Section 2 Symposium: Bruce Kobayashi on Predatory Pricing and the Relevant Measure of Cost

Bruce Kobayashi —  6 May 2009

kobayashiBruce Kobayashi is a Professor of Law at George Mason Law School.

Dimming the Court’s Brooke Group Bright Line Administrable Rule?

As noted in my earlier post, the Supreme Court’s Brooke Group rule is held out as the primary example of an administrable bright line rule aimed at controlling the costs of type I error. In practice, the Brooke Group above cost rule is not as bright as one might wish. The Achilles heel of the Brooke Group cost based rule is the failure to clarify what the relevant cost is.

Chapter 4 of the Section 2 report for the most part does a nice job of setting out the leading alternatives. The report notes that there is a broad consensus that prices above Average Total Cost (ATC) should be per se legal (Section 2 Report at 61). The report also discusses the measures preferred by Areeda-Turner, Marginal Cost (MC), and Average Variable Cost (AVC) as an administrable proxy for MC, and criticisms of these measures. The Report endorses Average Avoidable Costs (AAC), which includes both the variable and non sunk product-specific fixed costs of producing the incremental output as the preferred measure. AAC is the preferred measure because it correctly measures the avoidable cost of producing the incremental predatory output.

While Chapter 4 of the Section 2 Report provides a very useful review and analysis of the alternative cost measures, I thought the discussion simplified away several important issues. The nature of problems created as a result of this oversimplification can be illustrated by considering the numerical example used to illustrate the difference between the various cost measures (Section 2 report at 64). One problem is that the constant cost of producing the incremental output is constant, so that the AAC is equal to MC. Thus, the example fails to clearly illustrate the differences between the Areeda Turner preferred measure (MC) and the DOJ’s preferred measure (AAC). Moreover, the example suppresses several other important issues. For example, it measures the incremental output relative to the predating firm’s pre-entry output instead of measuring it relative to the but-for post entry output. The simplicity of the example has the advantage of being easier to understand, but it suppresses issues that make use of the AAC measure less administrable than an accounting measure such as AVC.

A more serious issue is the Report’s failure to clearly address the opportunity cost issue, a critical issue in the recent airline cases (U.S. v AMR Corp., 355 F.3d. 1109 (10 Cir. 2003) and Spirit Airlines v. Northwest Airlines, 431 F.3d. 917 (6th Cir. 2005)). In AMR, the DOJ’s position was that when an airplane is shifted from a profitable route (Route S) to expand capacity in the alleged predation route (Route P), avoidable costs for Route P should include the forgone profits from Route S as an opportunity cost. These costs would be added to the flight costs (cost of fuel, crew, passenger costs, etc.). The AMR court rejected inclusion of such forgone profits, but the Spirit court accepted forgone revenues as part of the incremental costs of expanding output (see Areeda & Hovenkamp, 2006 304-11).The report notes general agreement among panelists at the Section 2 hearings that opportunity costs should be included in the calculation of avoidable costs, but far less agreement with regard to whether lost inframarginal revenues should be considered. (Section 2 report at 66). Later, the Section 2 Report concludes “that consideration of foregone revenues [as part of avoidable costs] is neither appropriate nor likely to be administrable. The Department consequently will not consider the lost revenues on inframarginal sales as a cost when evaluating predatory pricing claims.” (Section 2 Report at 67). The first line suggests that the Department would not consider the forgone profits from Route S or the forgone inframarginal profits from Route P as part of avoidable costs for Route P. However, the second line suggests that the limitation applies only to consideration of reduced inframarginal profits for Route P, which would allow consideration of the forgone profits for Route S.

Given the Report’s clear statement that forgone inframarginal profits for Route P would not be considered, coupled with the Departments position in AMR that forgone profits for Route S should be part of avoidable costs, the second interpretation may be a plausible one. However, it does not seem to be an analytically consistent one. If one used a profit sacrifice test in predatory pricing cases, one would want to consider both the forgone profits for Route S and the forgone inframarginal profits for Route P in calculating whether the shift of capacity and altered prices constitute a profit sacrifice. However, one could conclude that use of such a test in predatory pricing cases would not be administrable. Given that the administrability of the forgone profit calculations for Route P and Route S would in general be similar (e.g., Route S is reduced from four flights a day to three, while Route P is increased from two flights to three), administrability concerns would suggest eliminating consideration of both the Route S and Route P forgone profits when evaluating predatory pricing claims. The DOJ approach would eliminate the latter on administrability grounds, but allow the former, which would generally present the same issues of administrability.

In general, the airline cases illustrate the potential complications encountered in applying the Brooke Group cost test. In addition to the issue of forgone profits, any incremental revenue calculations will have to account for the fact that passengers in hub and spoke systems will generate revenue by flying connecting segments, as well as the airlines’ complex yield management systems. In addition, narrow definition of markets that include distinct classes of flyers (e.g., the leisure versus business travel markets used in Spirit) requires that the courts address the difficult issue of how joint and common costs are to be allocated between these markets. All of these complicate the analysis and increase the number and complexity of the issues litigated in these cases. And while such developments may improve the economic analysis in the cases, one has to wonder whether these “advances” will “through the vagaries of administration, prove counterproductive, undermining the very economic ends they seek to serve” (Barry Wright Corp. v. ITT Grinnell Corp., 724 F.2d 227, 234 (1st Cir. 1983)). If so, the Department may be wise to adopt the unconditional version of its decision not to consider foregone profits in predatory pricing cases in order to preserve the administrability benefits of the Brooke Group rule.