Given myriad business practices and conditions, establishing certain antitrust harms requires context. This context often comes from framing the effects of the challenged conduct against the backdrop of some relevant market. In particular, a challenged practice’s anticompetitive impact may be outsized in a smaller market but insignificant in a larger one. For instance, the impact of a merger of two firms might be major if their combination captures 80 percent of a smaller market but relatively minor if that same combination represents only 10 percent of a larger market. In the latter, the combination is a drop in the bucket, while in the former it is a tidal wave. Plaintiffs thus have incentives to plead narrower relevant markets while defendants prefer broader ones. Yet this process of defining the relevant market can be highly technical, thrusting on judges the task of reining in increasingly complex economic and statistical analyses.

Challenging as this may be, the Supreme Court may face a taller task this term in Ohio v American Express Co.1 In particular, recent recognition of business methods known as “multisided platforms” further muddies the water by introducing additional complexity during market definition.2 These business methods connect two or more distinct groups of consumers that would benefit from interacting but face barriers to doing so. For example, payment-card platforms connect cardholders and merchants by issuing cards on one side of the platform and providing card-processing services on the other.3 Video game platforms connect gamers with developers by selling game consoles on one side of the platform and development kits and licenses on the other.4 Newspapers similarly connect subscribers and marketers by selling news content on one side of the platform and advertising space on the other.5 In this manner, selling more to one group influences demand by the other group.6 Often, the way platforms attract suitable participation is by offering discounts to the harder-to-attract group at the expense of the more ready-and-willing group. For instance, payment-card users frequently receive rewards—a sort of negative price—to ensure their participation.7 These are funded through fees assessed to merchants that desire cardholder business.8

To the technical process of market definition, multisided platforms thus add the challenge of determining whether the relevant market should incorporate all, some, or none of these interconnected groups. Because certain forms of antitrust analysis allow courts to trade off pro- and anticompetitive effects within the relevant market, defining that market serves to establish the space of permissible trade-offs.9 Notably, defining the relevant market to include all sides of the platform creates a broader space of allowable trade-offs than when the definition encompasses fewer sides. This is one reason that winning at market definition often means winning the case. Thus, it matters crucially whether and how courts incorporate “multisidedness” during market definition. This Comment suggests an approach to that inquiry.

Part I explores the challenges inherent in market definition. It situates this definitional process within the framework of the major antitrust enforcement laws to clarify the judicial requirements for defining a relevant market. It then frames market definition as both central and contested, pausing to chronicle Supreme Court jurisprudence designed to structure the inquiry and mitigate likely pitfalls. Part II explores the added challenge of whether the relevant market should incorporate all, some, or no sides of the platform. It describes the economic concept of a multisided platform and details the different ways that lower courts have addressed this challenge. Because plaintiffs and defendants often have opposing incentives concerning the number of sides that should be incorporated, knowing when to disregard or accept arguments concerning multisidedness is important. In this context, the courts could benefit from a test that seeks to determine when the relevant market should incorporate all sides of the platform. Hence, Part III proposes a two-stage test for multisidedness.

In the first stage, a court would ask (1) whether the business method can explicitly charge different prices to the distinct groups to which it provides goods or services; (2) whether each group’s benefit depends on the extent of participation by the other groups provisioned by the same business method and whether that participation varies based on market conditions; and (3) whether the platform is capable of, and generally does, set uniform prices in the markets in which each group participates. All three factors must hold, or multisidedness should be excluded from market definition.

Assuming all three stage one factors are satisfied, the court moves on to the second stage and asks whether the challenged conduct is designed principally to ensure the continued availability of the platform’s differentiated products. If so, the relevant market should encompass the market segments in which all sides of the platform operate.

  • 1. 2017 WL 2444673 (US). For the decision below, see United States v American Express, 838 F3d 179, 188–89 (2d Cir 2016) (“Amex”).
  • 2. See text accompanying note 83 (defining a platform as a business method that enables interactions between distinct groups, each of which cares about the extent of the other group’s participation).
  • 3. See text accompanying notes 85–89.
  • 4. See Marc Rysman, The Economics of Two-Sided Markets, 23 J Econ Persp 125, 125, 129–31 (Summer 2009) (asserting that both video game systems and payment cards are examples of multisided platforms).
  • 5. See E. Glen Weyl, A Price Theory of Multi-sided Platforms, 100 Am Econ Rev 1642, 1642–43 (2010) (describing credit cards and newspapers as canonical examples of multisided platforms).
  • 6. See notes 97–98 and accompanying text.
  • 7. See Rysman, 23 J Econ Persp at 129 (cited in note 4).
  • 8. See notes 136–37 and accompanying text.
  • 9. See notes 109–10 and accompanying text.