By Herbert Hovenkamp. Full text here.
The development of marginalist, or neoclassical, economics led to a fifty-year long crisis in competition theory. Given an industrial structure with sufficient fixed costs, competition always became “ruinous,” forcing firms to cut prices to marginal cost without sufficient revenue remaining to pay off investment. Early neoclassicists such as Alfred Marshall were not able to solve this problem, and as a result many economists were hostile toward the antitrust laws in the early decades of the twentieth century. The ruinous competition debate came to an abrupt end in the early 1930’s, when Joan Robinson and particularly Edward Chamberlin developed models that took product differentiation into account. The emergent theory of monopolistic competition came with its own problems, however—namely, “excessive” product variety and advertising, chronic excess capacity, and prices above short-run marginal cost. In sharp contrast to the ruinous competition model, the monopolistic competition model called for aggressive antitrust enforcement. This change of model largely explains the Roosevelt administration’s abrupt shift in antitrust policy between the First and Second New Deals. Only with John Maurice Clark’s theory of workable competition in 1940 and the Mason-Bain structure-conduct-performance paradigm developed in the 1950s did neoclassical competition theory begin to reach a new equilibrium that attempted to calibrate the amount and kind of competition policy necessary to produce satisfactory results in diverse markets. The subsequent debate between Harvard structuralism and the emergent Chicago School occurred largely within this paradigm.