Social Safety Net Programs and Household Behavior

This dissertation consists of two essays analyzing the effect of social safety net programs on beneficiary behavior in India. The first paper uses data from the National Sample Survey, a nationally representative survey of Indian households, to study the impact of expanding government food aid on child labor and school attendance. Beginning 2001, the Indian state of Chhattisgarh implemented reforms to the traditional public distribution system (PDS) that purchased foodgrain staples from farmers at a minimum support price and distributed it to poor families at subsidized prices. The first major reform expanded the number of retail stores selling subsidized foodgrains. The second major reform shifted government procurement of rice from out-of-state farmers to local farmers. I find that these reforms significantly increased paid work among older boys aged 12-17 years in Chhattisgarh. These effects are primarily driven by boys belonging to socially disadvantaged groups, living in rural areas and who are legally permitted to work as per the child labor ban law. For young boys aged 5-11 years, I find an increase in schooling and much smaller increases in paid employment. I do not find a significant impact on the schooling and paid employment of young or older girls. These findings are consistent with PDS reforms increasing family resources as well as labor market opportunities, with the latter having a greater impact on older boys such that the net effect is an increase in paid employment for them. The second paper evaluates the effect on savings-consumption behavior of beneficiary households due to the introduction of a nationwide health insurance program named Rashtriya Swasthya Bima Yojana (RSBY) in 2008 that provided free hospital insurance to the poor population. I use two rounds of the India Human Development Survey (IHDS), a nationally representative survey in India. Using a difference-in-differences strategy based on the phased roll-out of the program across different districts and the panel structure of the data, I estimate the intention to treat effects for the eligible population. I find that the program increased RSBY take-up by 8 percentage points among the eligible population. I do not find statistically significant effect of RSBY on consumption, savings, income or asset holdings of the beneficiary households.