The case hinges on whether American Express’s anti-steering policy, which prevents merchants who accept their cards from encouraging customers to use competitor’s (like Discover) cards instead, violates existing competition law.
Retailers have claimed that American Express is engaging in anti-competitive behavior by not allowing them to encourage the use of cards which charge them lower merchant fees, thereby exerting market power. The Second Circuit court, which ruled in favor of American Express, stated that credit card companies serve “two sided markets,” which means that their consumers are not just merchants, but also card holders. It therefore held that the enforcement of antitrust law requires analysis of whether the policy harms both sides. The evidence brought before the court suggested that American Express’s policy in fact benefitted its card holders, by allowing them to freely use a card that gives them greater perks.
This may seem like a dull technical question, but the introduction of this form of analysis into antitrust law is a significant development, and one that would significantly reduce the rates of antitrust “false positives” (antitrust actions against pro-competitive behavior). This would increase consumer welfare.
The theory surrounding two-sided markets is fairly new, pioneered by Nobel Prize winner Jean Tirole, and has started gaining traction with the rise of platform firms like Uber, AirBnB, and Amazon, which exemplify this type of structure. Two-sided markets, unlike traditional one-sided ones, are more complicated in their pricing structure and competition dynamics, since the service they provide to each side of their market depends on how they treat the other one.
Credit card companies which match merchants and cardholders can only be effective if they have enough merchants who take them to be a benefit to cardholders, and enough cardholders that use them to be a benefit to merchants. Other examples of two-sided markets include newspapers (readers and advertisers), nightclubs (men and women), shopping malls (retailers and shoppers), testing services (universities and students), ride sharing apps (drivers and riders), search engines (websites and searchers), among numerous others.
In terms of antitrust enforcement, these types of markets have not been defined, which has led to cases of enforcement that mischaracterize the market, which lead to false positives for enforcement. In the early 2000s, antitrust action against credit card companies in Australia and the United Kingdom fell short as a result of this.
MasterCard, a company that provides a similar service to American Express, was charged with being anticompetitive since it charged interchange fees above marginal cost (the cost of each additional use), which ended up costing merchants a significant amount. High interchange fees effectively reduce the cost to cardholders of using the card, while they raise the cost of merchants for accepting the card. Regulators believed that this showed MasterCard had market power to charge merchants more than the market would allow. While this analysis would generally make sense in a one-sided market, it did not apply to the realities of two-sided markets.
An illustrative example from economist Julian Wright helps explain this difference in pricing structure. Nightclubs are a type of two-sided matching market between men and women. Assume that men decide which nightclub they go to based on the number of women there, and prefer those with fewer men. If this is the case, then a nightclub, in order to maximize revenue, will charge women less than men, in order to increase the number of women in the club. This will increase the number of men willing to pay the higher amount. This example is not just theoretical, as the widespread prevalence of “ladies’ nights” at clubs across the world attests. The fact that the nightclub can get away with charging men more is not because it has significant power over men (places with more nightclubs are not less likely to have ladies’ nights), but reflects the decisions the club needs to make in order to appeal to its customers.
The nightclub would not be able to get away with charging men above cost prices if there were no women in the club, since they are paying more specifically because they expect there to be many women there. This means that charging women less actually increases the value of going to that club for men, which then justifies the increased cost to them. Credit cards in a similar way cannot charge the higher fees to merchants unless doing so increases the number of cardholders that will shop with their card, thereby increasing merchant revenues. By charging merchants above cost, MasterCard reduced fees for cardholders, increasing their propensity to shop, increasing the value of MasterCard to merchants. (A detailed explanation of the mistakes made by regulators in Australia is available in this study by Professors Jason Potts and Sinclair Davidson of RMIT in Melbourne.)
American Express is following this exact same logic in the contracts it draws up with retailers today, which ends up benefiting its customers overall. Restricting their ability to create these sorts of anti-steering policies with retailers would make it harder for the company to give perks to cardholders, which would reduce their likelihood to shop, hurting those exact same retailers in the long-run.
The Second Circuit’s analysis of the case was not only correct, but impressive in bringing up to date the understanding of complex market structures in antitrust analysis. The Supreme Court would be wise to affirm this ruling, thereby setting a precedent for those in two sided markets to improve their services without fear of antitrust suits against them.
As more and more services move into platform business models as the lower transaction costs they facilitate increase consumer welfare, platform firms need assurance that the law understands how they operate, and does not unfairly hinder their development. While much more needs to be done to improve the regulatory environment for these companies, acknowledging the competitive structure of two sided markets is a significant step in the right direction.