A cash flow and macroeconomic model of financial distress

The primary objective of this study was to determine whether a failure model derived from a working capital basis Statement of Changes in Financial Position converted to a cash basis, can classify and predict with reasonable accuracy firms that will fail. A secondary objective of this study was to evaluate the contribution that changes in specific economy-wide variables have in determining which firms will experience financial distress by including macroeconomic variables in the prediction model.; To accomplish the objectives, five different models were developed. The cash flow variables consisted of six cash inflow variables and seven cash outflow variables. Model 1 consisted of the cash flow variables. Model 2 consisted of changes in the cash flow variables. Model 3 consisted of the cash flow variables together with two macroeconomic variables. Model 4 consisted of changes in the cash flow variables together with the macroeconomic variables. Model 5 consisted of the cash flow variables and changes in the cash flow variables together with the macroeconomic variables.; Stepwise procedures were used with both discriminant analysis and logistic regression analysis to reduce the number of variables for each model. The resulting models were then compared for both their classification accuracy (using data from 1974-1980) and prediction accuracy (using data from 1981-1988).; The best model, as determined by its prediction accuracy, was Model 3. This model was compared to the models of Altman (1968), Gentry, Newbold, and Whitford (1984) and Bahnson and Bartley (1989) in terms of both classification and prediction accuracies. The best model outperformed the other models in classification accuracy using both discriminant and logistic regression analysis, but the prediction accuracies of all of the alternative models were only slightly better than chance.; In this study, the amount of dividends, investments and changes in long-term debt were found to discriminate between financially healthy and unhealthy firms. The inclusion of the macroeconomic variables added only slightly to the ability of the models to discriminate between the financially distressed and nondistressed firms.