Professor Tim Wu of Columbia University, who invented the term “net neutrality,” is a highly regarded expert on the internet and a former adviser to the Federal Trade Commission, which enforces federal antitrust laws. He is a recognized authority on the markets for internet services. This week (June 29, 2015) he opined that Google is engaged in anti-competitive conduct because its own platform fails to provide appropriate placement of search results yielded by a rival search engine called “Yelp.” Not coincidentally, Yelp is paying the worthy professor undisclosed fees for his opinion on this very topic.

According to Professor Wu, Yelp’s search engine provides the most helpful, relevant results in response to various kinds of queries, particularly those that concern businesses that provide services to personal consumers and households. When these queries are run on Google’s platform, however, Google first provides its own inferior responses and thereafter provides links to Yelp’s website, where the supposedly superior, more useful results can be viewed. Professor Wu therefore finds that Google’s service is causing harm to consumer welfare by failing to list the Yelp links before the less helpful findings generated by its own search algorithms, and on this basis he concludes that Google has somehow committed an arguable violation of antitrust law. I believe that he is mistaken.

Before proceeding with the antitrust critique, I observe that the best remedy to this supposed problem is as simple as they come: Customers disaffected with Google’s inferior search results need only go directly to Yelp’s site to run the same searches and receive the supposedly superior responses. Problem solved, and marathon antitrust litigation averted finally and forever.

I also add the following: It is not clear that Yelp provides better results for the queries in question. Professor Wu’s assertion of the point appears to depend on shaky premises. But even if this point is assumed true, his challenge to Google’s practices likely cannot succeed under the antitrust laws of the United States. The proper remedy would instead lie in more competition, rather than a costly antitrust intervention. If Yelp is truly better for certain kinds of searches, then it should make its case to the public and convince us that its search engine is better than Google’s for various kinds of searches, and that we should go directly to its site for these searches. It would then be the customers who decide which search engine they prefer for the searches in question. If Yelp is right, it likely would win this battle and gain the preference of loyal customers who appreciated its service. It costs nothing in money or effort to switch from Google’s search engine to Yelp’s.

This approach to the matter is called “competition on the merits,” while Professor Wu’s special pleading for Yelp is mere “lobbying for a marketplace advantage by seeking antitrust intervention on specious grounds.” Regardless, the special pleading faces insuperable obstacles.

There is case law already directly on point that settles the question. The Supreme Court decided the issue, or at least established the terms of the debate, in two landmark cases: Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 U.S. 585, 105 S. Ct. 2847 (1985) and Verizon Commc’ns Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 U.S. 398, 124 S. Ct. 872 (2004). In Aspen, the high court held that a skiing resort in Aspen, Colorado had committed unlawful monopolization in violation of Section 2 of the Sherman Act by refusing to continue a cross-ticketing arrangement on commercially reasonable terms with its smaller competitor. Rather, the skiing monopolist discontinued the arrangement and lost profits in direct consequence, doing so in order to destroy its smaller, vulnerable rival and emerge as the only remaining skiing resort of its kind in Aspen.

In Verizon, the Supreme Court observed that Aspen set forth a narrow exception to the general rule that any firm, if acting alone, may rightly refuse to do business with any other firm. See Verizon, supra, 540 U.S. at 399, 124 S. Ct. at 874 (“Aspen is at or near the outer boundary of § 2 liability….”) The Supreme Court thus affirmed in Verizon the general rule on refusals-to-deal, which is that every firm enjoys nearly unqualified liberty to refuse to do business with other commercial firms. Broadly speaking, unilateral refusals-to-deal are lawful, even when implemented by a monopoly firm. Only on rare occasion must a monopoly firm collaborate with its direct competitor. There is a therefore a rebuttable presumption that a monopoly firm’s unilateral refusal to do business with a direct competitor does not constitute an antitrust offense. There are exceptions to the general rule, and therefore the presumption is rebuttable, but the exceptions are narrow and rarely afforded. See id. 

This means that presumptively Google cannot be obliged to give prominent placement to search results generated by a rival’s search engine. This presumption can be overcome only if Google’s present refusal to show its rival’s search results departs from its past practice and appears calculated to consolidate its own monopoly in the market for search engines by snuffing out the promise of a threatening rival; even then, the affected competitor must show that Google lacks any reasonable pro-competitive justification for its new practice.

More broadly, the law on refusals-to-deal is pretty well established. The general rule is that every firm can do business with those whom it wishes to serve, and may refuse to do business with anyone else for any reason at all, except for an unlawful reason, such as unlawful discrimination or a violation of the antitrust laws. The laws against discrimination are sweeping and broad, and they are not an antitrust challenge at all and therefore lie beyond the scope of this post. As for the antitrust laws, a refusal-to-deal violates them only under very narrow, well-established  circumstances:

1. Unlawful Group Boycotts. The first antitrust challenge might arise under the rule against “group boycotts.” An unlawful group boycott occurs when two or more firms jointly refuse to make purchases from specified suppliers or to make sales to specified customers, and only when they do so in order to exclude the targeted suppliers or customers from an entire line of commerce. This challenge can be made under Section 1 of the Sherman Act, California’s Cartwright Act, and many other state antitrust statutes.

2. The Essential Facilities Doctrine. The second antitrust challenge might arise when one firm controls an essential facility, such as an electricity grid, and it refuses to give potential rivals access to this facility in order to prevent them from competing against it. The conduct arises to the level of an antitrust offense only if the firm that refuses the access is thereby able to maintain a monopoly position in a downstream market and avert all possible competition in this market. Such a challenge can be made under Section 2 of the Sherman Act.

3. Cessation of Mutually Profitable Collaboration. Lastly, under specified circumstances a dominant or monopoly firm cannot discontinue an established course of dealing with a direct competitor when it does so merely to run the competitor out of business so that it can consolidate or perpetuate a monopoly position. Such a challenge can likewise be made under Section 2 of the Sherman Act.

So much for the exceptions to the general rule. Otherwise, a refusal-to-deal does not give rise to an offense under our antitrust laws. Yet Professor Wu’s challenge concerns a refusal-to-deal — Google’s alleged refusal to give appropriately prominent placement to Yelp’s search results for certain kinds of searches.

This challenge to me seems doomed. As explained above, a firm such as Google, which holds a monopoly in the search engine markets, owes no duty to provide comfort to its rivals or display their search results on its platform, much less give these results priority under pain of being condemned for antitrust offenses. Rather, there exists a rebuttable presumption that Google owes no obligation to present Yelp’s search findings anywhere on its site, much less give prominent mention to them. If Yelp wishes to challenge Google’s refusal to give this prominent placement on antitrust grounds, it must first acknowledge the above law and then explain why its particular challenge is sufficient to overcome the rebuttable presumption that Google has acted lawfully by not listing its search results in the manner that Yelp would prefer. Unless I am missing something, Yelp has failed to do so or even to state why its challenge should receive any hearing at all.

Such a challenge has at best a dim prospect of success only if the following showing can be convincingly made: That Google previously collaborated with Yelp by prominently displaying links to Yelp’s search results among Google’s own search findings; that Google has stopped doing so by changing its search algorithms specifically to undermine Yelp and consolidate its own monopoly position as the dominant provider of internet searches; that by this practice it has diminished its own service in order to undermine Yelp; and that by so acting it has effectually consolidated its monopoly position in the search engine markets. If Yelp can make its challenge along these lines, then there is a small, fleeting prospect that it might succeed, but I expect that any such challenge would be very problematic, if not impossible to prove. I suggest that Yelp not bother and instead strive to compete better by making its case to the public that its search engine is better for certain kinds of searches. If customers are convinced, they will proceed directly to Yelp’s site and conduct their searches there. Perhaps that was the real point of Professor Wu’s statement — to draw attention to Yelp’s claim that its own search engine provides better information than does Google for queries about businesses that serve consumers.

My above analysis does not mean that Google is entirely in the clear. On the contrary, it faces withering scrutiny in Europe right now for its alleged practice of diverting its search customers to its own businesses rather than to sites owned by others that its own algorithms would otherwise suggest. This challenge is different and might succeed under European competition law. The challenge is that, first, Google is the dominant or monopoly provider of search engine queries; and, second, when its customers ask about particular services, Google’s site does not direct them to the most informative or “relevant” sites, but instead diverts them to its own businesses. Even if this charge is true (which is by no means established) it presents a case for European antitrust authorities, but not necessarily for aggrieved American consumers or businesses.

Under US law, it must be shown that Google not only employed such practices (which arguably constitute “anti-competitive conduct”), but also that by these practices it successfully monopolized one or more targeted markets, or that it has attempted to do so, specifically intends to do so, and will likely succeed unless there is an antitrust intervention. In contrast, under European law it need only be shown that Google has a dominant position in the markets for search results (search engine markets), and that it has abused this position in order to gain a foothold in other markets.

Some explanation is in order.

A firm commits a violation of Section 2 of the Sherman Act, if it holds a monopoly position in one market and abuses or “leverages” this monopoly in effort to establish a second monopoly position in a another market. Conduct of this kind can be condemned as attempted monopolization or monopolization in violation of Section 2 of the Sherman Act. See Alaska Airlines, Inc. v. United Airlines, Inc., 948 F.2d 536, 547 (9th Cir. 1991) (In a “leveraging” case, “a plaintiff cannot establish a violation of Section 2 without proving that the defendant used its monopoly power in one market to obtain, or attempt to attain, a monopoly in the downstream, or leveraged, market.”)

Under this doctrine, a challenge to Google’s practices could succeed only upon a showing that (1) Google is a monopoly provider of search engine services; (2) Google skews its search results in order to direct consumers to its own services in other markets (i.e., services other than search engine services); and (3) in this manner Google has attempted to monopolize one or more of these other target markets, or has succeeded in doing so. The challenge would depend on showing the necessary elements of attempted or actual monopolization of the target market or markets, such as markets for bargain travel services. Google’s mere “leveraging” of its monopoly does not constitute an antitrust offense under the Sherman Act if it merely helps Google to make sales in other markets. It rises to the level of a Sherman Act violation only if it is used in furtherance of a scheme to monopolize a targeted market. Even so, this leveraging can arguably be condemned by our Federal Trade Commission under the broad powers vested in it under Section 5 of the FTC Act. This result nearly occurred. The FTC investigated Google’s alleged steering of its search-engine customers to its own businesses rather than to the most informative sites. At the close of the investigation, its staff formally recommended that the FTC bring proceedings against Google under Section 5 of the FTC Act. Unusually, the FTC commissioners rejected these recommendations.

In Europe, however, the same kind of leveraging exposes Google to antitrust liability. The European Competition Commission has already raised the challenge under the amorphous and expansive doctrine called “abuse of dominant position,” which does not exist under United States antitrust law. The European challenge might prove more problematic for Google, and all the more so because in Europe there seems to be a prevailing bias these days against technology companies from the Silicon Valley – a circumstance that cannot help Google in this matter.

In the meantime, Professor Wu’s challenge, as it has been stated, likely cannot go very far in the US.