Predicting Credit Default among Micro Borrowers in Ghana

Abstract

Microfinance institutions play a major role in economic development in many developing countries. However many of these microfinance institutions are faced with the problem of default because of the non-formal nature of the business and individuals they lend money to. This study seeks to find the determinants of credit default in microfinance institutions. With data on 2631 successful loan applicants from a microfinance institution with braches all over the country we proposed a Binary logistic regression model to predict the probability of default. We found the following variables significant in determining default: Age, Gender, Marital Status, Income Level, Residential Status, Number of Dependents, Loan Amount, and Tenure. We also found default to be more among the younger generation and in males. We however found Loan Purpose not to be significant in determining credit default. Microfinance institutions could use this model to screen prospective loan applicants in order to reduce the level of default.

Keywords: Microfinance, Loan Default, Default Prediction, Logistic Regression