In the past weeks, the chatter surrounding a possible FTC antitrust case against Google has risen in volume, thanks largely to the FTC’s hiring of litigator Beth Wilkinson. The question remains, however, what this aggressive move portends and, more importantly, why the FTC is taking it.
It is worth noting at the outset that, as far as I know, Wilkinson has no antitrust experience; she is a litigator. Now, there’s nothing wrong with an agency enlisting a hired gun to help litigate its cases, but when the hired gun is not hired for her substantive expertise but rather her ability to persuade, it perhaps suggests something about the strength of the agency’s case.
It’s reading tea leaves (a time-honored, if flawed, DC practice), but Wilkinson’s hiring suggests to me that the FTC views its case as one that will require some serious rhetorical handling in order to win. While on its Sherman Act Section 2 merits that would be true anyway, it also suggests to me that the FTC intends to use the case as an opportunity to push – and seek court approval for – the ambitious plans of some of the Commissioners to expand the agency’s powers under Section 5 of the FTC Act. This would be a costly mistake for consumers.
Last year, in an interview with Global Competition Review, FTC Chairman Leibowitz was asked whether the agency was “investigating the online search market.” He declined to answer directly but instead offered this suggestive comment:
What I can say is that one of the commission’s priorities is to find a pure Section Five case under unfair methods of competition. Everyone acknowledges that Congress gave us much more jurisdiction than just antitrust. And I go back to this because at some point if and when, say, a large technology company acknowledges an investigation by the FTC, we can use both our unfair or deceptive acts or practice authority and our unfair methods of competition authority to investigate the same or similar unfair competitive behavior . . . .
Commissioner Rosch has likewise suggested that Section 5 could and should be expanded, precisely to reach activity that would be unreachable under current Section 2 standards. The effort to expand the FTC’s antitrust enforcement under Section 5, and to write out the jurisprudential standards of Section 2, is a troubling one.
Following Sherman Act jurisprudence, traditionally the FTC has understood (and courts have demanded) that antitrust enforcement under Section 5 (as a technical matter, the FTC does not directly enforce Section 2 of the Sherman Act but instead enforces the Act via its Section 5 authority) requires demonstrable consumer harm to apply. But this latest effort reveals an agency pursuing an interpretation of Section 5 that would give it unprecedented and largely-unchecked authority. In particular, the definition of “unfair” competition wouldn’t be confined to the traditional antitrust measures—reduction in output or an output-reducing increase in price—but could expand to, well, just about whatever the agency deems improper.
Most problematically, Commissioner Rosch has suggested that Section Five could address conduct that has the effect of “reducing consumer choice” without requiring any evidence that conduct actually reduces consumer welfare—a theory that only a vanishingly few commentators (essentially one law professor and one FTC lawyer have written the entire body of scholarship on this topic) support. Troublingly, “reducing consumer choice” seems to be a euphemism for “harm to competitors, not competition,” where the reduction in choice is the reduction of choice of competitors who may be put out of business by a competitor’s conduct.
Under Section 2 standards, the FTC would have a tough time winning its case. This is because the agency doesn’t seem to have a theory of harm that reaches consumers—and none of Google’s competitors that have been stoking the flames has offered one. Instead, all of the propounded theories turn on harm to competitors. But the U.S. has a long tradition of resisting enforcement based on harm to competitors without a showing of harm to consumers. If all that were required were harm to competitors, then all pro-competitive conduct would be actionable under the antitrust laws; for what is the aim and effect of competition if not the besting of one’s competitors? The competitive process is by definition one that can “reduce consumer choice.” This is why the great economist Joseph Schumpeter famously called the competitive process one of “creative destruction.”
In fact, the theoretical case against Google depends entirely on the ways it may have harmed certain competitors rather than on any evidence of harm to consumer welfare. For example, Google’s implementation and placement within its organic search results of its own shopping results is alleged to make it difficult for competing product-specific search sites (like Nextag or Amazon, for example) to reach Google’s users. Leaving aside the weakness of the factual allegation (I challenge you to perform a search for a product on Google that doesn’t offer up a mix of retailers, manufacturers, review sites and multiple product search engine results on the first page), it is hard to see how consumers are harmed here.
On the one hand, users have easy access to competing sites directly from their browser’s address bar and, increasingly importantly, to more persuasive product reviews from friends and colleagues via social media. In this way even the basic factual predicate is faulty, and it’s not even clear that consumer choice itself is reduced if Nextag is absent from Google searches, as the site can be reached by, among other things, links from reviews, links from friends on social media, other general search engines, and every browser address bar.
On the other hand, users are by no means foreclosed from access to actual products (and there is no evidence that I know of that consumer prices or supply are in any way affected) if any particular product search engine doesn’t appear in the top results. Placement of Google’s own product search results in fact streamlines consumers’ access, and Google’s comprehensive and effective search engine ensures that its shopping results are probably better than anyone else’s anyway. The same is true for travel searches, maps, and the range of other complained-of results. Flight information and reservations, location information and maps are widely available online and off through myriad sources other than Google.
The bottom line is that harm to competitors is at least as consistent with pro-competitive as with anti-competitive conduct, and simply counting the number of firms offering competing choices to consumers that happen to appear in the top few Google search results is no way to infer actual consumer harm.
One of the most important shifts in antitrust over the past 30 years has been the move away from indirect and unreliable proxies of consumer harm toward a more direct, effects-based analysis. Like the now archaic focus on market concentration in the structure-conduct-performance framework at the core of “old” merger analysis, the consumer choice framework substitutes an indirect and deeply flawed proxy for consumer welfare for assessment of economically relevant economic effects. By focusing on the number of choices, the analysis shifts attention to the wrong question.
The fundamental question from an antitrust perspective is whether consumer choice is a better predictor of consumer outcomes than current tools allow. There doesn’t appear to be anything in economic theory to suggest that it would be. Instead, it reduces competitive analysis to a single attribute of market structure and appears susceptible to interpretations that would sacrifice a meaningful measure of consumer welfare (incorporating assessment of price, quality, variety, innovation and other amenities) on economically unsound grounds. It is also not the law.
Commissioner Rosch has suggested that the Supreme Court in its 2007 Leegin decision provided a green light for consumer-choice-reducing antitrust theories without a showing of traditional (output-reducing) harm. But as Josh pointed out, the Ninth Circuit has held (in last year’s Brantley v. NBC Universal decision, which Thom has also blogged about here and here) that Leegin more accurately holds precisely the opposite, and coupled with the Court’s 2006 Independent Ink decision, seems clearly to restrict, rather than authorize, a consumer choice claim:
The Supreme Court has noted that both [reduced choice and increased prices] are “fully consistent with a free, competitive market,” [citing Independent Ink] and are therefore insufficient to establish an injury to competition. Thus even vertical agreements that prohibit retail price reductions and result in higher consumer prices . . . are not unlawful absent a further showing of anticompetitive conduct [citing Leegin].
Modern antitrust analysis, both in scholarship and in the courts, quite properly rejects the reductive and unsupported sort of theories that would undergird a Section 5 case against Google. That the FTC might have a better chance of winning a Section 5 case, unmoored from the economically sound limitations of Section 2 jurisprudence, is no reason for it to pursue such a case. Quite the opposite: When consumer welfare is disregarded for the sake of the agency’s power, it ceases to further its mandate. No doubt Beth Wilkinson could help make the rhetorical argument for a Section 5 case against Google based on a tenuous consumer choice theory. But economic substance, not self-aggrandizement by rhetoric, should guide the agency. Competition and consumers are dramatically ill-served by the latter.
Full disclosure: I worked briefly with Beth Wilkinson at Latham and Watkins. Further full disclosure: The International Center for Law and Economics, of which I am the Executive Director, has received support to make research grants from Google, among many other companies and individuals.
[Cross-posted at Forbes]