What accounts for the discrepancy between the microfinance impact claims of development practitioners and the far smaller impacts found in experimental studies? We demonstrate in a simple theoretical framework why "before-and-after" observations of practitioners overstate microfinance impacts and why estimations in some recent randomized trials understate the average treatment effect on the treated (ATT). Our empirical study uses a unique data set from eastern Nepal to study the impact of microfinance in villages where microfinance did not previously exist. We find that approximately three-fourths of the apparent impact of microfinance observed by practitioners is an illusion driven by correlated unobservable factors.
Rajbanshi, Ram D.; Huang, Meng; and Wydick, Bruce, "Measuring Microfinance: Analyzing the Conflict between Practitioners and Researchers with Evidence from Nepal" (2014). Economics. Paper 5.