Accounting Ratios and the Prediction of Failure: Some Behavioral Evidence

Researchers have investigated the predictive ability and behavioral impact of accounting information separately. For example, Beaver (1968) and Deakin (1972) examined the predictive ability of accounting information in the prediction of business failure, whereas Bruns (1966) and Hofstedt (1972) investigated the behavioral impact of accounting variations on decision making. Given that the predictive power of the measurements (the accuracy of the signals) and the ability of the decision maker (DM) to use the information (the accuracy of the DM's response to the signals) jointly determine the quality of decisions, it would seem beneficial to use a methodology that examines both factors jointly.' Failure to consider the accuracy of responses could result in judging as relevant information that cannot be utilized effectively by the DM because of his limitations as an information processor. In this paper I report the results of a field study designed to jointly evaluate the predictive power of ratio information and the ability of loan officers to evaluate that information in the business failure prediction context. Previous studies by Beaver (1966, 1968), Altman (1968a, 1968b). Edmister