Industry Structure, Market Rivalry, and Public Policy: A Comment

Abstract

A WIDELY accepted explanation for the positive association found between industry concentration and profitability has been that high seller concentration facilitates coordination of price and output decisions among industry members. 1 In a recent article in this journal, however, Harold Demsetz has suggested that the positive association may reflect merely the ability of some firms to operate more efficiently than others.2 Some firms, it is argued, have unusual luck or ability in exploiting scale economies, in organizing their resources, or in satisfying consumer demands. Because such firms -re more efficient than their competitors, they grow to dominate their industries, which in the process become more concentrated. Moreover, whether because the sources of the superior efficiencies are difficult to isolate or because accounting practices understate the value to the firms of their individual or combined inputs, the accounting statements of such efficient firms will be likely to show higher than normal rates of return. Thus, the argument concludes, highly concentrated industries appear to have higher than normal rates of return. Most of the existing structure-performance studies offer limited insight into whether the collusion or efficiency hypothesis is the more plausible. Studies relating industry profitability to industry concentration typically have incorporated profitability data that reflect primarily the performance of the leading firms in each industry. Where small samples have been used, the profitability data have been drawn primarily from the income statements and balance sheets of the largest firms in each industry. Where census data have been used, the price-cost margins of high concentration industries are necessarily dominated by the price-cost margins of the largest firms. Thus, the widely confirmed positive relationship between concentration and