Friday, March 15, 2019
Herb Hovenkamp weights in on AMEX's RULE OF REASON. Worth reading.
ABSTRACT: In Ohio v. American Express Co., the Supreme Court applied antitrust’s rule of reason to an allegedly anticompetitive practice on a two-sided platform. The challenge was to an “anti-steering” rule, a vertical restraint preventing merchants from shifting customers who offered an AmEx card from to a less costly alternative such as Visa or Mastercard.
A two-sided platform is a business that depends on relationships between two different, noncompeting groups of transaction partners. For example, a printed periodical such as a newspaper earns revenue by selling both advertising and subscriptions to the paper itself. Depending on its business model, the periodical might obtain very different mixtures of advertising and subscriber revenue. In some cases, such as most credit cards, card holders are actually negative contributors of revenue: they receive awards for using a particular card, paid through merchant fees. Success depends on a platform’s ability to maintain the appropriate balance between participation on one side and the other. For example, if Uber, a two-sided platform offering transportation services, sets too high a fare it will have enough drivers but too few passengers. If the price is too low, it will have too many passengers in relation to available drivers.
Administering antitrust under the rule of reason depends on careful fact finding. Under antitrust’s per se rule, once a practice is proven little evidence of anticompetitive effects is relevant. By contrast, the rule of reason requires a searching factual examination of a record, enabling the court to understand the effects of the defendant’s activities. This obliges appellate courts to review the record, and the Supreme Court’s AmEx opinion should be tested against this requirement. The Court simply ignored specific district court fact findings to the effect that AmEx’s anti-steering rule imposed higher costs on everyone, including customers who purchased with cash.
The Court also mistakenly concluded that AmEx’s anti-steering rule combatted free riding, because otherwise competing card issuers could profit from AmEx’s business model. That might be true if one could obtain AmEx’s perks merely by owning the card, but the record and AmEx’s own literature were clear that one received the perks only for transactions actually made with the card, so there was no free riding.
More generally, the Court’s analysis assumed that costs on one side of a two-sided platform are offset by gains on the other side. In some situations, such as the Uber example above, that may be true. But in the AmEx case both sides of the platform were harmed by the anti-steering rule. Suppose that the merchant fee for use of an AmEx card was $30 but only $20 if the customer purchased with a Visa card. This $10 difference creates bargaining room—a Coasean “surplus”—for the merchant and the cardholder to strike a mutually beneficial deal. The merchant might offer the customer a $6 discount for using a Visa card instead, which would make the customer $6 better off for that particular transaction and the merchant $4 better off. The customer would agree if the value it placed on AmEx’s perks was less than the $6 price discount.
The anti-steering provision would prohibit this transaction. As a result, the customer stays with the AmEx card and experiences a $6 loss. The merchant loses $4 as well. Far from being a situation where value goes up on one side and down on the other, it actually goes down on both sides. In addition, the competing platform, Visa, is also worse off because it was denied the opportunity to offer a lower cost substitute transaction. The only entity that is better off is AmEx—the owner of the platform itself, but not the dealing parties on one or the other side of the platform.
Competition always exists at the margin. One cannot evaluate the competitive effects of a particular restraint by considering whether the overall costs of a defendant’s business practices exceed the benefits. For example, in the NCAA case the challenge was not to the existence or legitimacy of the NCAA as an association, which no one was challenging. Rather it was to the effect of a limitation on the televising of games. The AmEx challengers were not trying to tear down AmEx’s entire business model, but only to enjoin its anti-steering rule. That requires assessing the marginal costs and benefits of the anti-steering rule. The record was clear that at the margin each merchant affected by the steering rule was worse off, and each cardholder was worse off as well.
The Court’s assumption about offsetting cost and benefits also led it to reach the economically absurd conclusion that in a qualifying two-sided platform, both sides should be included in the market definition. The two sides are, of course, complements in production, not substitutes. This holding is thus in conflict with an idea that is central to antitrust analysis, which is that relevant markets consists of a “collusive group,” or substitutes.
Further, the Court exacerbated this problem by holding that in a case involving a vertical restraint a relevant market must be defined, even if proof is proceeding through examination of direct effects. That conclusion, which was never properly briefed or argued, flies in the face of decades of progress toward more accurate methodologies for assessing power. Further, it turns into a question of law something that is clearly an issue of fact – namely whether the defendant has sufficient power to produce an anticompetitive restraint.
The Court did limit the scope of its holding to situations involving a simultaneous one-to-one transaction across the two sides of the platform. That would likely include credit card networks as well as other direct transactional networks such as Uber. However, it would exclude situations where there is not a simultaneous one-to-one relationship between transactions on the two sides, such as Netflix or music streaming services, periodicals or search engines partially supported by advertising, or most exchanges on sports networks. Neither would it include situations where the relationship between the transactions on the two sides is actuarial, such as health insurance networks.