Converging or Diverging Paths – The EU and U.S. Antitrust Approaches to Unilateral Refusals to Deal by Dominant Firms in a Transatlantic Comparison

Autor: Lennart Levita*

Artikel herunterladen


It is a well-known – if not globally recognized – principle, that every individual, and therefore every business, has the freedom to choose to contract or not to contract with others. It is neither favored nor intended under the tenets of globally recognized antitrust principles to override the freedom of contract in private business transactions by judicial intervention.[1] Therefore, antitrust law does not impose an affirmative obligation on a market participant to deal with its rivals. Instead, most legal systems interpret unilateral refusals to deal (RTDs) as an affirmative right to refuse, with doctrinal foundations in the freedom of contract and free market economy.[2] Nonetheless, limitations of a single-firms right to refuse to deal might be necessitated, when undertakings are acting under monopolizing conducts and thereby harmful.

Even though both EU competition law and U.S. antitrust law generally restrict abusive conducts by dominant undertakings concerning RTDs, they diverge significantly on the question under what conditions a single firm should have a duty to deal with another. While the EU approach favors judicial intervention to safeguard free market competition, U.S. authorities, both in word and deed, have set a higher threshold for qualifying refusals to deal as anticompetitive and casted doubts on the effectiveness of judicial intervention in RTD cases.[3]

This article aims to shed light on the debate on the ‘unilateral refusal to deal’ doctrine by critically review the respective legal approaches to the doctrine taken under U.S. and EU antitrust law, the relevant case law and the underlying principles. By evaluating the similarities and differences between both approaches and their respective legal and economic effects, it tries to assess which system provides a more satisfactory framework. In the end, it will consider whether recent technological developments provide a chance for a converging path.

B. Unilateral Refusals to Deal in U.S. and EU Law

RTD is neither legal nor illegal under EU or U.S. antitrust law per se. Both jurisdictions developed own standards to distinguish competition on the merits from abusive conduct, which will be elucidated in the following.

I.Single-Firm Refusals to Deal under U.S. Antitrust Law

The epicenter of the conflict between freedom to deal and antitrust enforcement in the U.S. antitrust treatment of RTD cases is monopolization.

1. Section 2 of the Sherman Act

Section 2 of the 1890 adopted Sherman Act[4] prohibits monopolization, attempted monopolization and the conspiracy to monopolize by single-firms, i.e. acts that change or entrench the structure of the market in a way that harms free competition.[5] Monopoly power can be seen in a businesses’ ability to raise prices without losing profit or business to competitors.[6] As both the offenses of monopolization and attempted monopolization do not outlaw the existence of monopolies itself but the act of “monopolization”, Section 2 of the Sherman Act does not condemn pure ‘size’, but rather certain types of conduct and intention to create or threaten to create monopolies.[7] As a result, RTDs can only be seen as anticompetitive actions under Section 2 of the Sherman Act if they target to acquire monopoly power.

2. Refusals to Deal under Section 2 of the Sherman Act

The criteria to classify RTDs as monopolization or attempted monopolization evolved through the decisions made by the Federal and Supreme Court’s over time.

a) Before Trinko

The evolution of RTD jurisdiction has started more than a century ago, when the Supreme Court first addressed the issue in United States v. Colgate & Co. In its decision, the Court acknowledged that “the [Sherman] act does not restrict the long recognized right of trader or manufacturer […] freely to exercise his own independent discretion as to parties with whom he will deal”.[8] Nonetheless, RTD can create antitrust violations, when the company refusing has a “purpose to create or maintain a monopoly”.[9] In other words, a violation of Section 2 of the Sherman Act can be found where there has been clear intent or purpose to seize or exert monopoly power to control price or exclude competition by refusing to deal.[10] The Colgate ‘intent dictum’ became the benchmark in subsequent decisions, under which the Supreme Court upheld allegations of monopolization where undertakings refused to deal with the purpose or intent of monopolization.[11]

In Aspen Skiing v Aspen Highlands Skiing Corp.[12], the Supreme Court applied the Colgate ‘intent test’ as well. Here, the Court ruled that a monopolist’s refusal to continue in an earlier voluntarily entered commercial marketing arrangement with its only competitor on the market shows clear unlawful monopolization intent, unless there are “valid business reasons for the refusal”.[13] As the agreement between the two competitors was mutually “presumably profitable”, the only valid reason for its discontinuation could only be viewed in the intent to sacrifice short-term profits to achieve a long-term anticompetitive benefit on the market. The refusal was thereby unlawful.[14] The Aspen ‘profit-sacrifice indicator’ formed the law’s dominant paradigm for addressing single-firm RTDs for half a century.[15]

Further Federal Court’s decisions implemented a duty to supply regarding facilities and thereby widened antitrust enforcement even to once lawfully acquired monopolies, the so-called ‘essential facility doctrine’, and increasingly identified essential facilities as an instance of a duty to deal.[16] The doctrine states that Section 2 of the Sherman Act is violated when a firm, that uses or controls a facility or infrastructure on the primary market that is crucial (or essential) for its rivals’ activity in a secondary market, refuses to supply this facility so that its rivals cannot compete on the downstream market – so called ‘bottleneck monopoly’.[17] While the Supreme Court never recognized the doctrine terminologically,[18] still all signs pointed towards a stricter antitrust control of RTDs by U.S. authorities.

b) Supreme Court’s Decision in Trinko

The 2004 Supreme Courts’ decision in Verizon Communications Inc. v Law Offices of Curtis V. Trinko[19] stands for a retreat from this development as it outlines the Supreme Court’s reluctance to impose positive obligations on parties to deal with their competitors. Till today, it represents the leading case on RTDs under Section 2 of the Sherman Act and thereby provides considerable guidance on the recent contours of U.S. law on the doctrine.[20]

Justice Scalia delivered the majority opinion of the Court. First, he still recognized that RTDs raise antitrust concerns under certain circumstances. Referring to both Colgate and Aspen, he explicitly outlined that a refusal to deal, for violating Section 2 of the Sherman Act, must both be performed by an undertaking having monopoly power and show clear abusive conduct.[21]

The Court then emphasized the significance of monopolies for the U.S. economic system, following its explicit wording, as “the mere possession of monopoly power […] is not only not unlawful; it is an important element of the free-market-system. The opportunity to charge monopoly prices […] is what attracts business acumen […]; it induces risk taking that produces innovation and economic growth”.[22] A company that provides better services or products for its customers can acquire a monopolistic posture, with beneficial effects on the dynamic efficiency – the innovative strength of a market[23] – as it encourages competitors to develop alternative resources. Therefore, an overly excessive interpretation of the duty to deal under Section 2 of the Sherman Act in antitrust law could lessen the incentives for the monopolist, the rival or both to invest in economically beneficial facilities.[24]

The Supreme Court opined that judicial compulsion of undertakings to deal would require the Courts to set the terms of the compulsory agreement, thereby acting as “central planners”, a task to which they seem ill-suited.[25] Since anticompetitive conduct is difficult to identify and the virtue of forced sharing is uncertain in most RTD cases, the judicial system must act cautiously when assessing competing actions.[26] An excessive antitrust enforcement could produce “mistaken interferences” of robust competition, and thus “chill the very conduct the antitrust laws are designed to protect”.[27] Where actions are not clearly beneficial to competition, interventions should be rejected.

Due to the above-mentioned difficulties in identifying anticompetitive RTDs and the importance of monopolies for a dynamic market, in the eyes of the Court, monopoly power is not unlawful unless exceptional circumstances are clearly met. The Court identified IMS as the leading case for such exceptions, although it localized it “at or near the outer boundary” of Section 2 liability.[28] The Court then compared the respective facts of the cases and concluded that Aspen’s limited exceptions were not applicable in Trinko due to the factual disparities. In Aspen, the relationship between the parties was existing before, voluntarily entered and profitable, indicating that a termination of this relationship sacrificed short-term profits and therefore followed monopolization purposes only. Such an anticompetitive benefit could not be seen in Trinko.[29] Hence, the duty to deal at least could not “be enforced by means of an antitrust claim”.[30]

At last, the Supreme Court for the first time mentioned the essential facility doctrine and expressed its strong reservations about it noting that it never accepted the doctrine in any case before as it was only crafted by “some lower courts” and will not apply it here.[31]

With Trinko, therefore, the Supreme Court severely limited the scope of unlawful RTDs under Section 2 of the Sherman Act.[32] No competitor, once lawfully acquired a monopoly position, should be enjoined from exploiting it.[33] While it accepted the exceptional circumstances set up in Aspen to grant a duty to deal, it focused on the distinction of the facts of both cases. The Supreme Court saw the slight benefits of intervention outweighed by its costs and risks.[34] A necessity exists to evaluate the reasoning in Trinko, pursuant to which RTDs are actionable under Section 2 of the Sherman Act postTrinko.

aa) Prior Course of Dealing and Short-Term Profit-Sacrifice

While, the Court in Aspen treated the sacrifice of profit from a prior course of dealing only as an objective indicator for anticompetitive intent, in Trinko, the profit-sacrifice test was established as the distinctive tool between Trinko and Aspen. Circuit Courts, following Trinko, have required defendants to prove the cessation of prior voluntary business dealings and thereby the scarification of short-run profits as a necessary condition for a Section 2 claim based on RTDs.[35]

If a monopolist engages in an unprofitable RTD, it was assumed that the firm has taken that course of action only to increase barriers to competition, to earn greater monopoly profits in the future.[36] The indicators created in Aspen therefore became the mandatory minimum-requirement for U.S. Courts to estimate anticompetitive conducts as they shed light on the motivation of the firm.[37]

bb) Anticompetitive Intent equals Anticompetitive Effect

Following Aspen, Trinko mainly focused on the intent behind the defendants conducts as a predictor of the effect of the conducts on competition on the market. The Court set “anticompetitive bent” and “dreams of monopoly”[38] as apparently necessary conditions for application of the ‘Aspen exception’ to the general principle of freedom not to deal.[39] As the refusing party in Trinko did not terminate a previously entered voluntary deal, the Court deduced from the lack of anticompetitive intent the absence of anticompetitive effect of the conduct.

c) Interim Result

Trinko has been recognized as a landmark decision with significant implications for determining the conditions under which RTDs violate Section 2 of the Sherman Act. The profit-sacrifice standard became a necessary condition to sustain RTDs. Notwithstanding Aspen’s narrow exceptional circumstances, a wide free-refusal-to-deal principle applies. That is still the law in the U.S. today.

II. Single-Firm Refusals to Deal under EU Competition Law

Unlike their American counterparts, European institutions have expressed a greater willingness to use antitrust law to oblige firms to deal with (non)competitors. For both the European Commission and the Court of Justice (CoJ), the protection of competition remains paramount as dominant firms have a “special responsibility not to allow its conduct to impair competition on the common market”.[40]

1. Article 102 of the Treaty on the Functioning of the European Union

The responsibility to protect competition stems directly from Art. 102 of the Treaty on the Functioning of the European Union (TFEU) (initially: Art. 86 EEC, later: Art. 82 EC), which stipulates that “any abuse by […] undertakings of a dominant position within the internal market […] shall be prohibited as incompatible with the internal market in so far as it may affect trade between Member States”. In other words, Art. 102 TFEU precludes firms from taking unilateral actions, which are abusive of its dominant position in the market.[41]

Again, paralleling Section 2 of the Sherman Act, a dominant position is not in itself prohibited, but rather conduct that is abusive of it. Art. 102 lit. a-d TFEU provides non-exhaustive examples of what abusive conduct “may, in particular, consist in”. A common theme being drawn from the examples given is to ensure that dominant firms do not impair effective competition by excluding their competitors in an anti-competitive manner (exclusionary conduct).[42] Considering whether conduct has an anti-competitive effect depends on the specific circumstances of each individual case.

2. Refusals to Deal under Article 102 TFEU

a) Before Bronner and IMS Health

Against this backdrop, a dominant firm’s RTD constitutes an exclusionary abuse under Art. 102 TFEU if it excludes competitors from entering or remaining in the downstream market.[43] The CoJ[44] came to this result in its decision of Commercial Solvents, when it had to determine the application of Art. 102 TFEU on RTDs, for the first time.[45] The Court held that an undertaking in a dominant position for the production of raw material used to produce chemicals could not cease to supply the materials to a downstream customer who manufactures derivatives, simply because the dominant firm planned to supply derivatives on the downstream market on its own and wanted to already exclude future competitors.[46] As the decision concerned the termination of supplies to an existing customer, it was uncertain whether, and to what extent the principles set up in Commercial Solvents could be broadened to “non-termination” cases.[47]

The Court went further in United Brands[48], where it held that even in the case of a refusal to supply to a distributor in the upstream market, a duty to deal could arise under certain conditions. A duty to deal was issued as the dominant manufacturer could “not stop supplying its long standing customer who abides by regular commercial practice, if the orders placed by that customer are in no way out of the ordinary”.[49] Even though both parties were not competitors on the downstream market, the refusal still had anticompetitive effects, as it was done with the intention to induce the distributors to buy exclusively from the refusing manufacturer.[50] A duty to deal did not require the indications of a price increase in the downstream market or other direct consumer harm, as long as it could not be justified.[51]

EU competition authorities applied the ‘essential facility doctrine’ in a similar direction as taken by the U.S. Federal Courts before.[52] While the doctrine originated in U.S. antitrust law but has remained highly contentious there, it has influenced (especially) the Commission’s decisions in a number of cases. The Commission initially applied the doctrine to port and other transport infrastructures,[53] but subsequently expanded the doctrine’s application to several other facilities. This wide approach placed greater responsibility on the owners of such facilities and a duty to supply was more likely to appear to new as well as existing customers, sometimes even if they could compete without access to the facility.[54]

b) ECJ’s Decision in Bronner

The ECJ’s watershed decision in Bronner[55] in 1998 limited the EU approach to RTDs. The Court was considering, in a preliminary ruling, whether a dominant undertaking could be required to supply new customers seeking access to its products or services for the first time. It found that the obligation for dominant firms to grant access to their facilities is limited to a narrow set of ‘exceptional circumstances’.[56] There could only be an abuse of a dominant position if the product or service was “indispensable to carrying on the rival’s business” and thus RTDs could “eliminate all competition on the part of” the undertaking seeking access without being objectively justified.[57]

The ECJ saw the indispensability requirement’ only met in the case that it was not economically viable for a rival to create an own facility capable of gaining a market power comparable to the consisting one.[58] The refusal to access the existing scheme, therefore, must constitute a barrier to entry such that it could eliminate all competition in the relevant market.[59] The ECJ had to consider if there were alternative solutions to the current structure and if so, whether technical, legal or economic obstacles made it impossible or unreasonably difficult for the competitors to create these alternative solutions – regardless of whether those were less advantageous.[60] The Court emphasized that it is not enough for the refused party to argue that the establishment of its own facility was simply too expensive for it due to the small size of his business.[61] Rather, it is necessary “at the very least” to establish that it was not economically viable for a competitor of ‘equal size’ to create its own product or service.[62] In other words, indispensability could only be assumed, if the creation of an alternative solution was impossible even for a hypothetically similar sized competitor of the dominant undertaking.

While the ECJ seemed to apply the ideas of the U.S. ‘essential facility doctrine’, it limited the scope of its application as facilities were not lightly be termed ‘indispensable’. Whereas the Commission assumed the essential facility doctrine to be satisfied in the present case, the ECJ did not pursue this wide approach.[63]

c) ECJ’s Decision in IMS Health

In the postBronner decision of IMS Health[64] in 2004, the ECJ reinforced the Bronnerindispensability test’ and transformed it into a substantive test for the question of refusals to license intellectual property rights (IPR).

After confirming that a refusal to license IPR can fall within the exceptional circumstances of the RTD doctrine, the Court set out three conditions to be satisfied so that a refusal to license ‘indispensable’ IPRs constitutes an abuse. Those circumstances require (1) the prevention of the emergence of a new product for which there is a potential consumer demand, (2) that it is unjustified and (3) such as to exclude any competition on a secondary market.[65] The Court explicitly clarified that it reads the criteria cumulatively.[66] In addition to the Bronner conditions, which emphasized the importance of freedom of contract for the market, the Court preserved IPRs in the sense that no compulsory license can be imposed if the competitor does not propose to market a new product to benefit consumer welfare. From setting up the conditions, the indispensability and new product requirement are such that everything will turn on the application of these conditions to the facts of the case. For the definition of the ‘indispensability’ requirement, the Court relied on its earlier decisions in Bronner.[67]

The ECJ underlined the significance of IPRs for the competitiveness of many industrial sectors. It made clear that a plain obliged license would only harm the authors’ incentive to invest and innovate without creating sufficient benefits for the market.[68] While the Court generally emphasized that a refusal to license IP can be subject to EU competition law, the expansion of exceptional circumstances to IP cases intended to limit a duty to license to situations in which the consumer’ benefits of licensing are large, while the negative effects on the incentives to innovate are small. This should prevent an over-extensive application of Art. 102 TFEU towards IP licensing.[69]

The IMS Health condition of a ‘new product’ was highly criticized by scholars[70] and got watered down three years later in the Court of First Instance’s Microsoft[71] decision. The Court in Microsoft took the view that the new product requirement would be satisfied as long as the adjustment of a slightly modified product somehow benefits consumer demands and is not simply duplicative.[72] This requirement thereby mostly lost its importance as a necessary precondition in refusal to license cases. The main elements of the test will therefore remain the exclusion of any competition due to the facilities’ indispensability without any justification.

d) Interim Result

EU competition law recognizes a potential duty to deal in both refusal to supply and refusal to license IP cases. Still, Bronner and IMS Health, significantly narrowed the approach by requiring special circumstances to be fulfilled cumulatively to grant a duty to deal. Only were RTDs concern an indispensable facility, owned by a dominant firm, which is likely to exclude all competition, a duty to deal might occur. Potential business alternatives – even less advantageous ones – preclude any antitrust claim by a competitor, as long as they are possible and not unreasonably difficult to achieve.[73]

C.Comparison of the EU and U.S. Approaches

As shown, both the U.S. and EU approach respect the freedom of contract regarding RTDs but differ on the scope of use of antitrust enforcement to protect competition. The different requirements produce disparities in the respective outcomes, to put it in the words of some scholars, as far as “[c]oncepts which appear now to be mythical beasts in s.2 jurisprudence – price squeezes, essential facilities, refusals to deal – are alive and well and living in Europe”.[74]

I. The Main Differences between EU and U.S. Antitrust Enforcement on RTD Cases

Comparing both wordings, Art. 102 TFEU and Section 2 of the Sherman Act seem to correlate with each other. However, upon further assessment, the antitrust law’s approaches to RTDs vary and contain significant differences.

Both Art. 102 TFEU and Section 2 of the Sherman Act require defining the relevant market and a firm’s dominant position in it. The EU’s threshold for establishing market dominance is generally lower than the U.S. benchmark. In U.S. antitrust law, sometimes even if a firm has a market share of 100%, it can be treated as insufficient where there is ease of entry, as further factors, other than a calculation based purely upon market share, must also be taken into account.[75] By contrast, the EU commission prescribes, that a firm’s attainment of 50% market share – under certain conditions already approximately 40%[76] – is strong evidence that a firm has a dominant position.[77] Other factors such as the likelihood of expansions play a (mostly) subordinate role.[78]

Art. 102 TFEU parallels Section 2 of the Sherman Act by requiring anticompetitive effects to distinguish the proportionate use of a lawfully acquired dominant position from an abusive conduct.[79] While both jurisdictions prohibit so-called ‘exclusionary conducts’ as anticompetitive practices, the threshold for what sort of conduct qualifies as such is the crucial point of distinction between the two approaches. Some scholars even see a “exclusionary conduct ‘definition war’” among the positions.[80]

EU competition law aims to protect consumer welfare and competition by the establishment of a trans-European fair and equal market.[81] Art.102 TFEU as part of the EU Treaties follows a long run socioeconomic movement that promotes Member State integration to create an “ever closer union” and a level playing field through economic ties between all States.[82] Consumers are protected indirectly by protecting the uniform market structure, maintaining an open market and preserving alternative products to the dominant firm’s facility.[83] Thus, a conduct is already seen as exclusionary, if it somehow creates obstacles affecting the open access of other firms to the market or harm a level playing field in the EU.[84] Anticompetitive intent by the dominant firm is not required as the focus is on the objective effect of the firm’s action. Following the decisions in Bronner and IMS Health, the spotlight is on the indispensability of a facility by which a refusal to gain access to a service or product prevents rivals from competing by means other than “normal competition” or “competition on the merits”. If the refusal excludes all competition on the market for the product, it is assumed that the conduct indirectly harms consumers too and a duty to deal is likely to be granted. Thus, the termination of a former voluntary agreement is not a requirement under EU law and a distinction between the RTD with non-competitors or competitors is rather formal as it has no real impact on the outcome of the case.

U.S. antitrust law acknowledges a duty to deal only when a firm with monopoly power makes an unjustified change in a course of dealing which it voluntarily entered before, to the detriment of the market and its consumers. The mere fact that a conduct by a dominant firm excludes or injures a competitor and thereby reduces competition does not create unlawfulness. Even aggressive competition strategies are welcomed as long as they benefit consumer welfare.[85] Therefore, U.S. antitrust law does not condemn conduct as exclusionary unless it violates the standard of appropriate behavior. In the cases of Aspen and Trinko, the Supreme Court heavily relied on the profit-sacrifice-test to define this standard. In circumstances of doubt about whether an RTD is an anticompetitive conduct or a competitive merit, the legal assumption of a lawful competitive conduct will mostly prevail.[86] RTDs, thus, are not illegal, even in situations where they are arbitrary or have no justification.[87]

Due to the legally obtained exclusive position of IP holders, a fortiori, there is even greater reluctance to impose an obligation to license IPRs in the U.S. [88] The Court of Appeals for the Federal Circuit categorically rejects a duty to license IP rights in cases they were legally obtained and not used beyond the rights statutorily granted.[89] This unqualified protection of IPRs, characterized as triggers for innovation, precludes a priori to consider a refusal to license IP an antitrust violation by itself.

To conclude, a dominant manufacturer has significantly less room to refuse to deal with rivals or customers in the EU than similarly situated firms have in the U.S. The threshold for a firm’s position being dominant is mostly higher under Section 2 of the Sherman Act than under Art. 102 TFEU. U.S. case law raised the bar by requiring qualified anticompetitive effects of profit-sacrifice or clear damage on consumer welfare. Under EU competition law, pure market power can suffice to impose a duty to deal as long as it excludes fair competition on the market.

II. Discussing both Approaches

Having portrayed the differences between the approaches, an analysis can be made as to which system provides more certainty and effectiveness and should therefore be given priority to.

1. (Economic) Effects of Mandatory Market Access

Whilst conducting such a comparison, one must keep in mind that the aim of antitrust law is to protect markets from anticompetitive behavior.[90] Thus, the effects of duties to deal on undertaking’s competitive behavior are necessary to be enlightened. The mandatory access debate is framed around a trade-off between ex ante and ex post effects of judicial interventions.[91]

On one hand, duties to deal can reduce firm’s ex ante incentives to innovate and invest on the market.[92] While both the ECJ and the Supreme Court have taken into account its detrimental effect, U.S. Courts especially emphasized the alleged positive effects of monopolies for economic welfare. Monopoly power is viewed as a catalyst for competitive processes in a free market and as an important driving force to induce risk taking by competitors.[93] As Judge Hand stated, “[t]he successful competitor, having been urged to compete, must not be turned upon when he wins.”[94] By creating the legal ability to exclude others from using the invention and charge monopoly prices, successful inventors are rewarded and competitors are driven to follow their example. Thus, the market tends toward self-correction without a need of excessive regulative intervention (the so-called “Chicago School” approach).[95] Duties to deal could facilitate entry of inefficient producers not required to develop varied products to enter the market leading to a decline of innovation.[96]

On the other hand, duties to deal can also have the ex post effect of an increase of competition and consumer welfare on the downstream market if it enables competitors to gain access to an otherwise inaccessible market and facilitates the development of new products.[97] EU competition aims to protect the individual rights of (especially smaller) competitors against harms which are not part of the normal “competition on the merits” but stem from exclusionary conducts that impede fair competition.[98] Fair access to the market is seen as a guarantee for a functioning competition resulting in consumer welfare in the long run.[99] This effect is especially observed in circumstances whereby a product is indispensable to carry on business on that market in any manner so that a duty to deal is necessary to create competition in the first instance.

Taking both effects into account, monopolists should be allowed to compete on the merits as this is what generally drives innovation and lowers consumer prices. Forced sharing of facilities diminishes the incentive to create alternatives and thereby chills innovation. However, innovation is not harmed by compulsory supply where no alternatives exist or are possible to be created. In markets without reasonable alternatives to the provided facility – which is exactly the requirement for indispensability of the facility in EU law – no undertaking could otherwise compete with the monopolist.[100] Instead, it is, after all, not effective if every competitor would be required to build up its own telephone or port infrastructure or other indispensable facility to gain access to the market. Sitting monopolists, that do not fear competition, do not feel innovative pressure.[101] A likelihood arises for dominant firms to focus more on the oppression of potential competitors than on consumer needs.[102] There is no guarantee, that excessive revenues of dominant firms would be reinvested in innovation. A duty to deal under those conditions is not, as in Judge Hand’s understanding, a dominant firm’s punishment for its success but for its abusive restriction of competition. A too restrictive approach by a sole faith in the market’s self-regulation leaves no space to adequately address and react to negative market influences by monopolies.[103]

An effective protection of competitors against exclusionary conducts much more guarantees the individual rights of every undertaking and increases the incentives of smaller market players and potential newcomers to invest and compete as well.[104] Consumers can benefit from divergent competition on the market as well as more market players can fosters competition and drive prices down. Thus, the argument that a duty to deal undermines the incentives to innovate and invest is often made too readily.

2. Costs of Legal Errors in Refusal to Deal Cases

As it is mostly difficult to balance with certainty those beneficial and adverse effects of antitrust intervention, mistakes are likely to arise and incorrect justifications may be taken.[105] Two kinds of misjudgment must be considered hereby. EU law accepts more the risk of false positives (wrongful judicial intervention), but pays highly attention to false negatives (wrongful judicial nonintervention) and its costs. U.S. antitrust law conversely is mostly concerned with the risk of false positives and its harm on competitors’ free actions. According to Forrester, “[t]he [EU] Commission attributes comparatively lower weight to a dominant player’s freedom to run its own business, and comparatively more weight to the protection of competitors than U.S. courts”.[106] It seems that to balance between U.S.’s and EU’s approach, the question that is required to be considered is what harms free competition more. The exclusion of smaller competitors and the harm of consumer welfare resulting from supra-competitive prices due to a lack of judicial intervention, or the chilling effects on incentives for investment and innovation by overzealous antitrust enforcement.

The answer cannot lie in a simple “Better” or “Worse” to the respective false intervention. From an error cost perspective, an appropriate equilibrium between both errors can be tailored to the specific case and is superior to one focusing solely on one of the errors. Aggressive competition is, without doubt, damaged by false positives. And aggressive competition can benefit innovative dynamism and consumer welfare. Advocate-General Jacobs in Bronner acknowledged this risk.[107] He noted that the necessity to ensure that procompetitive strategies are not wrongly thwarted and that competitors are not protected to the detriment of competition.[108] Thus, Bronner and IMS Health effectively narrowed compulsory sharing to “careful justification” to limit the costs of expected errors and require impossibility of any alternative solution to the indispensable facility so that a duty to deal is most likely to cause welfare increase.[109] An accused over-excessive antitrust enforcement under Art.102 TFEU, thus, must be objected. The U.S. approach, while effectively excluding false positive, is cautious in identifying and remedying exclusionary conduct.[110] It leaves a broad space for judicial nonintervention in cases exclusionary conducts are taken but not identified as such and thus does not effectively contain false negative.

3.The Exceptional Circumstances Test – The Legal Standard in Both Jurisdictions

On both sides of the Atlantic, the Courts condemn RTDs as antitrust violations only in “exceptional circumstances”. Both impair high, albeit different hurdles. The high level of consciousness by U.S. Courts of the possibility to inappropriately override property owner’s rights has led to an RTD test postTrinko, which seems too narrow to guarantee an adequate protection of competition.

Trinko used the discontinuance of the prior agreement as a principle to distinguish its facts from Aspen on the basis that an RTD case can succeed only, if a defendant reneged from an earlier voluntary deal, not minding sacrifice short-term profits for later anti-competitive gain. Such a requirement discourages monopolists to deal with competitors in the first place and thereby supports imbalance in the market.[111] It is contradictory if a system that on the one side tries to promote monopolists to deal with competitors and third parties for a competitive market, on the other side makes discontinuance from a dealing a necessary ingredient for antitrust violation and thereby induces monopolists not to enter agreements as a precautionary measure against antitrust actions.[112] While the evidence of profit-sacrifice may support a finding of exclusionary conduct, not all monopoly firm’s exclusionary strategies require a profit-sacrifice.[113] Trinko itself illustrates the limitations of the profit-sacrifice requirement as a helpful predictor of anticompetitive intent.[114] While in Trinko, the dominant firm’s refusal was facially anticompetitive and harmed consumers. Still, the Court refused to facilitate competition as the firm never voluntarily shared its facilities before and so terminated no previously entered agreement.

Following those disadvantages in the U.S. approach, the exceptional circumstances established in Bronner and IMS Health, in many respects, allow for a more appropriate and nuanced judicial intervention. The indispensability and objective justification requirements do not involve a termination of a former voluntary agreement or the sacrifice of profit, thus, allow a more flexible application of the law. The ECJ does not entirely exclude IP from antitrust control and allows judicial intervention where competition and consumers benefit from a duty to license that otherwise excludes all competition. Even if one were not to share the application of antitrust law on IP licensing, at all events competitors’ access to indispensable IPs would make economic sense to at least guarantee the effective allocation of resources and reasonable pricing from which consumers can benefit[115] – by antitrust enforcement or other means.

III. Conclusion

The different approaches to the RTD doctrine are highly debated in the transnational antitrust community. This is not at least attributable to the fact that the effect of a duty to deal has the potential to both enhance and impede competition in the relevant market.

There is consensus between both approaches that competition law should not intervene where the market can be expected to self-correct exploitative practices. Judicial intervention should only be taken where fair and effective competition is no longer ensured due to RTDs.[116] The little room left in U.S. Supreme Court and lower court caselaw post-Trinko seems too small and uncertain to effectively protect competition and consumer welfare where it is needed. There is a certain risk that, if Trinko is understood the way outlined in this article, incumbent monopolists will be insulated from anticompetitive abuses that have a real possibility of inflicting antitrust law. Therefore, EU competition law proves to be more balanced and thus favorable in its exceptional circumstances. It does concern both false negative and false positives and creates manageable criteria to impose proportionate duties to deal whenever competition in a market depends on having access to a specific facility.

  1. “Big Tech” – Creating a more Converging Path

The latest technological developments in high-tech industries might create an opportunity for U.S. antitrust law to join the EU in effectively combatting abusive RTDs. The so-called “Big Tech” companies, including Google, Facebook, Amazon, Microsoft and Apple – to mention only the most popular – have built worldwide dominant market structures and, to accelerate growth, tend to acquire actual or potential competitors (i.e. startups) or oust competitors from using their platforms in order to preserve their dominant positions.[117] Due to the immense investment requirements, network effects and the big data necessity to develop accurate algorithms, access to their platforms and online facilities seems indispensable for undertakings and a refusal to grant access can significantly harm competition on the digital markets.[118] Access to what once was regarded as being essential, railways, harbors etc., nowadays turned into a necessary access to digital platforms, the essential facilities of the 21st century. Gaining monopoly positions in these digital markets was made, also and especially, possible by the laissez-faire antitrust enforcement by U.S. authorities.

Though, the high reluctance of judicial intervention in RTD cases postTrinko is observed by some U.S. authorities as backfiring. In 2009, the U.S. Department of Justice (DoJ) changed its policy due to the developments by withdrawing its Section 2 Report. The report favored extreme caution and granted high obstacles for antitrust enforcement to apply to RTD cases. Assistant Attorney General Varney stated that it would no longer be sufficient to “rely upon the marketplace[‘s self-correction] alone to ensure that competition and consumers will be protected”.[119] The DoJ stated that it is necessary to preserve an “open and level playing field” to foster innovation through competition, which needs to be guaranteed through antitrust enforcement.[120] Whether antitrust enforcement in RTD cases should therefore also be expanded to IP law remains uncertain. However, a tendency towards a renunciation of the common legal practice of U.S. antitrust authorities can be seen.

In 2019, the Federal Trade Commission (FTC) reached a $5 billion settlement with Facebook caused by concerns not only about antitrust issues but concerns about the abuse of market power regarding customers data privacy.[121] Only about a year later, in October 2020, the DoJ filed a lawsuit against Google, a company against which the European Commission has launched three separate antitrust investigations and charged fines totalling about €8 billion since 2010, creating the agency’s first major antitrust case against Big Tech since 1998.[122] Due to these latest developments, whilst there is still no clear step-back of the U.S. “Chicago School-shaped” Courts from their reluctant antitrust enforcement, there is room for optimism that EU and U.S. antitrust law are on converging paths.


*Lennart Levita studiert Rechtswissenschaften an der Albert-Ludwigs-Universität Freiburg und ist studentische Hilfskraft bei Herrn Prof. Dr. Boris P. Paal M. Jur. (Oxford)
am Institut für Medien- und Informationsrecht, Abt. I:
Zivil- und Wirtschaftsrecht, Medien- und Informationsrecht. Der Artikel basiert auf seiner Seminararbeit im Seminar zum Transnational Antitrust Law bei Prof. Dr. Alexander Riesenkampff, Prof. Laurence T. Sorkin und Dr. Sebastian Steinbarth, LL.M.


[1] Areeda, 58 Antitrust (1989) 841, 852f.

[2] Schoen, 80 NYU LRev (2005) 1625, 1638.

[3] Cf. 540 US 398 (2004).

[4] 15 U.S.C. § 2 (2004).

[5] Schweitzer in Ehlermann/ Marquis (Eds.), Europ. Comp. L.A. (2007) at 6.

[6] Meadows, 25 Fordham Intell. Prop. Media & Ent. L.J. (2015) 795, 806.

[7] Raybould/Firth, Law of Monopolies (1991) at 99; Barnes/Dworkin/Richards, Law for Business, 7th Ed. (2000) at 874;

U.S. v U.S. Steel Corp., 251 US 417, 451 (1920).

[8] U.S. v Colgate & Co., 250 U.S. 300, 307 (1919).

[9] Ibid.

[10] Gellhorn/Kovacic/Calkins, Antitrust Law and Economics (2004), at 140.

[11] Cf. Eastman Kodak Co. of New York v Southern Photo Materials Co. 273 U.S. 359 (1927); Loraine Journal Co. v U.S. 342 U.S. 143 (1951).

[12] Aspen Skiing v Aspen Highlands Skiing Corp. 472 U.S. 585 (1985).

[13] Ibid. at 597.

[14] Ibid. at 608.

[15] Fox, 73 Antitrust L.J. (2005) 153, 155.

[16] Cf. Hecht v Pro-Football, Inc. 570 F.2d 982 (1977).

[17] Georges/Bay, 17 IP & Tech L.J. No. 12 (2005) at 1.

[18] Troy, Columb. LRev (1983), 441; Temple Lang, 18 Fordham Int’l L.J. (1994) 437, 446.

[19] 540 U.S. 398 (2004).

[20] Katz/Veel, 79 Antitrust L.J. (2013) 139,144.

[21] 540 U.S. 398, 407 (2004); quoting 384 U.S. 563, 570f (1966).

[22] 540 U.S. 398, 407 (2004).

[23] Candeub, 66 UPitt LRev (2005) 821, 835f.

[24] 540 U.S. 398, 408 (2004).

[25] Ibid.

[26] Ibid.

[27] Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 226 (1993).

[28] 540 U.S. 398, 408, 409 (2004).

[29] Ibid. at 399.

[30] Ibid. at 406.

[31] Ibid. at 410.

[32] Hovenkamp, UPenn Legal Scholarship Repository (2008) at 31.

[33] Nagy 32 Europ. LRev (2007) 664, 666.

[34] Jones, Handbook (2006) 236, 256.

[35] See cf. Covad Commc’ns Co. v Bell Atl., 398 F.3d 666, 673 (D.C. Cir. 2005).

[36] Lao,73 Antitrust L.J. (2005) 171, 186.

[37] See post-Trinko Am. Cent. E. Tex. Gas Co. v. Union Pac. Res. Group Inc., 93 F. App’x 1, 9 (5th Cir. 2004).

[38] 540 U.S. 398, 409 (2004).

[39] Fox, 73 Antitrust L.J. (2005) 153, 164.

[40] C-322/81, NV Nederlandsche Banden Industrie Michelin v Commission of the European Communities, 1983 E.C.R. 03461.

[41] Wish/Bailey, Competition Law, 8th Ed. (2015) at 170.

[42] EC Guidance 2009/C 45/02, para.19.

[43] Temple Lang, 18 Fordham Int’l L.J. (1994) 439, 445.

[44] Later European Court of Justice (ECJ) – therefore, it will be referred to as ECJ.

[45] Case 6 & 7/73, Instituto Chemioterapico Italiano SpA & Commercial Solvents Corp. v Commission, 1974 E.C.R. 225.

[46] Ibid. at 256.

[47] Leistner, 3 ZWeR (2005) 138, 140.

[48] C-27/76, United Brands Company and United Brands Continentaal BV v Commission, 1978 E.C.R. 207.

[49] Ibid. at 292.

[50] Temple Lang, 18 Fordham Int’l L.J. (1994) 439, 476.

[51] C-27/76, 1978 E.C.R. 207 at 196.

[52] Hedberg (1999) at 9.

[53] Cf. Decision 94/19/EC relating to a proceeding pursuant to Article 86 of the EC Treaty (IV/34.689 – Sea Containers v. Stena Sealink).

[54] Decision 2002/165 relating to a proceeding pursuant to Article 82 of the EC Treaty (COMP D3/38.044 — NDC Health/IMS Health) para. 63.

[55] C- 7/97, Oskar Bronner GmbH & Co. KG v Mediaprint Zeitungs- und Zeitschriftenverlag GmbH & Co. KG, 1998 E.C.R. I-7791.

[56] Leistner, 3 ZWeR (2005) 138, 144.

[57] C- 7/97, 1998 E.C.R. I-7791, para. 38, 41; The ECJ set up the requirement already in earlier decisions, see cf. C-311/84, CBEM-Télémarketing v Compagnie Luxembourgeoise de Télédiffusion and Information Publicité Benelux, 1985 E.C.R. 3261.

[58] C- 7/97, 1998 E.C.R. I-7791, para. 46.

[59] Van Bael/Bellis, 5th Ed. (2009) at 843; C- 7/97, 1998 E.C.R. I-7791, para. 66.

[60] Leistner, 3 ZWeR (2005) 138, 144; C- 7/97, 1998 E.C.R. I-7791, para. 44.

[61] Ibid. para. 45.

[62] Ibid. para. 46.

[63] Leistner, 3 ZWeR (2005) 138, 149.

[64] C-418/01, IMS Health v NDC Health, 2004 E.C.R. 1-05039.

[65] Ibid. para. 38.

[66] Ibid.

[67] Ibid. para. 40.

[68] Derclaye, 29 Europ. LRev (2004) 687, 695.

[69] Ahlborn/Evans/Padilla, 28 Fordham Int’l L. J. (2005) 1109, 1110; Leistner, 3 ZWeR (2005) 138, 153.

[70] Cf. Conde Gallego, Research Handbook (2008) 215, 559, stating: “requiring the […] undertaking to intend to offer a new product will make no sense”.

[71] T-201/04 Microsoft Corp v Commission of the European Communities, 2007 E.C.R. 11-03601.

[72] Ibid. para 647.

[73] Leistner, 3 ZWeR (2005) 138, 154.

[74] Duncan/Coleman/Daniel/Haleen, 2 GCLitRev (2009) 148, 158.

[75] U.S. v Syufy 903 F.2d 659 (9th Cir. 1990).

[76] T-219/99 British Airways v Commission, 2003 II E.C.R. 5917.

[77] C-62/86 AKZO Chemie BV v Commission, 1991 E.C.R. I-3359.

[78] EU Commission Guidance on Art. 82 EC, Official Journal of the European Union, 2009/C 45/02, at 1, 3.

[79] Leistner, 3 ZWeR (2005) 138, 156.

[80] Gavil, 72 Antitrust L.J. (2005) 3, 5.

[81] Ehlermann, 29 CM LRev (1992) 257, 261ff; Fox, 59 Antitrust Bulletin (2014) 129, 132.

[82] Ohlhausen, Remarks at the Georgetown Global Antitrust Symposium, 19.09.2016, at 9.

[83] T-201/04 Microsoft v Commission, 2007 E.C.R. II-03601; C-95/04 P British Airways v Commission, 2007 E.C.R. I-2331.

[84] See T-219/99 British Airways v Commission, 2003 E.C.R. II-5917.

[85] U.S. DoJ, Sec. 2 Report (2008) at 14.

[86] 540 U.S. 398, 414 (2004).

[87] Nagy, 32 EL Rev (2007) 664, 672.

[88] Katz/Veel, 79 Antitrust L.J. (2013) 139, 144.

[89] Cf. In re Independent Serv. Org. Antitrust Litig., 203 F.3d 1322 (Fed. Cir. 2000).

[90] Fox, 26 World Comp. (2003) 149.

[91] Geradin, 41 CM LRev (2004) 1519, 1538.

[92] Ahlborn/Evans/Padilla, 28 Fordham Int’l L.J. (2004) 1109, 1137.

[93] 540 U.S. 398, 407 (2004).

[94] Hand J., U.S. v Aluminium Co. of America 148 F.2d (2d Cir. 1945) 416, at 430.

[95] Schweitzer in Ehlermann/ Marquis (Eds.), Europ. Comp. L.A. (2007) at 28; Jacobs, 74 NC LRev (1995) 219, 222.

[96] Ahlborn/Evans/Padilla, 28 Fordham Int’l L.J. (2004) 1109, 1138.

[97] Ibid. at 1137.

[98] Schweitzer in Ehlermann/ Marquis (Eds.), Europ. Comp. L.A. (2007) at 41.

[99] Fox, 26 World Competition (2003) 149, 155f.

[100] Candeub, 66 UPitt LRev (2005) 821, 836.

[101] Meadows, 25 Fordham Intell. Prop. Media & Ent. L.J. (2015) 795, 808; Conde Gallego, Research Handbook (2008) 215, 220.

[102] Drexl, IIC (2004) 788, 796.

[103] Apel (2015) at 516.

[104] Schweitzer in Ehlermann/ Marquis (Eds.), Europ. Comp. L.A. (2007) at 41.

[105] Ahlborn/Evans/Padilla, 28 Fordham Int’l L.J. (2004) 1109, 1111.

[106] Forrester, 28 Fordham Int’l L. J. (2005) 919, 920.

[107] Opinion of Advocate-General Jacobs in Bronner, E.C.R. I-7791, I-7805-06 (1998) at 44ff.

[108] Jones, Handbook (2006) 236, 241.

[109] Ahlborn/Evans/Padilla, 28 Fordham Int’l L.J. (2004) 1109, 1144.

[110] Jones, Handbook (2006) 236, 264.

[111] Fox, 73 Antitrust L.J. (2005) 154, 160f.

[112] Drexl, IIC (2004) 788, 799.

[113] Lao, 73 Antitrust L.J. (2005) 171, 188.

[114] Ibid. at 189.

[115] Drexl, IIC (2004) 788, 805.

[116] O’Donoghue/Padilla, 2nd Ed. (2013), at 637.

[117] Wu/Thompson, The Roots of Big Tech Run Disturbingly Deep, The New York Times, June 7, 2019, (accessed Dez 29, 2020).

[118] Freeman/Sykes, Antitrust and “Big Tech” (2019) at 21.

[119] Press Release of the DoJ of May 11, 2009: “Justice Department withdraws report on Antitrust Monopoly Law”.

[120] Pozen, Antitrust Division DoJ (2012) at 14.

[121] Press Release of the DoJ of Jul 24, 2019: “Facebook Agrees to Pay $5 Billion and Implement Robust New Protections of User Information in Settlement of Data-Privacy Claims”.

[122] Press Release of the DoJ of Oct 20, 2020: ”Justice Department Sues Monopolist Google For Violating Antitrust Laws”; the FTC sued Microsoft for illegally maintaining its monopoly position in the PC market in 1998, cf. U.S. v. Microsoft Corporation, 253 F.3d 34 (D.C. Cir. 2001).