A Test of Accounting Bias and Market Structure

Abstract

Empirical research in industrial economics generally supports the hypothesis that there is a positive relationship between market concentration and profitability.' In recent years, however, this evidence on the correlation between profits and concentration has been challenged2 on grounds of alleged deficiencies in the experimental design, measurement errors in census data, and in the accounting definitions of profits. The purpose of this paper is to test whether or not accounting bias could have affected the results of the previous profitability-concentration studies. There is an abundance of examples in the accounting literature which show that the use of different accounting procedures may produce different profit figures whether profits are measured as dollars or rates of return. For example, suppose all firms in concentrated industries treat research and development costs as assets and amortize these costs over several years, while firms in the unconcentrated industries treat research and development costs as an expense when incurred. In this case, the unconcentrated industries' stated asset values would be lower and accounting profits (measured as return on assets) would be systematically lower than those of the concentrated industries because of accounting rather than economic differences. Thus it is important to determine if concentrated and unconcentrated industries use different accounting principles. Our hypothesis is that if there are no significant differences in the choice of accounting procedures by firms in concentrated and unconcentrated industries, then ceteris paribus the difficulties inherent in accounting measurement will not affect the conclusions that can be drawn from the positive correlation between profit and concentration.