IL RAFFORZAMENTO DELLA CAPACITÀ PREVISIVA DEI MODELLI DI ANTICIPAZIONE DELLE CRISI DI IMPRESA: IL CASO DEL SISTEMA DI ALLERTA PROPOSTO DAL CNDCEC

As the pandemic urged further investigations on the prediction of firms’ financial distress, this study develops and tests an alternative alert system which combines the benefits of the Z-score’s multivariate discriminant model and the National Council of Chartered Accountants and Accounting Experts’ predictors. Using a sample of 43 viable and 43 non-viable Italian SMEs, we first compare the financial distress predictive accuracy of the models mentioned over the period 2015-2019. On the basis of the results, we elaborate and compare the revised versions of both approaches which align them to the current socio-economic conditions. Also, we provide an alternative measure which embeds a Z-score calculated using the ratios elaborated by the National Council of Chartered Accountants and Accounting Experts for the alert system. The analysis of the two baseline approaches showed complementary results as the Z-score overperformed the alert system when predicting the status of non-viable firms whereas the opposite emerged as regards viable firms. The revised version of both approaches pointed out an enhanced predictive accuracy with respect to baseline models. In particu- lar, the complementary role of the Z-score has been integrated into the new alert system as major contribute to its enhancement which pointed it out as the best measure employed. Our analysis enriches the post-pandemic debate on refined financial distressed prediction methods by pointing out the limits of the alert system as designed by the National Council of Chartered Accountants and Accounting Experts and suggests an alternative and better performing measure that may be used by third-party bodies to predict financial distress.