David Evans, Sep 10, 2013
Last week a court appointed expert in the interchange fee litigation in the United States submitted a memorandum that evaluated the economic arguments presented by the plaintiffs and defendants. His memorandum highlights the importance of the new economic learning on two-sided platforms, and some of the questions courts will be grappling with in cases involving these platforms. These platforms constitute a large and growing part of the economy.
NYU Professor Alan Sykes, who has a Ph.D. in economics and a law degree, both by way of Yale, submitted a 51-page memorandum to Judge John Gleeson. Judge Gleeson is overseeing the proposed settlement in the antitrust case brought by merchants against the four-party payment card networks and merchants over interchange fees and other network rules. The proposed settlement has gotten highly contentious. Many merchants have opted out of it. Some are trying to block it altogether. Professor Sykes’ assignment was to assess whether the settlement was a reasonable deal for the merchants. To do so, he examined the likelihood of the plaintiffs prevailing on establishing liability and proving damages.
The proper antitrust analysis of two-sided platforms was central to his findings. His discussion provides an excellent road map of issues that are likely to arise in other cases. It also emphasizes that, at least among economists, there remains little controversy that sound economic analysis needs to take the features of two-sided platforms into careful account.
This short article summarizes those aspects of his memorandum that touch on two-sided platforms; it does not address the other aspects of the interchange fee case and proposed settlement. However, it is useful to summarize his key conclusions since the consideration of two-sided issues were key to reaching them. He found that:
“[T]he plaintiffs face considerable difficulty in establishing that the core practices at issue in the case and left in place by the proposed settlement—such as default interchange and honor all card rules—cause anticompetitive harm that outweighs their pro-competitive benefits…. In addition, even if the core practices at issue were deemed anticompetitive on balance, the plaintiffs face considerable difficulty in proving their damages, as it is exceedingly challenging to identify the proper counterfactual for purposes of a damages computation.”
Professor Sykes’ key point in support of this conclusion is that the economic analysis of these two-sided platforms needs to account for the following considerations:
- payment card networks have two sets of customers ultimately: merchants and cardholders;
- a change in price for one type of customer will likely change the price for the other type of customer in the opposite direction;
- the relevant price to examine is the total price charged to both types of customers; and,
- the relevant notion of welfare is that obtained by the two types of customers together.
These considerations affect the proper analysis of market power, competitive effects, and damages.
Market Definition and Market Power
Professor Sykes concluded that the analysis of market definition for two-sided platforms must account for the fact that an increase in price to customers on one side could be offset by a decrease in price to customers on the other side:
“In a two-sided market, however, the possibility arises that a price increase on one side of the market will be wholly or substantially offset by a price decrease on the other side of the market.”
Therefore, for the interchange fee case the analysis must consider whether MasterCard and Visa could raise the total price of payment services through an increase in the interchange fee. One reason for this conclusion is that the card networks would not necessarily increase profits by raising prices:
“With reference to the general purpose card market, if, hypothetically, any increase in interchange fees by the ‘hypothetical monopolist’ were offset by a commensurate decrease in cardholder fees or an increase in cardholder benefits, supra-competitive returns would not follow.”
He goes on to caution that in a two-sided market one must be careful in applying the SSNIP test and explains the issue in the context of payment cards:
“Accordingly, one must be careful applying the hypothetical SSNIP test in a two-sided market. In particular, one cannot draw an inference of market power simply from past increases in interchange rates relative to processing costs on the merchant side of the market, or the fact that such increases led few if any merchants to drop out of the Visa and MasterCard networks.”
One issue is that an increase in price to one side could make the product more valuable to members on the other side.
“In a two-sided market, in theory, merchants might accept increases in interchange not because of the lack of actual or potential competition from other payment mechanisms, but because interchange increases are accompanied by increased benefits to cardholders that make them more likely to use credit cards, so that higher merchant discounts are offset by increased sales.”
The plaintiffs’ experts conducted an analysis of total prices that responded to these two-sided issues. Professor Sykes found in this case:
“[P]laintiffs have provided a reasonable basis for concluding that a hypothetical monopolist of general purpose credit and charge card services could impose a SSNIP, based on a property conceptualized two-sided or total price.”
The Existence of Anticompetitive Effects
Professor Sykes continued his emphasis on the total price in his analysis of anticompetitive effects. He found that the proper analysis of welfare effects should account for the effect on both sides of the two-sided platforms.
To begin with he recognized the important role of skewed pricing structures in two-sided markets:
“[A]n important body of theoretical work on two-sided markets suggests that it may be socially desirable for prices to be higher on the side of the market that is less price sensitive.”
In the case of general purpose payment cards:
“If merchants are less price sensitive than consumers, this body of theory implies that economic welfare can be enhanced if merchants bear a substantial proportion of the total costs incurred by credit card issuers (through a mechanism such as interchange). A corollary is that socially desirable interchange fees need not be tied to the costs of processing merchant transactions.”
The problem with the plaintiffs case is that they would have difficulty establishing that the practice of setting the interchange fee actually reduced consumer welfare properly understood as consisting of the welfare of both the merchant and the consumer side.
“If the theoretical literature offers one robust conclusion, however, it is that the determinants of socially optimal interchange rates are complex and dependent on a variety of subtle factors… There is little basis for believing that socially optimal (or competitively determined) interchange fees are zero, for example, or some other amount that may have been chosen by any particular set of national regulators.”
One of the most interesting aspects of Professor Sykes’s discussion involved the calculation of damages in part because this topic has not been addressed much in the literature an the antitrust of two-sided platforms. He is sympathetic to the view that, at least as a matter of economics, damages claimed by one group of customers to a two-sided platform should be offset by gains to the other group of customers. He assumes in discussing damages that there has been a finding that behavior is anticompetitive.
“From a purely economic standpoint, however, an argument can be made for focusing on the two-sided or total price as the better way to conceptualize any overcharge, as it captures the extent to which all affected parties in the aggregate have been harmed by any anticompetitive practices…. To the degree that increases in interchange fees result in increased cardholder benefits, reductions in interchange may have the opposite effect, so that any benefits to merchants from reduced interchange fees might be offset in whole or in part by a loss to cardholders.”
Professor Sykes does not go into the reasoning here. However, if the purpose of damages is to deter companies from engaging in anticompetitive behavior then the proper measure of damages should be related to the total reduction in consumer welfare caused by that behavior. To put another way, given that two different sets of behavior by a platform are anticompetitive, courts should impose higher damages on the behavior that causes a greater harm to the total welfare of both groups of customers.
Relevance to Other Platform Cases
Professor Sykes’ memorandum to the court in the interchange fee litigation emphasizes the importance of taking the economics of two-sided platforms seriously in antitrust cases. He showed the importance of two-sided considerations in the assessment of market definition, liability, and damages.
There is nothing controversial in the issues that Professor Sykes raised in his memorandum although, of course, other economists might have viewed some of the arguments and evidence differently. As I noted in my remarks at the ABA meetings earlier this year there is now a consensus among economists on the general economic principles concerning two-sided platforms. Going forward, I would expect that there will be an increasing number of cases in which the economists for both plaintiffs and defendants adopt two-sided analyses. Professional economists really cannot defend apply one-sided models to two-sided platforms. The debate going forward will concern how these economic principles should be incorporated into an antitrust case law that has largely taken shape before the development of the two-sided literature.
Evans, David S., The Consensus Among Economists on Multisided Platforms and its Implications for Excluding Evidence that Ignores It (April 13, 2013). Available at < ahref=”http://ssrn.com/abstract=2249817″ target=”_blank”>SSRN
Evans, David S. et al., PLATFORM ECONOMICS: ESSAYS ON MULTI-SIDED BUSINESSES, Competition Policy International, 2011. Available at SSRN
Evans, David S. and Schmalensee, Richard, The Antitrust Analysis of Multi-Sided Platform Businesses (January 30, 2013). Roger Blair and Daniel Sokol, eds., Oxford Handbook on International Antitrust Economics, Oxford University Press, Forthcoming; University of Chicago Institute for Law & Economics Olin Research Paper No. 623. Available at SSRN
Evans, David S. et al., INTERCHANGE FEES: THE ECONOMICS AND REGULATION OF WHAT MERCHANTS PAY FOR CARDS, Competition Policy International, 2011. Available at SSRN