Analysis of Platform Vertical Contracts: Price Versus the Competitive Process in American Express

The continuing debate surrounding the Supreme Court’s 2018 landmark decision in Ohio v. American Express Co. (Amex) indicates that a great deal of uncertainty remains regarding how the anticompetitive effects of platform vertical contracts should be analyzed. Much of this debate concerns what particular price increase should be used to evaluate the alleged anticompetitive impact of the American Express (Amex) contract provisions that forbid merchants from “steering” their customers to another credit card at the point of sale. The dissent focused on evidence it claims demonstrates that these contractual restraints increase merchant fees, arguing that both antitrust precedent and economic analysis imply that this result is sufficient evidence of anticompetitive effects caused by the antisteering restraints. In contrast, the majority emphasized that merchant fees are only one element of the payment system’s total price charged to merchants and cardholders considered together on the two sides of the credit card platform, and that demonstrating anticompetitive effects therefore requires evidence that the antisteering restraints increased this total price of an Amex transaction. “Price increases on one side of the platform . . . do not suggest anticompetitive effects without some evidence that they have increased the overall cost of the platform’s services,” the majority held.

A large number of articles have discussed Amex, some agreeing with the majority and some with the dissent regarding how anticompetitive effects should be evaluated. A recent article in this Journal misrepresents the state of the debate by totally dismissing the Amex decision, describing it in the article’s first sentence as “what may be the worst antitrust decision in many decades.” Because my co-authored article is favorably cited by the majority fourteen times in support of its analysis, I have decided to enter this debate.

Perhaps surprisingly, I argue in this article that neither the merchant price nor the total price should be the focus of the anticompetitive effects analysis, and that an increase in either price should not be considered sufficient direct evidence of anticompetitive effects of the Amex vertical restraints at issue. Instead, I argue that one should specify the anticompetitive theory of how the vertical restraint is alleged to distort the competitive process in the sense of significantly decreasing the ability of rival platforms to effectively compete. Once that anticompetitive theory is specified, the requisite evidence of anticompetitive effects of the vertical contracts is then implied. This framework for assessing evidence of anticompetitive effects is shown to be fundamentally consistent with current U.S. antitrust law of vertical contracts, the underlying principles of which are applied in this article to vertical contracts instituted by platforms.

Establishing the anticompetitive effects of a vertical contract requires that the firm (or platform) instituting the contract possess sufficient market power. This safe-harbor market-power requirement is important because it permits a firm that clearly lacks significant market power—but sells a differentiated product and thus faces a negatively sloped demand—to use a vertical contract to earn increased rents on its firm-specific assets as an essential part of the normal competitive process. Vertical contracts are commonly used for this purpose, with the often-resulting increase in the firm’s price or profits not serving as an appropriate measure of anticompetitive effects in the sense of creating or maintaining market power. In these circumstances, the alternative of focusing solely on the effect of a vertical restraint as increasing price would involve what I refer to as “microregulation” of the competitive market process. This would render suspect all vertical contracts where market results deviate from the perfectly competitive benchmark of price equal to marginal cost, which is not the established or appropriate role of antitrust analysis of vertical contracts.

Neither the majority nor the dissent in Amex conducted a market-power analysis and reached a conclusion regarding whether Amex possessed market power in the credit card platform relevant market. The majority avoided what it considers to be this required first step of the indirect antitrust standard it claims to employ. Instead, it argued that there is an absence of evidence of anticompetitive effects of the Amex antisteering restraints in terms of a higher total price (or a reduction in quantity or quality) of Amex transactions in the relevant credit card platform market. The majority therefore implicitly adopted the direct-evidence antitrust standard used by the dissent, where “proof of actual adverse effects on competition is, a fortiori, proof of market power. Without such power, the restraints could not have brought about the anticompetitive effects that the plaintiff proved.” The essential question underlying the disagreement between the majority and dissent therefore is: What should direct evidence of anticompetitive effects of vertical restraints from which market power may be inferred consist of?

The antisteering vertical restraints at issue in Amex restricted the ability of merchants to “steer” customers who present the Amex card to a rival credit card. Part I presents the basic economics and institutional background of credit card platforms. Part II then discusses the ostensible disagreement between the Amex majority and dissent regarding whether an indirect- or direct-evidence antitrust standard should be used to evaluate the presence of anticompetitive effects of the Amex antisteering restrictions. Part III describes the economic and legal basis for a market-power safe-harbor requirement, and Part IV describes what should constitute the direct evidence of anticompetitive effects from which the anticompetitive exercise of market power can be inferred. It argues that such evidence should demonstrate use of the vertical restraint to distort the competitive process in the sense of causing a significant decrease in the ability of rivals to compete effectively. Part V summarizes the claimed evidence of anticompetitive effects in the Amex record presented by the dissent and the majority. Part V shows that in both cases, the evidence referred to does not meet the necessary condition of demonstrating such a distortion in the competitive process.

Continue reading the full text of this article in PDF format.