Does the FTC have grounds to pursue a Section 5 case against Google? Comment by Bob Litan

Posted by Robert E. Litan (Kauffman Foundation and Brookings)

The FTC hasn’t enjoyed much luck over the years convincing the courts that various parties have committed standalone violations of Section 5 of the FTC Act, which punishes “unfair methods of competition” and “unfair and deceptive practices.” This string of losses occurred despite the Supreme Court’s broad language in FTC v. Indiana Federation of Dentists (1986), which states that Section 5 of the Act gives the Commission broader authority than what exists under the basic antitrust laws (Sections 1 and of 2 of the Sherman Act) to “stop in their incipiency acts which, when full blown, would violate those Acts.”

The large gap between this broad rhetoric and actual case law is not hard to understand. “Incipiency” is hard to define outside the merger context, where the DOJ and FTC have set forth relatively hard numerical guidelines for suggesting when mergers might violate Section 7 of the Clayton Act. And apart from the equivalent of common law fraud and misrepresentation, it is hard to know what “unfairness and deception” under Section 5 of the FTC Act actually mean.

These definitional problems notwithstanding, certain critics have charged that Google’s relatively recent entry beyond generalized search into “specialized” or “universal” search – which provides, among other things, price and product comparisons (like “Sony Camera” or “best airline fares from Kansas City to Dallas”) -- runs afoul of Section 5. Google asked me and Hal Singer to look into this claim, based solely on publicly available information, and see whether there is much to it. In a recent paper posted on SSRN, we concluded there isn’t. In brief, apart from the generic problems in bringing Section 5 cases just noted, here are the most notable allegations and why, in our opinion, they are without either legal or economic basis.

First, some have claimed that by placing its universal search results above those of independent search engines – like Kayak, Yelp, and so on – Google has acted “unfairly” and should somehow be punished, even perhaps by being prohibited from engaging in universal search altogether. That Google’s motives are anticompetitive is undermined by the fact that Google’s search rivals, notably Bing and Yahoo, engage in the very same behavior. Moreover, as antitrust scholars well know, the antitrust laws were designed to protect competition, not specific competitors. For the FTC to push Section 5 beyond this basic principle would open up unbounded inquiries that could threaten aggressive competition throughout the economy.

A second potential critique is that Google has engaged in something deceptive by seemingly reneging on previous statements in its IPO filings and its corporate mission statement that it wanted users to quickly leave its site, and now through universal search wants them to stay a little longer. To punish Google’s change in business model because it is seemingly inconsistent with statements made long ago would set a dangerous precedent: subjecting companies to potentially draconian remedies (in Google’s case, either prohibition of universal search or regulation of the order in which those search results appear) that clearly would chill innovation. People change their minds when they get new facts. John Maynard Keynes even once famously proclaimed that as a virtue. If that is the case for people, it should certainly be true for companies.

Finally, to prevail in a traditional antitrust tying or bundling claim (outside of Section 5), plaintiffs must establish that the two services are distinct antitrust products. Independent of the two foregoing allegations, proponents of antitrust scrutiny claim that Google’s universal search is somehow fundamentally different from the company’s original general search offering. This premise is wrong and the distinction between the two types of search is largely semantic. As antitrust scholar Einer Elhauge has pointed out, antitrust economics tells us that two seemingly different products are really a single one if consumers would not reassemble them in the absence of a combined offering. Because search users would not likely integrate universal search with general search if Google didn’t do it for them, the two services are best understood as a unified offering (and thus as constituting a single market).