Abstract. This paper attempts to investigate the determinants of productivity in microfinance institutions (MFIs) in India using the empirical Bayesian technique. To do this, we utilize an unbalanced panel dataset covering the period 2005-2011 with 292 observations from 64 institutions. Based on theoretical grounds, three broad factors are specified: institutional characteristics, outreach, and efficiency. We find convincing evidence that institutional characteristics and outreach have both positive and negative effects on the productivity of MFIs, depending on the proxy used in the analysis. However, the efficiency of MFIs affects the productivity negatively. Specifically, we find that the age of the institution positively influences the productivity by 6.1581 points, while number of offices and number of personnel negatively affect it by 26.41% and 8.77%, respectively. Of the outreach variables, numbers of active borrowers positively influence productivity by 0.04%, whereas average loan size appears to have an inverse relationship with productivity. We further find that cost per loan – a proxy for efficiency, has a negative and statistically significant impact of 1.9604 points on the productivity of MFIs. Overall, our investigation suggests that there is a need to build client confidence and pursue innovative credit delivery techniques in reaching out to the poor and achieving high levels of productivity.

Keywords: productivity; microfinance institutions; efficiency; outreach; empirical Bayesian estimation technique.