On the validity of accounting rates of return in cross-sectional analysis: Theory, evidence, and implications

This paper reports the results of an empirical examination of the relationship between firm accounting rate of return (ARR) and firm internal rate of return (IRR). The evidence presented shows that some of the analytically derived properties of the ARR-IRR relationship hold in a sample of U.S. manufacturing firms. This evidence can be interpreted as documenting the existence of a potentially important degree of measurement error in the ARR for a sample of actual firms and increases the credibility of those who have questioned the use of accounting rates of return as the dependent variable in cross-sectional studies of firm profitability.