I am participating in an online “debate” at the American Constitution Society with Professor Ben Edelman. The debate consists of an opening statement and concluding responses to be posted later in the week. Professor Edelman’s opening statement is here. I am cross-posting my opening statement here at TOTM, and will cross-post my closing statement later this week.
The theoretical antitrust case against Google reflects a troubling disconnect between the state of our technology and the state of our antitrust economics. Google’s is a 2011 high tech market being condemned by 1960s economics. Of primary concern (although there are a lot of things to be concerned about, and my paper with Geoffrey Manne, “If Search Neutrality Is the Answer, What’s the Question?,” canvasses the problems in much more detail) is the treatment of so-called search bias (whereby Google’s ownership and alleged preference for its own content relative to rivals’ is claimed to be anticompetitive) and the outsized importance given to complaints by competitors and individual web pages rather than consumer welfare in condemning this bias.
The recent political theater in the Senate’s hearings on Google displayed these problems prominently, with the first half of the hearing dedicated to Senators questioning Google’s Eric Schmidt about search bias and the second half dedicated to testimony from and about competitors and individual websites allegedly harmed by Google. Very little, if any, attention was paid to the underlying economics of search technology, consumer preferences, and the ultimate impact of differentiation in search rankings upon consumers.
So what is the alleged problem? Well, in the first place, the claim is that there is bias. Proving that bias exists — that Google favors its own maps over MapQuest’s, for example — would be a necessary precondition for proving that the conduct causes anticompetitive harm, but let us be clear that the existence of bias alone is not sufficient to show competitive harm, nor is it even particularly interesting, at least viewed through the lens of modern antitrust economics.
In fact, economists have known for a very long time that favoring one’s own content — a form of “vertical” arrangement whereby the firm produces (and favors) both a product and one of its inputs — is generally procompetitive. Vertically integrated firms may “bias” their own content in ways that increase output, just as other firms may do so by arrangement with others. Economists since Nobel Laureate Ronald Coase have known — and have been reminded by Klein, Crawford & Alchian, as well as Nobel Laureate Oliver Williamson and many others — that firms may achieve by contract anything they could do within the boundaries of the firm. The point is that, in the economics literature, it is well known that self-promotion in a vertical relationship can be either efficient or anticompetitive depending on the circumstances of the situation. It is never presumptively problematic. In fact, the empirical literature suggests that such relationships are almost always procompetitive and that restrictions imposed upon the abilities of firms to enter them generally reduce consumer welfare. Procompetitive vertical integration is the rule; the rare exception (and the exception relevant to antitrust analysis) is the use of vertical arrangements to harm not just individual competitors, but competition, thus reducing consumer welfare.
One has to go back to the antitrust economics of the 1960s to find a literature — and a jurisprudence — espousing the notion that “bias” alone is inherently an antitrust problem. This is why it is so disconcerting to find academics, politicians, and policy wags promoting such theories today on the basis that this favoritism harms competitors. The relevant antitrust question is not whether there is bias but whether that bias is efficient. Evidence that other search engines with much smaller market shares, and certainly without any market power, exhibit similar bias suggests that the practice certainly has some efficiency justifications. Ignoring that possibility ignores nearly a half century of economic theory and empirical evidence.
It adds insult to injury to point to harm borne by competitors as justification for antitrust enforcement already built upon outdated, discredited economic notions. The standard in antitrust jurisprudence (and antitrust economics) is harm to consumers. When a monopolist restricts output and prices go up, harming consumers, it is a harm potentially cognizable by antitrust; but when Safeway brands, sells, and promotes its own products and the only identifiable harm is that Kraft sells less macaroni and cheese, it is not.
Understanding the competitive economics of vertical integration and vertical contractual arrangements is difficult because there are generally both anticompetitive and procompetitive theories of the conduct. One must be very careful with the facts in these cases to avoid conflating harm to rivals arising from competition on the merits with harm to competition arising out of exclusionary conduct. Misapplication of even this nuanced approach can generate significant consumer harm by prohibiting efficient, pro-consumer conduct that is wrongly determined to be the opposite and by reducing incentives for other firms to take risks and innovate for fear that they, too, will be wrongly condemned.
Professor Edelman has been prominent among Google’s critics calling for antitrust intervention. Yet, unfortunately, he too has demonstrated a surprising inattention to this complexity and its very real anti-consumer consequences. In an interview in Politico (login required), he suggests that we should simply prevent Google from vertically integrating:
I don’t think it’s out of the question given the complexity of what Google has built and its persistence in entering adjacent, ancillary markets. A much simpler approach, if you like things that are simple, would be to disallow Google from entering these adjacent markets. OK, you want to be dominant in search? Stay out of the vertical business, stay out of content.
This sort of thinking implies that the harm suffered by competing content providers justifies preventing Google from adopting an entire class of common business relationships on the implicit assumption that competitor harm is relevant to antitrust economics and the ban on vertical integration is essentially costless. Neither is true. U.S. antitrust law requires a demonstration that consumers — not just rivals — will be harmed by a challenged practice. But consumers’ interests are absent from this assessment on both sides — on the one hand by adopting harm to competitors rather than harm to consumers as a relevant antitrust standard and on the other by ignoring the hidden harm to consumers from blithely constraining potentially efficient business conduct.
Actual, measurable competitive effects are what matters for modern antitrust analysis, not presumptions about competitive consequences derived from the structure of a firm or harm to its competitors. Unfortunately for its critics, in Google’s world, prices to consumers are zero, there is a remarkable amount of investment and innovation (not only from Google but also from competitors like Bing, Blekko, Expedia, and others), consumers are happy, and, significantly, Google is far less dominant than critics and senators suggest. Facebook is now the most visited page on the Internet. Many online marketers no longer view Google as the standard, but are instead increasingly looking to social media (like Facebook) as the key to advertisers’ success in attracting eyeballs on the Internet. And at the end of the day, competition really is “just a click away” (OK, a few letters away) as Google has no control over users’ ability to employ other search engines, use other sources of information, or simply directly access content, all by typing a different URL into a browser.
Finally, even if there is a concern, there is the problem of what to do about it. Even if Google’s critics were to demonstrate that bias is anticompetitive, it is relevant to any analysis that bias is hard to identify, that there is considerable disagreement among users over whether it is problematic in any given instance, that a remedy would be difficult to design and harder to enforce, and that the costs of being wrong are significant.
Tom Barnett — who was formerly in charge of the Antitrust Division at the DOJ and who now represents Expedia and vociferously criticizes Google (including at the Senate hearings in September) — has himself made this point, observing that:
No institution that enforces the antitrust laws is omniscient or perfect. Among other things, antitrust enforcement agencies and courts lack perfect information about pertinent facts, including the impact of particular conduct on consumer welfare . . . . We face the risk of condemning conduct that is not harmful to competition . . . and the risk of failing to condemn conduct that does harm competition . . .
Condemning Google for developing Google Maps as a better form of search result than its original “ten blue links” reflects retrograde economics and a strange and costly preference for the status quo over innovation. Doing so because it harms a competitor turns conventional antitrust analysis on its head with consumers bearing the cost in terms of reduced innovation and satisfaction.
