Wednesday, May 1, 2013
Posted by Spencer Weber Waller
A first rate law review article should be like a visit to the optometrist. Like a phoropter (the machine with the many different lens options), outstanding legal scholarship should ask the question “Is it clearer with this lens, or lens #2?” And so on, until the problem comes into the sharper focus and the right prescription is given to the patient.
By this measure and most other more traditional ways to judge legal scholarship, Hemphill and Wu have produced a first rate piece of scholarship. It is clear, exhaustive, rigorous in arguing and supporting its thesis, generous in case studies and applications in the real world, reasonable in considering contrary arguments, and persuasive in rebutting opposing views.
The particular setting that clarifies much of antitrust doctrine is the notion of parallel exclusion. According to the authors, “parallel exclusion is self-entrenching conduct, engaged in by multiple firms, that harms competition by limiting the competitive prospects of an existing or potential rival to the excluding firms.” (p.1189). Hemphill and Wu have convinced me that many of the types of parallel exclusion that they discuss are more durable and more harmful than normally acknowledged, particularly when used to exclude important types of innovation. They haven’t however entirely persuaded me that this is necessarily a path to revived enforcement of the Sherman and FTC Acts against shared monopoly. But they have made the case for an increased incorporation of parallel exclusion into less controversial areas of antitrust.
One of the key contributions of this article is addressing a disturbing trend in antitrust doctrine that the wide use of a practice in an industry (exclusive dealing contracts, bundling, RPM, whatever) is a sign of its pro-competitive value and a reason to refrain from aggressive enforcement. Hemphill and Wu have provided a critical counter-weight to these arguments and important support to the longstanding notion that 1) the exclusionary practices of dominant firms matter regardless of the behavior of the competitive fringe and 2) if all or most of the important firms in an industry engage in an exclusionary practice there is more, rather than less, cause for concern.
Their arguments support one of the consistent lines of antitrust doctrine going back to the Standard Stations case of the 1940s where the Supreme Court held that the extent of exclusive dealing contracts throughout the industry was a relevant consideration in deciding the legality of the exclusive dealing contracts of a leading firm with an individual share of 16% of the relevant market. Not surprisingly, Hemphill and Wu cite Standard Stations in several places in the article as one of many examples of non-collusive parallel exclusionary practices and the need to consider the full scope of the industry behavior in deciding exclusionary effects and any relevant offsetting efficiencies. Similarly, theories of parallel practices were also cited by the Supreme Court more recently in Leegin as one of a number of suggested rules of thumb for lower courts to use in applying the rule of reason in resale price maintenance cases going forward. Finally, even Frank Easterbrook in The Limits of Antitrust identified parallel industry practices as one of his key filters in identifying problematic anticompetitive conduct. And yet, several recent courts and commentators have drawn the opposite conclusion that wide-spread adoption of a particular practice makes the situation less problematic rather than more so. The D.C. Circuit in Microsoft thought the practice of fringe firms in bundling or tying new features into their operating systems cut in favor of applying a rule of reason, rather than the Supreme Court’s quasi per se rule, to such allegations. Earlier this year, former FTC Chairman Jon Leibowitz in his comments accepting Google’s commitment letter to end the competition investigation into alleged search bias noted: “Tellingly, Google’s search engine rivals engaged in many of the same product design choices that Google did, suggesting that this practice benefits consumers.”
There are two reasons why this type of reasoning is not sound antitrust policy. First, the exclusionary effects of a dominant firm’s practices simply are not the same as its rivals at the fringe of the industry, any more than the gravitational attraction of a black hole is the same as that of an asteroid. Dominance, like gravity, means that a consumer or customer may not be able to avoid the anticompetitive pull of a practice not in their interests. In contrast, a fringe firm’s use of the same practice tells one nothing of exclusionary effect by such firms, which is likely to be minute and easily avoidable where other options and models of doing business are available throughout the industry in question. Hemphill and Wu are at their best when explaining why the widespread use of an exclusionary practice make the matter of far greater, rather than lesser, antitrust concern. For widespread parallel industry practices, the exclusionary effect is magnified regardless of whether by agreement, interdependent decisions, or independent decision; consumer choice is limited; cheating more easily detected; profits need not necessarily be sacrificed; arrangements tend toward longer term stability; and longer term harms to competition are more likely. They recognize that any such identifiable harms may still be offset by legitimate efficiencies, but that is a question of fact rather than theory.
As a result, I highly recommend this article, its clarifying lens, and its prescriptions for more effective competition policy. Please make an appointment to revisit the issue in a year, but be sure to come back sooner if things still remain blurry.