Efficiency of microfinance institutions in Sri Lanka: a two-stage double bootstrap DEA approach

- We employ a bootstrapped DEA method to measure the efficiency in microfinance institutions (MFIs).
- We analyze financial and social efficiency to capture outreach and sustainability.
- Most Srilankan MFIs are financially and socially inefficient.
- Bootstrapped based DEA is more informative than a deterministic one.

This study examines technical efficiency and its determinants of 36 microfinance institutions (MFIs) in Sri Lanka using a two-stage double bootstrap approach. In the first-stage, bias-corrected Data Envelopment Analysis (DEA) efficiency estimates for the individual MFI are obtained by means of the smoothed homogeneous bootstrapped procedure (Simar and Wilson, 2000) and then they are regressed on a set of explanatory variables employing the double bootstrap truncated regression approach (Simar and Wilson, 2007). Two different DEA models are designed to obtain DEA scores along financial and social perspectives. According to the results from the first stage, many MFIs in Sri Lanka do not escape criticism of financial and social inefficiency. Second stage regression reveals that age and capital-to-assets are significant determinants on financial efficiency whereas age, type of the institution and return-on-assets are the crucial determinants of social efficiency.