Gender and other repayment determinants of microfinancing in Indonesia and Sri Lanka

Abstract

Purpose – The repayment performance of microfinancing loans funded by donors amounting to hundreds of millions of dollars is an important issue, because it indicates the effectiveness of utilising these funds to alleviate poverty. The purpose of this paper is to develop models to predict the repayment success of microfinancing loans.

Design/methodology/approach – Analysing data relating to 1,109 random loan records from Indonesia and Sri Lanka, the study develops models to predict the repayment probability of microfinancing loans using logistic regression.

Findings – There are significant differences between the two countries. In Sri Lanka, the time to approve and disburse the loan, loan cycle, gender and age of the borrower, whether a group or individual borrower, the purpose for which the loan is used and visiting frequency by the loan officers were found to be significant when predicting the repayment. Only three factors were significant in Indonesia: time to approve and disburse the loan, interest repayment frequency and gender. Both models have over 70 per cent prediction accuracy.

Originality/value – The models developed can be used in the loan appraisal stage to improve the repayment performance of microfinancing institutions saving hundreds of millions of dollars in bad debt write offs.