Growing Pains in EU Antitrust Enforcement

by Jonathan Hettleman 1

Antitrust law is notorious for vague doctrinal boundaries. 2 The United States Supreme Court and the European Court of Justice (ECJ) (the highest court in the European Union (EU) on matters of EU law) have been reticent to offer concrete standards that can guide lower courts, industry-members, and legal counsel. In part, this uncertainty can be attributed to the fact that antitrust statutes tend to be vague and have, therefore, left the judiciary to chart its own path towards drawing doctrinal boundaries. 3Perhaps even more importantly, much of antitrust law involves speculative predictions of the future effects of a firm’s conduct on competition in given markets. 4

A debate has emerged in both the United States and the EU over the proper method by which courts should assess whether a firm is foreclosing its competition in a given market. 5  More specifically, this broader debate is highlighted by the jurisdictions’ differing approaches to various kinds of rebate schemes and their relative likelihood to foreclose competition.

Although circuits are split on how best to resolve this issue in the United States, 6 most courts, commentators, and regulators favor an approach that seeks to determine whether, after taking account of the rebate, a product or a bundle of products is being sold for a price below the cost of their production such that the price will foreclose equally efficient competitors. 7 In the EU, while the ECJ has consistently applied a test that asks whether a firm’s conduct is hypothetically capable of foreclosing its competitors in the future (regardless of whether it is being sold below cost), 8 the European Commission for Competition has advocated for an approach that is closer to the favored approach in the United States: 9 specifically, the Commission counseled courts in the EU to apply a version of the effects-based “as efficient competitor” test. 10 While the ECJ has never adopted the approach proposed by Commission, its rulings have slowly drifted in favor of a more effects-based analysis and against the more conduct-based hypothetical approach. Most recently, for example, this shift was exemplified in the ECJ’s ruling in Post Danmark A/S v. Konkurrencerådet (hereinafter Post Danmark II). 11

This Contribution will argue that the ECJ should continue to favor an effects-based approach in its analysis of whether rebates foreclose competition: such an approach favors analyzing actual effects rather than speculation, protects efficient companies that make pro-competitive business decisions, and provides clarity.


Historically, in EU courts, once a firm was found to be dominant in a particular market, it had a heightened responsibility not to impede competition. 12 This meant that, by definition, non-dominant firms could engage in behavior that would be unlawful for dominant firms. 13 This heightened responsibility also meant that market share held by a firm (the extent of a firm’s dominance) took a backseat to the binary question of whether or not a firm was dominant.

In addition, EU courts explicitly resisted inquiry into the degree to which a market had actually been foreclosed. In Intel v. Comm’n, for example, the court explained that under EU competition law, “it is unnecessary to undertake an analysis of the actual effects of the rebates on competition.” 14 In Microsoft v. Comm’n, too, the court made clear that satisfying the anticompetitive foreclosure element for the purposes of finding an abuse of Article 102 TFEU – the source of authority for antitrust regulation in the EU – does not depend on the analysis of actual foreclosure effects. 15 Instead, for the purposes of finding a rebate to be in violation of EU competition law, it was sufficient for the court that the dominant firm’s behavior was capable of foreclosing competition.

Instead of applying an equally efficient competitor test based on whether a rebate rendered the price of a product or group of products below their cost of production, courts relied on nebulous factors to try to predict whether the pricing scheme was capable of foreclosing competition. For example, the Microsoft court looked to whether competition was being driven by price or the merits of the products or services being offered. 16 The court found that “if Microsoft had not [tied Windows Media Player to the Windows operating system] competition between RealPlayer and Windows Media Player would have been decided on the basis of the intrinsic merits of the two products.” 17 Ultimately, the court in Microsoft found that the company’s success in selling Windows Media Player did not “come about because that player is of better quality than competing players or because those media players . . . have certain defects.” 18 Instead, the court found that Microsoft’s increased success in the media player market resulted from its decision to tie Windows Media Player to Windows (a product in a market in which Microsoft was dominant). 19

Another factor that the ECJ historically focused on was whether a rebate scheme induced loyalty. 20 In a seminal decision in 1979, the ECJ held that rebates that were conditioned on a customer purchasing exclusively from a specific provider were per se violations of EU competition law. 21 The ECJ later held that “any loyalty-inducing rebate system applied by an undertaking in a dominant position has foreclosure effects prohibited by [EU competition law].” 22 But the skepticism towards loyalty-inducing rebates did not stop at those that forced exclusivity. In British Airways, for example, the ECJ held that British Airways’ rebate scheme foreclosed competition because it was “fidelity-building.” 23

In Post Danmark II, however, the ECJ applied a decidedly more effects-driven analysis to determine whether a rebate scheme foreclosed competition. 24 First, the court made clear that, in order for a rebate to be found anticompetitive, it must be “likely” to foreclose competition. 25 Second, to determine whether a rebate scheme is likely to foreclose competition, the Post Danmark II court looked to the extent of the dominance of the firm offering the rebate. 26

In addition, the Post Danmark II decision described three categories of rebate schemes, providing much-needed clarity on how to assess different types of rebates. 27 Rebates falling within the first category, “quantity rebates,” are based solely on the volume of purchases made by a customer. 28 Quantity rebates, according to the Post Danmark II court, are not per se anticompetitive. 29 Second, “loyalty rebates” are those which force customers to buy all or nearly all of their requirements from the dominant firm. 30 Rebates that fall within this second category are anticompetitive because, by definition, they prevent customers from purchasing all or most of their requirements from the dominant firm’s competitors. 31 Finally, the court recognized a “third category” of rebates that cannot simply be categorized as either quantity rebates or loyalty rebates. 32 Rebates that fall within Post Danmark II’s “third category” are not per se anticompetitive or per se pro-competitive, but instead require additional circumstantial analysis. 33When deciding whether these “third category” rebates are anticompetitive, Post Danmark II held that, when determining whether a rebate is justified, courts must consider whether the rebate creates economic efficiencies; whether the rebate restricts the customer’s freedom to choose a supplier; whether the rebate prevents competitors from entering the market; and whether the rebate strengthens the firm’s dominant position. 34


The ECJ should build on its decision in Post Danmark II by continuing to move towards an effects-based approach when deciding whether rebate schemes foreclose competition for three reasons.

First, an effects-based approach forces courts to go beyond mere speculation and instead requires them to determine the actual or likely effects of a firm’s conduct – not just whether the pricing scheme is capable of foreclosing competition at some hypothetical point in the future. Insofar as it held that a pricing scheme must be “likely” to foreclose competition in order to violate EU law, the court in Post Danmark II took a step in the right direction towards looking at effects of a rebate rather than just focusing on a firm’s conduct. Without this new requirement, rebate schemes could be found to be anticompetitive even if there were a very low likelihood that such foreclosure would ever occur, so long as a rebate was merely “capable” of foreclosing competition. Also, by holding that the extent of a firm’s dominance is relevant to the analysis of whether a pricing scheme is anticompetitive, the ECJ moved towards assessing the actual likelihood that foreclosure will occur: after all, the more market share a firm holds, the more likely it is that its aggressive rebate scheme will foreclose its competition.

Second, by looking to the effects of a rebate scheme rather than hypothesizing about whether a firm’s conduct is capable of foreclosing competition, the law would protect efficient companies. Competition law does not and should not protect inefficient companies. 35 But requiring dominant firms to “ensure competition” is inconsistent with encouraging companies to be as efficient as possible in order to provide benefits to consumers. 36 If a dominant firm must preserve its competitors, foreclosing them by being more efficient would be a violation of its heightened responsibility – a result that would be unquestionably bad for consumers, which is why the law does not, and should not, protect inefficient competitors. By continuing to eliminate the responsibility of dominant firms to ensure competition, and by only requiring dominant firms to avoid foreclosing equally efficient competitors, an effects-based approach protects and encourages efficient business.

Third, an effects-based approach provides businesses, courts, and lawyers with clarity about what separates pro-competitive behavior from anticompetitive behavior. The ECJ should build on the factors set forth in Post Danmark II to ensure that anticompetitive foreclosure is measured by the actual or likely effects of a given rebate scheme, not just its hypothetical impact on competition. This approach would provide clear metrics, thereby giving clarity to judges and interested parties. In turn, the effects-based approach would foster more consistent decision-making.


For decades, EU competition law has involved too much speculation. Gradually, the ECJ has begun to move towards analyzing the effects of a rebate scheme instead of trying to determine whether certain conduct might hypothetically foreclose competition in the future. The Post Danmark II case marked progress in moving towards an effects-based analysis to determine whether a rebate forecloses competition. The ECJ should continue to steer the law towards the adoption of a purely effects-based analysis.


  1. Jonathan Hettleman is a 2L at New York University School of Law. This piece is a commentary on the 2017 Problem at the Global Antitrust Institute (GAI) Invitational Moot Court Competition held in Washington, D.C. The problem presented a fact pattern focused on whether a bundled pricing scheme violated either United States or European Union antitrust laws. The problem asked first whether the bundled pricing package violated § 2 of the Sherman Act, and second whether it violated Article 102 of the Treaty on the Functioning of the European Union (TFEU). The views expressed in this article do not necessarily represent the views of the author on either of these points of law. Rather, this article is a distillation of one side of an argument assigned to the team the author represented at the GAI Invitational.
  2. See, e.g., Richard A. Posner, How Judges Think 5 (2008) (“American antitrust law is far more the creation of judicial decisions than of antitrust legislation: the most important antitrust laws are as skimpy and vague as most provisions of the Bill of Rights.”).
  3. Herbert Hovenkamp, Federal Antitrust Policy: The Law of Competition and Its Practice 69 (5th ed. 2015) (referring to antitrust statutes as “vague and malleable”).
  4. See Joshua D. Wright, Comm’r, Fed. Trade Comm’n, Prediction in Antitrust is Hard (But Some Predictions are Harder than Others), Address at the Washington Bar Association’s 31st Annual Antitrust, Consumer Protection, and Unfair Business Practices Seminar (Nov. 13, 2013) (presentation available at (“Antitrust often requires . . . agencies to make predictions about how conduct today will influence competition in the future.”).
  5. See, e.g., Derek W. Moore & Joshua D. Wright, Conditional Discounts and the Law of Exclusive Dealing, 22 Geo. Mason L. Rev. 1215, 1 (2015) (“The appropriate antitrust analysis of conditional discounts remains a subject of considerable debate.” (citation omitted)).
  6. Compare Cascade Health Solutions v. PeaceHealth, 515 F.3d 883 (9th Cir. 2008) (applying a price/cost analysis to determine whether the bundled rebate is being sold below cost in order to determine whether it forecloses competition), with Lepage’s Inc. v. 3M, 324 F.3d 141 (3d Cir. 2003) (holding that a bundle may be anticompetitive even if its price is not below its cost, if the bundle will foreclose competition in the future), cert. denied, 124 S. Ct. 2932 (2004).
  7. See Verizon Communc’ns Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 U.S. 398, 414 (2004) (“[A]bove-cost predatory pricing schemes, [are] ‘beyond the practical ability of a judicial tribunal to control,’”) (quoting Brook Grp. Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 223 (1993)).
  8. See, e.g., Case 286/09 Intel v. Comm’n, (2014) E.C.R. I___ (delivered June 12, 2014), ¶ 103 (explaining that “it is unnecessary to undertake an analysis of the actual effects of the rebates on competition” in order to find that a pricing scheme forecloses competition in violation of Article 102 of the Treaty on the Functioning of the European Union (TFEU)).
  9. European Comm’n, Guidance on the Commission’s Enforcement Priorities in Applying Article 82 of the EC Treaty to Abusive Exclusionary Conduct by Dominant Undertakings, COM (2009) 864 final, ¶ 50 (Feb. 9, 2009) [hereinafter Guidance Paper], available at (advocating that courts apply a price/cost analysis in order to determine whether a bundled rebate scheme forecloses equally efficient competitors).
  10.  Id.
  11. Case C-23/14 Post Danmark A/S v. Konkurrencerådet, ECLI:EU:C:2015:651 [hereinafter Post Danmark II].
  12. See e.g., Case 322/81, NV Nederlandsche Banden Industrie Michelin v. Comm’n (Michelin I), (1983) E.C.R. 3461, ¶ 57 (“the undertaking concerned has a special responsibility not to allow its conduct to impair genuine undistorted competition on the common market.”).
  13. See e.g., Case T-219/99, British Airways plc v. Comm’n, 2003 E.C.R. I-2331 ¶ 23 (“[A] dominant undertaking is subject to certain limitations that do not apply to other undertakings in the same form. Because of the presence of the dominant undertaking, competition on the market in question is weakened. Therefore—whatever the causes of its dominant position—that undertaking has a particular responsibility to ensure that its conduct does not undermine effective and undistorted competition in the common market. A practice which would be unobjectionable under normal circumstances can be an abuse if applied by an undertaking in a dominant position.”).
  14. Case 286/09 Intel v. Comm’n, (2014) E.C.R. I___ (delivered June 12, 2014), ¶ 103.
  15. Case T-201/04, Microsoft Corp. v. Comm’n, (2007) E.C.R. II-3601, ¶ 868, 1035 (“…the fact that the Commission examined the actual effects which the bundling had already had on the market and the way in which that market was likely to evolve, rather than merely considering—as it normally does in cases of abusive tying—that the tying has by its nature a foreclosure effect, does not mean that it adopted a new legal theory.”).
  16. See id. at ¶ 1046.
  17. Id. 
  18. Id. at ¶ 1057.
  19. Id. at ¶ 1046.
  20. See, e.g., British Airways, 2003 E.C.R. I-2331, ¶ 271.
  21. Case T-85/86 Hoffmann-Law Roche v. Comm’n, (1979) E.C.R. 461, ¶ 89 (holding that exclusivity rebates offered by dominant firms per se foreclose competition).
  22. Case T-203/01 Manufacture Française des Pneumatiques Michelin v. Comm’n, (2003) E.C.R. II-4071, ¶ 65.
  23. British Airways, (2003) E.C.R. I-2331, ¶ 271. The rebate at issue in the British Airways case was fidelity-building because British Airways was offering individualized retroactive rebates. In other words, British Airways was conditioning the rebate on individual ticketing agents meeting a specific sales target that was based on their prior sales numbers. Thus, the rebate induced the ticketing agents to purchase more and more of their requirements from British Airways in order to meet the target and qualify for the rebate.
  24. Post Danmark II, ECLI:EU:C:2015:651.
  25. Id. at ¶ 67.
  26. Id. at ¶ 39.
  27. Id. at ¶ 27-28.
  28. Id. at ¶ 27.
  29. Id.
  30. Id.
  31. Id.
  32. Id. at ¶ 28.
  33. Id.
  34. Id. at ¶ 29.
  35. Moritz Lorenz, An Introduction to EU Competition Law 216 (2013).
  36. See, e.g., Michelin I, (1983) E.C.R. 3461, ¶ 57 (“the undertaking concerned has a special responsibility not to allow its conduct to impair genuine undistorted competition on the common market.”).