May 5, 2003
"Class Warfare"--A Commentary by Prof. George L. Priest
(This essay was originally published in the May 5, 2003, edition of the Wall Street Journal.)
By George L. Priest, John M. Olin Professor of Law and Economics
The principal lesson of last week's $3 billion antitrust settlement by MasterCard and Visa is that, if a class can be certified that is massive enough, even defendants with a strong case can be bludgeoned into a huge payout.
Though there had been no trial on the merits -- indeed, not a single witness had testified -- and though the antitrust claims against them were highly questionable, MasterCard and Visa had no choice but to settle. Judge John Gleeson had certified a class of four million merchants -- led by Wal-Mart and Sears -- seeking damages that were estimated to be between $40 and $100 billion in a case to be tried to a Brooklyn jury. With potential damages at $100 billion, trebled under the Sherman Act, the Wal-Mart class needed only a 1% chance of victory for a $3 billion-settlement to make sense. MasterCard and Visa could have had Oliver Wendell Holmes and Clarence Darrow on their legal team and it still would have been foolish to try the case.
Wal-Mart and the merchants' basic complaint was that Visa's and MasterCard's interchange fee -- the amount discounted from the retail purchase for use of the card -- was too high for purchases made with debit cards. They claimed that it was an antitrust violation for the two associations to maintain the fee by enforcing their "Honor All Cards" rule, requiring a merchant who accepted a MasterCard or Visa credit card to also accept their debit cards, and asserted that the rule was a form of tying arrangement.
The plaintiffs had obtained certification of the four-million-member class by convincing Judge Gleeson to accept a radically simplified theory of antitrust damages. According to Wal-Mart, damages could be measured on a class-wide basis simply by comparing the interchange fee that the defendants had actually charged to a hypothetical fee that the plaintiffs' economic expert estimated would have been charged in the absence of the Honor All Cards rule. MasterCard and Visa had shown the unreality of the measure: It failed to account for differences in consumer use as banks changed their relative promotions of debit and credit cards based on the returns from interchange; it failed to account for the likely increase in card charges as debit interchange revenue declined; it failed to consider differences among the four million merchants in their proportions of debit and credit usage.
But Judge Gleeson brushed off these problems. And the Second Circuit, over a stinging dissent by Judge Dennis G. Jacobs, concluded that all that was necessary for class certification was to show that the plaintiffs' economic theory was "at least a 'colorable method' of proving" damages and that the economic methodology "was not fatally flawed," adding that, if necessary, Judge Gleeson could amend the damages calculation during trial.
In dissent, Judge Jacobs insisted that the court's standard was much too low, and accurately predicted that certification of the class alone would "coerce settlement." MasterCard and Visa sought Supreme Court review, but the Court declined, perhaps thinking that it could better evaluate the issues after trial, shutting its eyes to Judge Jacobs' sage understanding that, in a case of this magnitude, class certification is the only adjudication that a defendant will ever receive.
Wal-Mart's antitrust theory was weak, but given uncertainty in the law, it could not be immediately dismissed. The principal hope of the class appeared to be that Judge Gleeson would declare the alleged tying agreement illegal as a per se offense. But the per se offense of tying was last affirmed by the SupremeCourt in 1984, and in a 5-4 ruling at that. There has been substantial water under the dam in antitrust law over the past 19 years. Almost two years ago in the Microsoft case, the D.C. Circuit Court of Appeals rejected per se analysis of tying arrangements as too simplistic in network industries because it presumes that there are no efficiencies from the arrangement.
Without the per se rule, the trial would have had to evaluate whether the Honor All Cards rule helps or hurts consumers and thus the competitive process. Here the claim of violation is not so obvious. Industry requirements like the Honor All Cards rule are a commonplace in retail business. They are no different from McDonald's requiring its franchisees to offer the entire range of McDonald's products or auto manufacturers requiring their dealers to sell a full line of car models. Undoubtedly, hundreds of thousands -- perhaps millions -- of the plaintiff merchants enforced or complied with such policies themselves. Can we imagine that it is an antitrust violation for Wendy's to clamp down on a franchisee that wants to drop its chicken barbecue on a complaint about revenue?
In the payment industry, Honor All Cards is a rule adopted by every card system, including American Express, Discover, Diner's Club and the on-line debit networks that the merchants purport to prefer. This itself is proof of efficiency, but there are also good economic reasons underlying the practice. The point of establishing a unified card network, as opposed to a balkanized series of cards under different trademarks, is that both cardholders and merchants gain benefits where standardized cards are widely carried, used and accepted by merchants so that consumers can pay in any way they want.
The settlement will surely lead MasterCard's and Visa's debit interchange fees to decline. But that will hardly be the end of the story. Banks may reintroduce or increase consumer charges. Merchants may gain the ability to pick and choose among the cards they accept. But can we imagine that it is a better world where, though there is an American Express or MasterCard logo on the door, a consumer will not know whether the retailer only accepts AmEx corporate cards or refuses MasterCard debit?
Or, more accurately, can we imagine that these many payment card issues and their antitrust implications are best resolved to the benefit of society by the simple ruling that a plaintiff has satisfied the minimal procedural requirements for class certification? That was all the justice that our legal system afforded MasterCard and Visa in this litigation.
Mr. Priest teaches antitrust at Yale Law School. He has served as a consultant for Visa but was not involved in the litigation.