THE MARGINAL IMPACT OF CASH FLOW INFORMATION ON BANK FAILURE PREDICTION: AN EMPIRICAL INVESTIGATION

This study empirically examines the marginal impact of cash flow-based information on bank failure prediction. Previous empirical studies rely predominantly on accrual-based accounting ratios selected to measure the various categories of the FDIC's bank ranking system popularly known as CAMEL (Capital adequacy, Asset quality, Management, Earnings and Liquidity.) The research asked the question, "Does adding cash flow-based information to an accrual-based failure prediction model improve the model's ability to predict bank failure?".; The research also addresses the controversy surrounding bank cash flow reporting. In particular, FASB Statement No. 95 now requires that all banks include a Statement of Cash Flow as part of a complete set of financial statements. The accounting community generally argue that this statement provides investors with information useful for assessing the future financial performance of the reporting entity. The banking community argue that cash flow analysis is not useful for assessing bank performance and reporting of a bank's cash flows is superfluous.; Statistical testing required developing two models: (1) a CAMEL model employing only accrual-based financial ratios as explanatory variables, and (2) a mixed model (MM) combining these same accrual-based ratios with cash flow-based ratios as explanatory variables. The cash flow-based measures are based on a Cash Flow Identity developed by Lawson (1985). All test results are evaluated in the context of dichotomous choice probability models estimated using logit regression techniques.; The empirical findings show that accrual-based CAMEL variables are robust empirical predictors. The CAMEL and MM models classified failed and nonfailed banks at different rates as evidenced in their respective validative and predictive accuracies. However, the observed differences in these classification accuracies are not strong enough to establish statistical differences between the models. With regard to bank cash flow reporting, the empirical evidence supports the banking community's contention. The information in a bank's Statement of Cash Flow is of questionable value when assessment of future solvency and performance is the objective.