Further Thoughts on Critical Loss

In a recent issue of The Antitrust Source, David Scheffman and Joseph Simons presented their views on critical loss, responding to our earlier article and a closely related article by Daniel P. O'Brien and Abraham L. Wickelgren. 1 In the interest of advancing the debate on the use of critical loss in merger analysis, and to avoid any unnecessary confusion, we take this opportunity to respond to—and correct—Scheffman and Simons. We and they are both concerned with comparing the Actual Loss and Critical Loss faced by a hypothetical monopolist as part of a market definition exercise. In our earlier article, we showed that high observed margins strongly suggest that the hypothetical monopolist would face a small Actual Loss (i.e., demand would be unresponsive to a price increase). Scheffman and Simons assert that the process by which firms set prices is so complex and poorly understood that any such inference is unwarranted except in very special cases. As two professors of business strategy who have worked extensively with private clients as well as served in the Antitrust Division of the Department of Justice, we are well aware that firms often take many factors into account in setting prices, and actual pricing is more complex than can be explained fully by any simple model. However, we believe that the application of powerful economic logic combined with a careful examination of the facts in each case can allow us to make considerable headway in defining appropriate relevant markets and avoiding serious mistakes. More specifically: ● Scheffman and Simons fail to appreciate the generality of the analysis that leads to an equilibrium relationship in which high margins indicate low Actual Loss. As we show below, this analysis can readily incorporate strategic reactions by rival suppliers and applies perfectly well to bidding and bargaining situations. ● The analysis is not completely general, of course, and there are situations in which the equilibrium relationship between margins and demand elasticity is more complicated than in the fundamental model. Scheffman and Simons point to the kinked demand model of oligopoly as an example where they claim our methods break down. However, as we show below, the fundamental logic that we developed in our earlier article still applies to that setting. This example illustrates a more general point: it is not enough to say that the world is complicated. Before abandoning fundamental economic logic, one must identify and examine …