Reducing vulnerability through microfinance: Evidence from Indian self help group program

We investigate if participation in Indian Self Help Group microfinance program (SHG) results in reducing vulnerability. Vulnerability estimates are constructed using crosssectional SHG rural household survey data, collected in 2003. The potential selection bias is eliminated by propensity score matching to estimate the average treatment on treated effect using nearest neighbour matching and local linear regression algorithm. We find that despite a disproportionately high percentage of poor in the SHG members, vulnerability is not significantly different between the SHG and non-SHG members. This result is found to be robust using sensitivity analysis and Rosenbaum bounds method.