The Original Study
Santos et.al. (2014) assess the impact on food security of a program that provides microplots to landless families in West Bengal, India with the purpose of improving their incomes and food security status. The authors use data from a baseline survey applied to a sample of 1373 households from Coochbehar, Bankura and Jalpaiguri districts in West Bengal. For the analysis, the authors relied on a propensity score-weighted regression model.
They find statistically significant program effects on the four intermediate outcomes. They report that the average treated household was 12% more likely to get a loan from a formal bank. Also, they invested more in agriculture, for example they were 11% more likely to have used fertilizer or pesticides. Regarding women participation, a woman in a treated household was more likely to be involve in decisions regarding family land and 9% more likely to be involved in food purchase and consumption decisions. They did not find any statistically significant effect on food security measures. The authors argue that the null results may be due to the fact that only 25% of beneficiaries had moved; that some of the plots had poor land quality; or because more time is needed to see results on the food security outcomes.