In one of the year's most anticipated antitrust decisions, Ohio et al. v American Express Co., the U.S. Supreme Court has ruled that courts considering allegations of anticompetitive behavior by companies that operate "two-sided transaction platforms" must consider the economic interaction of the various sides of the platform in defining the relevant market. The decision also clarifies the obstacles for plaintiffs seeking to challenge vertical agreements and behavior; the Court confirmed that plaintiffs must bear the burden of proving a relevant market, and that the defendant has market power in such a market, even when relying on direct evidence of competitive harm such as increases in price.
Today's ruling by the Court, which came on a 5-4 vote, has potentially far-reaching implications for antitrust review of companies that operate multi-sided platforms, especially in the financial services sector. In addition, while the decision may be read narrowly as applying only to "transaction platforms," it does not foreclose potential application to other markets, given that the opinion also included sophisticated economic analysis that may apply more broadly to other "nontransaction platforms," such as in technology sectors.
Summary of Holding
The Amex litigation began in 2010 when the U.S. Department of Justice (DOJ) and several states (collectively, "plaintiffs") sued American Express (Amex), together with Visa and MasterCard, claiming that the anti-steering provisions in their respective merchant contracts violated Section 1 of the Sherman Act. The provisions at issue prohibited merchants who accept Amex cards from "steering" customers towards payment options with lower merchant fees, such as Visa or Discover. The Visa and MasterCard agreement had parallel provisions. They both settled when the complaint was filed.
According to the plaintiffs, the anti-steering provisions prevent competition among payment cards at the point of sale, allowing Amex to raise prices to merchants without losing market share to other card networks. After a seven-week trial, the district court ruled for the plaintiffs, finding that the challenged agreements had harmed competition for merchant payment services. The U.S. Court of Appeals for the Second Circuit reversed in 2016, holding that the district court had erred in defining the relevant market to include only merchant payment services and declining to consider the potential benefits of the challenged provisions—such as increased loyalty rewards and wider acceptance—for Amex cardholders and the Amex platform as a whole. The court of appeals concluded that, because every card transaction involved a merchant on one side and a cardholder on the other, the district court had erred into not focusing on the two-sided nature of the market.
The Supreme Court's decision upheld the Second Circuit by a 5-4 vote. The majority opinion, authored by Justice Thomas, focused on the close economic relationship between the two sides of Amex's "transaction platform": merchants and cardholders. Relying heavily on economic literature on the unique features of two-sided platforms, the Court noted that such platforms are frequently characterized by indirect network effects, where the value of the platform to customers on one side depends materially on the number of participants on the other side. More importantly, the Court suggested that a restraint on one side of the market (i.e., the anti-steering provisions in contracts with merchants) may be necessary to attract customers on the other side (i.e., cardholders, who expect "a frictionless transaction" at the point of sale). Transaction platforms like Amex represent a special case: because such platforms "cannot make a sale to one side of the platform without simultaneously making a sale to the other," the Court held, they are "better understood as 'supplying only one product'—transactions". To prevail, therefore, the Court held that the plaintiffs would have had to prove that Amex's agreements had caused an anticompetitive effect in the market for "credit-card transactions" as a whole, rather than merely proving a price increase to either merchants or cardholders in isolation. Specifically, a price increase to merchants on one side of the market is not enough. There must be evidence of harm to competition in the market as a whole, such as a reduction in output, i.e., transaction volume. Because there was no such proof, the Supreme Court affirmed the Second Circuit.
While its ultimate impact on the antitrust analysis of multi-sided platforms remains unclear, Amex clarifies the high bar for potential antitrust plaintiffs seeking to challenge vertical agreements (i.e., agreements "between firms at different levels of distribution"), policies, or mergers. Such plaintiffs will now need to prove the existence of market power in a specified relevant market, rather than merely showing price effects, and will need to support their allegations of harm with rigorous economic evidence. Future complaints also will need to seriously consider the role of indirect network effects in defining the relevant market, including whether such effects are sufficiently strong as to render a given two-sided platform a single market for market definition purposes.
Amex Clarifies the Substantial Burdens on Antitrust Plaintiffs in "Vertical Restraints" Cases
The Court in Amex confirmed the applicability of the widely accepted three-step Rule of Reason analysis, under which the plaintiff bears the initial burden of showing that the challenged practice has an anticompetitive effect.1 However, in contrast with prior cases allowing plaintiffs challenging horizontal agreements to meet their burden by showing "direct evidence" of harm to competition, the Court clarified that plaintiffs challenging vertical agreements must first prove the existence of a relevant market and show that the defendant has market power within it.2 Holding that "[v]ertical restraints often pose no risk to competition unless the entity imposing them has market power, which cannot be evaluated unless the Court first defines the relevant market," the Court rejected as inapplicable prior horizontal decisions which had allowed "proof of actual detrimental effects [on competition]" to substitute for a formalized market definition.3
Adding to the plaintiffs' difficulty, the Court also took a skeptical view of the direct evidence relied on by the district court, which consisted of showing that Amex had raised prices to merchants multiple times over five years.4 Citing Brooke Group for the proposition that when "output is expanding at the same time prices are increasing, rising prices are equally consistent with growing product demand," the Court held that a price increase to one side of a multi-sided platform, standing alone, is not sufficient to show an anticompetitive effect or market power, particularly in a context where transactions had grown by 30 percent over the same five-year period, and non-defendants Visa and MasterCard had increased prices as well.5 To rebut the inference that Amex's price increases were merely the result of increased demand, the Court found that the plaintiffs would have needed to show that "output was restricted or prices were above a competitive level"—e.g., that the "the price of credit-card transactions was higher than the price one would expect to find in a competitive market."6
Bringing "Indirect Network Effects" into the Foreground for Platform Market Definition
More broadly, the Court held that, when "indirect network effects" are strong enough (as demonstrated by "interconnected pricing and demand"), a two-sided platform should be seen as "supplying only one product—the transaction."7 Under this analysis, a prospective plaintiff challenging the behavior of a "transaction platform" would need to show harm to the market served by the platform as a whole—not just to one segment of users. Price increases—or, equally, below-cost prices, which might otherwise be seen as indicia of predatory pricing—are required to be assessed in the context of "overall cost of the platform's services." As the Court's heavy citation of economic testimony suggests, such an assessment is likely to require substantial economic sophistication on the part of potential plaintiffs, even in cases where evidence of price effects on a specific customer group would otherwise be readily available.
Unclear How Far Amex Will Reach with Respect to "Tech Platforms"
Despite these groundbreaking holdings, it remains unclear how far the Amex decision will reach, particularly with respect to so-called tech "platform companies." The Court's majority relied heavily on the fact that Amex's platform is designed to facilitate a "single, simultaneous transaction between participants." The Court declined to extend its holding to other platform-like businesses, such as newspapers, where indirect network effects are less pronounced. In such industries, where users on one side of the platform are "largely indifferent" to whether those on the other side participate, the Court held that the "market… behaves much like a one-sided market and should be analyzed as such."8
The Amex holding is therefore potentially narrower than it appears at first glance. To the extent that tech platforms are viewed by future courts and regulators as selling complements—or even services that lack the tight linkage between output on the two "sides" exhibited in credit cards—the impact of Amex may be more limited, at least with respect to market definition and assessment of competitive harm. Nevertheless, the level of indirect network effects necessary to qualify for two-sided market treatment was not quantified in the opinion and will need to be tested in future cases over time. Some platforms will qualify, others will not; and in every case, it will all depend on the specific claim.