Every serious student of industrial economics should read John Sutton's Sunk Costs and Market Structure: Price Competition, Advertising, and the Evolution of Concentration. This massive (577 pages overall), masterful, innovative, and thought-provoking study sets out to derive and test robust implications of game-theoretic models of free-entry equilibrium for the dependence of minimum viable levels of seller concentration on market size, set-up costs, and other factors. Empirical analysis is based on an extraordinary set of detailed, often fascinating histories of 20 food and drink industries in the six largest industrial economies. Formal econometric techniques and international comparisons are carefully and cleverly exploited.

The analysis in Sunk Costs distinguishes sharply between what I will refer to as type I and type II markets; advertising competition is important only in the latter. It is argued that models of type I markets generally imply that the lowest viable concentration declines to zero as market size increases, while models of type II markets generally imply that this lower bound converges to a strictly positive value. Empirically, six industries with low advertising–sales ratios are taken to be of type I, while the remaining 14 advertising-intensive industries are taken to be of type II. Cross-section relations between concentration and market size are estimated for these two groups of industries.