A Cointegrated Multi-Factor Semi-Log Regression Approach to Quarantine the Impact of Microfinance on Poverty Regionally

Abstract

This study aims to examine and segregate the impact of COVID-19, microfinance, and multiple macroeconomic variables on various poverty indicators in a single country at the macro level. Using a 35-year time series (1988-2022), the study applies unit root cointegration tests to address non-stationarity in the data. The semi-log regression method is employed to estimate poverty measures and disentangle the relative importance of different factors. The main research questions are: (1) What is the impact of COVID-19 on poverty indicators? (2) How does growth in microfinance borrowers, service availability and gross loan portfolio affect poverty levels? (3) What is the role of macroeconomic factors in poverty reduction? The findings reveal that the COVID-19 fixed effect is statistically significant across various poverty measures, while an increase in microfinance borrowers and service availability is associated with a reduction in poverty. The total loan portfolio has a significant effect on poverty levels despite its small size. Other macroeconomic variables have mixed effects on poverty indicators. The study concludes that governments should invest in expanding social policies such as education and training, support for entrepreneurs, and universal healthcare in addition to expanding microfinance services to reduce poverty effectively.