Default risk modelling for small-to-medium enterprises in the context of stressed conditions in an undeveloped economy

This paper designs stepwise logit models to predict the default probability for small-to-medium enterprises (SMEs) under downturn conditions in an undeveloped economy. The primary focus of this study is to recognise and interpret the determinants of default probability for SMEs. We apply an empirical data set of Zimbabwean defaulted and non-defaulted SMEs for applicability and effectiveness motives. Our experimental results indicate that the ratio of (current assets - current liabilities)/total assets, the earnings before interest and tax/total assets ratio, the book value of total assets, the real GDP growth rate, the inflation rate, the ratio of net sales/net sales last year, the age of the firm and the ratio of bank debt/total assets are all robust determinants of default probability for Zimbabwean SMEs. The implication here is that firm and loan features, accounting ratios and macroeconomic factors should be incorporated when forecasting default probability for SMEs in the context of stressed conditions.