Delinquency and Default Risk Modeling of Microfinance in Ghana

Abstract

The objective of this research was to identify the risk factors that influence loan default of customers in the microfinance sector and to develop model that links these factors to credit default for any customer in the sector. Data from a microfinance institution based in Accra was used. A binomial logistic regression analysis was fitted to a data of 472 customers who were granted credit from January 2011 to December 2012. Based on the Wald criterion, it was realized that among the variables that were considered only six out of the 16 predictor variables significantly influence the probability of loan default. One of the key findings of the study was the fact that the loan officer has a significant effect on loan default risk. Other factors include; Client‟s age, Assessment, Type of collateral, guarantor, and residential status. The findings suggest that, default rate is higher for trading and manufacturing sectors than for food vendors as well as those in the service sector. Clients in the service industry are relatively less risky. Clients with guarantors or security other than household items perform well in their obligations. The findings indicated that the responsibility of the loan officer have a tremendous impact on loan default. A test of the full model against a baseline model was statistically significant, indicating that the predictors, as a set, reliably distinguished defaulters and non- defaulters. The Receiver Operating Characteristics (ROC) that measures the sensitivity and specificity of the model was significant at 0.05 level. Using the hold out sample, the model is able to classify defaulters and non-defaulters with at least 80% accuracy. This means that for every four out of five clients, the model is able to predict correctly: default or otherwise. The model could serve as tool to manage and improve loan decision and ultimately enterprise profitability.