An Evaluation of the Altman Z-Score Model in Predicting Corporate Bankruptcy for Canadian Publicly Listed Firms

The purpose of this study is to assess the effectiveness of Altman’s Z score in predicting corporate bankruptcy for Canadian listed companies. First, we estimate Altman’s original model and test the efficiency of the cut-off region, coefficients and the variables used. Secondly, we add Cash Flow from Operation (CFO) to Total Liabilities (TL) ratio as a sixth variable to improve Altman’s (1968) standard Z score model. By testing and comparing these models using a sample of 70 bankrupt firms against a population sample of 1,047 non-distressed firms, this study determines which models has a higher discriminating ability. While Altman’s research focuses on manufacturing companies, for the purpose of this study, we selected firms from five sectors; energy, consumer discretionary, consumer staples, industrials and materials. We use Multiple Discriminant Analysis (MDA) to compare the predictive abilities of these models. Our results show that the cut-off regions and the coefficients used by Altman should be time-varying and that our augmented model produces better results, when removing the effects of outliers.