Microfinance industry: understanding the positive link between social andfinancial performances. Empirical evidence from the application of DEAand a mutivariate analysis to 28 Peruvian MFIs

The purpose of this paper is to extend earlier research on the assessment and the understanding ofmicrofinance institution (MFI) efficiency. It also aims at demonstrating that there can be a positivelink between social performance (SP) and financial performance (FP) in microfinance industry.MFIs are organisations that respond to a double bottom-line – i.e. complying with both SP and FPgoals. To date, their efficiency has been solely assessed by means of financial ratios. It is suggestedthat efficiency measurement of MFIs should include other indicators to better appraise the entirety oftheir activiy.The two-step methodological approach combines the Data Envelopment Analysis (DEA) tool and amulivariate analysis. These methods are applied to MIXMARKET data that correspond to a sample of28 Peruvian MFIs.In the first instance, DEA efficiency scores are obtained and ranked for every specification –particuliar set of inputs/outputs. A relevant remark must be formulated: the complete model (includingall inputs and outputs) incorporates both financial and social variables.In the second instance, the multivariate analysis carried out enables to go further into the interpretationof the DEA scores, that is to say beyond the sole comparison of simple efficiencies. The two principalcomponents extracted are significant vectors of information in so far they help understand better why aDMU reaches a certain level of efficiency under a given specification. In other words, the choice of apeculiar specification impacts the efficiency assessment.These results confirm preceding findings revealed with the same methodological design. Moreover,this investigation is believed to make a contribution to the question posed by the contradictory studiesupon the relationships between SP and FP. It provides empirical evidence that there is no trade-offbetween SP and FP and even it supports the hypothesis that the link between SP and FP is positive.Finally, it offers additional insights about this relationship by opening the "black box" of an MFI'ssources of efficiency: best practices are idendified, route to performance traced and status effecthighlighted (Regrouping of DMUs on the graph according to their status).To conclude, this work is part of an attempt to show that DEA approach is relevant to assess efficiencyand productivity beyond ratio analysis by incorporating other types of variables. The research can bereplicated to other sectors whose activies entail the necessity of measuring double or triple bottom-lineefficiency.