Technical efficiency of Microfinance Institutions (MFIs).

Purpose The paper aims to evaluate the technical efficiency of African Microfinance Institutions (MFIs) and examine if there is performance difference by ownership type.Design/methodology/approach The paper applies stochastic frontier analysis (SFA) assuming that the translog production functions to estimate the technical efficiency of 134 Microfinance Institutions operating in 36 African countries. The parametric SFA is preferred over the non-parametric, as it captures the random and inefficiency effects. Though the suitable approach is SFA, for the purpose of consistency and robustness of the results, the alternative data envelopment analysis (DEA) approach is also run and the results are compared with those derived from SFA.Findings In our analysis we have found that African MFIs are technically inefficient. The average technical efficiency for the sample institutions is 0.489, which is quite low and suggests that on average, African MFIs are achieving only 48.9 per cent of the maximum achievable output. Our results also revealed the presence of significant technical inefficiencies with considerable differences in inefficiency among the MFIs. Further, we found statistically significant difference in the efficiency performance among the different ownership types of MFIs. More importantly, the NGO and non-bank financial institutions are relatively more efficient, while the cooperatives/credit unions are the least efficient.Research limitations/implications The study contributes to the continuing debate on the effect of ownership type on performance of institutions. Moreover, it indicates the importance of using certain approaches and complementing them with other alternatives for a better insight.Practical implications The study found that the least efficient type of MFIs are the cooperatives/credit unions. This might be related to the nature of these institutions where the members are owners and borrowers. This might affect efficiency negatively, although it may somehow address the agency problem.Originality/value This paper provides an evidence on efficiency performance of African MFIs, taking a large data set and applying SFA. DEA was also used to complement the SFA results. It provides useful empirical evidence and perspective on this important issue for policy makers and analysts. [ABSTRACT FROM AUTHOR]