Abstract
The aim of this paper is to identify factors that affect credit risk of bank customers (based on time to default approach) with emphasis on macroeconomic factors in Iran. This paper tries to find answer for these two questions. 1) What is the effect of individual variables and variables related to loan on credit risk? And 2) what is the effect of macroeconomic variables on credit risk? In this study, a random sample of 5316 customers who borrowed from Maskan Bank of Iran during 1388-1393 is used. The result of the model without macroeconomic variables indicates that loan amount, number of installments, marital status, education, age, job type and number of children have a significant effect on default risk of customers. The result of the model with macroeconomic variables suggests that job type and number of children don’t have significant effect on default risk. But interest rate variable has a significant effect on default risk of customers. The effectiveness of other characteristic variables is like model without macroeconomic variables. In this model macroeconomic variables such as inflation, unemployment rate, house price index growth and economic growth have a significant effect on default risk. Comparisons of Cox-Snell residuals for both models show that the model with macroeconomic factors has better performance than the model without them. Also measuring Harell’s C statistics for test data set indicates that the model with macroeconomic variables has more predictive power than the other model. So when banks want to predict the status of costumers accurately, they should pay attention to macroeconomic conditions.
)