Business Failure Prediction Modeling and Firm-Territory Relationships : Preliminary Findings of a Statistical Study of Small Italian Firms

Purpose of the paper: To test whether the qualitative variables regarding the territory and/or the firm-territory relationship can improve the accuracy rates of small enterprise business failure prediction models.Methodology: We applied logistic regression to a sample of 736 Italian small enterprise and built 4 different business failure prediction models. We then tested the accuracy rates obtained with these models against a holdout sample.Findings: Including territory qualitative variables and, most importantly, those regarding the relationships between firms and their territories, significantly improves the accuracy rates of default prediction models. This improvement is most marked in the case of smaller firms.Research limits: The qualitative variables data collected are affected by subjective judgments made by managers in the firms studied. In addition, no other qualitative variables (such those regarding competitive strategies, or managerial skills, or inter-organizational characteristics) were included, nor were quantitative variables regarding the relationships between firms and financial institutions.Practical implications: The study suggests that financial institutions should include territory qualitative variables, and, above all, qualitative variables regarding the firm territory relationship, when constructing business default prediction models. Including this type of variable could reduce the tendency to place unnecessary restrictions on credit.Originality of the paper: In the field of business failure prediction modeling no research has been done into whether it would be useful to adopt qualitative variables regarding the firm-territory relationship. It is a unexplored area