Bankruptcy is a costly event. Holders of publicly traded securities can rely on security prices to reflect their risk. Other stakeholders have no such mechanism. Hence, methods for accurately forecasting bankruptcy would be valuable to them. A large body of literature has arisen on bankruptcy forecasting with statistical classification since Beaver (1967) and Altman (1968). Reported total error rates typically are 10%-20%, suggesting that these models reveal information which otherwise is unavailable and has value after financial data is released. This conflicts with evidence on market efficiency which indicates that securities markets adjust rapidly and actually anticipate announcements of financial data. Efforts to resolve this conflict with event study methodology have run afoul of market model specification difficulties. A different approach is taken here. Most extant criticism of research design in this literature concerns inferential techniques but not sampling design. This paper attempts to resolve major sampling design issues. The most important conclusion concerns the usual choice of the individual firm as the sampling unit. While this choice is logically inconsistent with how a forecaster observes financial data over time, no evidence of bias could be found. In this paper, prediction performance is evaluated in terms of expected loss. Most authors calculate total error rates, which fail to reflect documented asymmetries in misclassification costs and prior probabilities. Expected loss overcomes this weakness and also offers a formal means to evaluate forecasts from the perspective of stakeholders other than investors. This study shows that cost of misclassifying bankruptcy must be at least an order of magnitude greater than cost of misclassifying nonbankruptcy before discriminant analysis methods have value. This conclusion follows from both sampling experiments on historical financial data and Monte Carlo experiments on simulated data. However, the Monte Carlo experiments reveal that as the cost ratio increases, robustness of linear discriminant rules improves; performance appears to depend more on the cost ratio than form of the distributions.

In 1982, the Financial Accounting Board (FASB) issued Statment of Financial Accounting Standards No. 69 (SFAS No. 69) which required oil and gas producing companies to disclose supplementary information to the basic financial statements. These disclosures include, costs incurred, capitalized costs, reserve quantities, and a standardized measure of discounted cash flows. The FASB considered these disclosures to be necessary to compensate for the deficiencies in historical cost financial statements. The usefulness of the new signals created by SFAS No. 69, however, is an empirical question and research regarding that objective is lacking. The objective of the study is to test the usefulness of SFAS No. 69. The research strategy used to achieve that objective is to compare the discriminant and predictive power of SFAS No. 69 signals or SFAS No. 69 signals combined with financial signals to that of financial signals alone. The research hypothesized that SFAS No. 69 signals by themselves or as supplmentary to financial signals have more discriminant and predictive ability for business failure in oil and gas industry than do financial signals alone. In order to test that hypothesis, the study used the multiple discriminant analysis technique (MDA) to develop three equations. The first is based on SFAS NO. 69 signals, the second on financial statement signals, and the third on joint financial and SFAS No. 69 signals. Data were collected from the 10-K's arid the annual reports of 28 oil and gas companies (14 failed and 14 nonfailed). The analysis was repeated for four time bases, one year before failure, two years before failure, three years before failure, and the average of the three years immediately before failure. After assessing the discriminant and predictive ability of each equation in the four time bases, a t-test was used to determine a significant difference in the discriminant and predictive power existed between SFAS No. 69 signals or SFAS No. 69 signals combined with financial signals and financial signals alone. The study concluded that SFAS No. 69 signals by themselves or as supplementary to financial statements have more discriminant and predictive power for business failure than financial signals alone in the analyses of the third year before failure and of the average of three years before failure. The study, however, found no significant difference in the discriminant and predictive ability in the analyses of one year and two years before failure. The results indicated that SFAS No. 69 signals are useful for financial report users in detecting the deterioration of the financial position of an oil and gas company before failure.