Argument analysis: Clash over market definition, competitive harms and burden shifting
on Feb 27, 2018 at 5:12 pm
The Supreme Court heard oral argument yesterday in Ohio v. American Express Co., a challenge by 17 states to the network rules that prohibit merchants accepting Amex cards from steering customer-cardholders toward using other cards that charge the merchants lower transaction fees. My argument preview outlined the facts of the case and explained the relevant legal principles.
Justice Neil Gorsuch led off an active argument by questioning Ohio Solicitor General Eric Murphy about the central goal of the antitrust laws — protection of competition rather than specific competitors — and whether there was evidence in the record of any limitation of output. Assuming that the challenged provision increased prices to merchants, Gorsuch wanted to know whether or not there had been a net price increase to cardholders, taking into account any rewards benefits the cardholders received from using an Amex card. Murphy responded that, although it was not the plaintiff’s burden to prove competitive benefits from the challenged practice, there was record evidence that not all of the higher amount charged to merchants was transferred to cardholders. Gorsuch was not satisfied, reiterating that the question concerned net prices paid by cardholders.
The argument turned to the critical distinction between interbrand competition, among different brands, and intrabrand competition, among sellers of the same brand. Leegin Creative Leather Products Inc. v. PSKS Inc., Murphy emphasized, concerned resale price maintenance, a vertical restraint that limits price competition among sellers of the same brand of products, which may in turn increase interbrand competition. Murphy argued that the anti-steering provision restricts merchants from offering options to cardholders by providing small discounts — for using a different credit card that charges a lower fee to the merchant, for example — that would increase interbrand competition. The states did not claim that there had been a horizontal agreement among credit-card networks, but because the market is concentrated and the major networks all used the anti-steering provisions, the effect of the vertical agreement in this case was horizontal.
Justice Anthony Kennedy turned the subject to market definition and the understanding of competitive harm. In a multi-sided platform such as a credit-card network, should the proper antitrust analysis focus on output, as argued in the amicus brief for Antitrust Law & Economics Scholars supporting Amex? Murphy agreed that output in terms of expanding use of credit cards has expanded, but he disagreed about causation and responded that to make their prima facie case, the states were require to prove that the vertical restraint caused harm to the merchants by raising prices or restricting output. More specifically, Murphy pointed to Supreme Court precedent that forbids firms from justifying price restraints by pointing to other countervailing benefits, such as safety in National Society of Professional Engineers v. United States, or credit terms in Catalano Inc. v. Target Sales Inc.
Disputes about market definition are standard arguments in antitrust cases, in which markets must be defined in order to show direct harm to competition or to infer harm by showing market power. The key question of what the relevant market is in this case was teed up by Justice Ruth Bader Ginsburg. Murphy responded that the states were challenging a vertical non-price restraint imposed by credit-card networks upon merchants that accept the cards. That agreement between the network and the merchants constitutes the proper market for the analysis, he asserted. Even if the Supreme Court agrees with the U.S. Court of Appeals for the 2nd Circuit and defines the market as the total network transaction including services to merchants and to cardholders, he concluded, the states should still win because this vertical restraint limited interbrand as well as intrabrand competition, by foreclosing cardholders’ choice between high reward/high cost cards and low cost/low reward cards.
Deputy Solicitor General Malcolm Stewart argued in support of the states. The Department of Justice, it should be noted, was an original plaintiff at trial and on appeal, but opposed cert. When the petition was granted, DOJ filed an amicus brief and argued in support of the states. The case is a simple one, Stewart argued: The purpose and effect of the Amex anti-steering provision eliminated price competition. First, at the outset of the case, the issue is the competitive effect on the market for network services to merchants. The states and DOJ proved their prima facie case by showing competitive harm. Second, turning to the consumer-cardholder side, the issue is whether the anti-steering provision harmed or enhanced competition there by preventing merchants from offering options that cardholders might or might not value.
Gorsuch expressed concern that antitrust analysis of two-sided markets is new and, bowing in Judge Frank Easterbrook’s direction, pointed out that market errors are more quickly self-correcting than judicial ones. Stewart agreed that the Supreme Court should decide the case narrowly and not seek to adopt a standard for all two-sided markets at this stage.
Nevertheless, Stewart urged the court to recognize longstanding principles and precedent that define markets by identifying substitutes. He agreed with Justice Elena Kagan that at the second stage of the rule of reason, the effect, if any, on the cardholder side becomes relevant. The two cannot be conflated: Amex competes with other networks on the merchant side, but with issuing banks on the cardholder side. The three-step rule-of-reason approach necessarily recognizes this distinction, he clarified to Kennedy. Benefits, if any, are considered at step two of the rule of reason, and are the defendant’s burden to establish.
Justices Stephen Breyer and Sotomayor sought clarification of the model in an extended colloquy. Again stressing the advisability of a narrow decision, Stewart pressed the justices not to pronounce a categorical rule on multi-sided markets, but to clarify the meaning of competitive benefits, step two of the rule of reason. Gorsuch reminded Stewart that the individual credit-card networks remain free to set their own merchant fees, to adopt a high fee/high reward or low cost/low reward alternative as they choose. This case is unusual, Stewart explained, because when it was brought, the networks all used anti-steering provisions. Even absent such provisions by Visa and MasterCard, he added, the Amex anti-steering rule prevented merchants from informing customers about the merchant fees. Of course, the networks themselves can inform cardholders about their merchant fees, Gorsuch pointed out.
Arguing for Amex, Evan Chesler stressed that the relevant market is credit-card transactions, simultaneous transactions between cardholders and merchants facilitated by the network. But the anti-steering provision forecloses competition at the point of sale, inquired Sotomayor, doesn’t it? Merchants are precluded from offering discounts to consumers who use a different card. Cardholders may value a small discount, say one percent, or they may prefer any airline miles or other rewards offered by the credit card, for example. Competition between credit-card brands remains and is enhanced, Chesler responded. But consumer choice is the essence of competition, Sotomayor stated. Chesler reiterated that the product should be defined as “credit card transactions,” and that because the plaintiffs failed to prove competitive harm in that market, they did not carry their prima facie burden.
Breyer returned to first principles, remarking that he had been using classic antitrust language for 40 years. The credit-card networks are a variation on vertical restraints between producers and dealers that restrict intrabrand competition, limiting dealers, but may benefit ultimate consumers. Step one of the rule of reason, Breyer stated, looks at the anticompetitive effect of the vertical restraint. Proof of pro-competitive justifications occurs at step two. Although he acknowledged the standard rule-of-reason analysis, Chesler responded that the entire transaction (cardholders and merchants brought together by the network) is the market and at the time of trial, DOJ claimed that Amex had 26 percent of the market. More to the point, output in terms of dollar volume has increased significantly. Chief Justice John Roberts then intervened to question causation. Other factors, including overall economic growth, may increase credit card transactions, he suggested.
Sotomayor, Breyer and Kagan turned the focus back to consumer choice. Merchants were prevented by the anti-steering provision from informing consumers of the merchants’ fees or passing savings to consumers at the point of sale. Proving benefits to cardholders, Sotomayor pointed out, appropriately falls within step two of the rule of reason, because the defendants have that unique information and the government does not.
Gorsuch reiterated that in the absence of market power, a vertical restriction “is not within the cognizance of the antitrust laws.” And, without more, a 26 percent share of a market does not constitute power. Market power is a gremlin to throw in the gears, Breyer noted, but why is proof of power necessary in the face of proof of an anticompetitive effect? Chesler urged the court always to start by defining the market. He agreed with Stewart that to decide this case, the court does not have to adopt a market definition or analytic approach for every multi-sided market, but merely has to find that this credit-card platform constitutes the relevant market.
During Murphy’s rebuttal, Ginsburg asked what relief the states were seeking. Murphy asked the court to rule that the government has proved its prima facie case, and that, on remand, Amex has the burden of proving step two of the rule of reason: pro-competitive benefits. Gorsuch asked whether Murphy agreed that the states must prove market power to show anticompetitive effects. Yes, he responded, asserting that the states did show market power in the form of actual effect.
[Disclosure: Goldstein & Russell, P.C., whose attorneys contribute to this blog in various capacities, is among the counsel on an amicus brief in support of the petitioners in this case. However, the author of this post is not affiliated with the firm.]