal” model of competition.[54] Unfortunately, the Guidelines do the opposite and double down on the ordoliberal decentralized ideal of competition that will again fail Europe at what could not be a worse time. That is, the Guidelines come amidst yet another Schumpeterian gale of creative destruction in the form of artificial intelligence, an area where Europe is well already behind: “[a]round 70% of foundational AI models have been developed in the US since 2017 and just three US ‘hyperscalers’ account for over 65% of the global as well as of the European cloud market.”[55]

The urgency of this techno-economic moment is intensified by the great geopolitical challenge facing the West that will define the 21^st century: the rise of China. Specifically, whereas the digital revolution empowered America’s economy to remain strong, over the past 40 years, Western Europe’s share of global wealth has fallen by half as China’s has grown by nearly 1000 percent. Rather than pursue competition policy that is hostile to large firms, both America and Europe should thus be focusing on fostering scale-driven Schumpeterian innovation competition. Doing so will ensure that the West remains techno-economically on top and enjoys the scale necessary to compete with heavily subsidized Chinese rivals that, as ITIF has warned, in many areas “will likely equal or surpass Western firms within a decade or so” unless appropriate pro-innovation policies are put into place, including in competition policy.[56]

To be sure, despite recognizing the critical significance of the current moment for Europe’s political economy, the Draghi Report itself praises the Digital Markets Act which, much like the current Guidelines, imposes a rule regime for exclusionary conduct divorced from harm to consumers and competition. Of course, perhaps the primary effects of such policies will be on American firms, who are the lords of the digital age, and not European upstarts. But that would be a serious miscalculation: not only will European innovation and competitiveness suffer from overly burdensome unilateral exclusionary conduct standards, whether imposed through ex ante regulation or ex post enforcement, but as ITIF has explained, in this existential rivalry with China, Europe and America must either be in it together or not at all.[57]

Thank you for your consideration.

Endnotes