Measuring effects of SNAP on obesity at the intensive margin.

The effects of the Supplemental Nutrition Assistance Program (SNAP) on obesity have been the focus of much debate. However, causal interpretation of estimates from previous studies, comparing participants to non-participants, is complicated by endogeneity and possible misreporting of participation in SNAP. In this paper, we take a novel approach to examine quasi-experimental variation in SNAP benefit amount on adult obesity. Children of SNAP households qualify for free in-school meals, thus freeing some additional benefits for the household. A greater proportion of school-age children eligible for free in-school meals proxies for an exogenous increase in the amount of SNAP benefits available per adult. Using data from the National Longitudinal Survey of Youth-1979 we show that school meals represent a non-trivial part of the food budget for SNAP households. We find that increases in SNAP benefits have no effect on obesity levels for the full sample of those who report SNAP participation. To better isolate the effects of additional benefits from other potential changes we restrict our analysis to adults living in households with at least one child under 5 years of age. In this setting, we find that additional SNAP benefits reduce BMI and the probability of being obese for SNAP adults. Specifically, when one child in a household of four becomes school-aged, adult BMI is expected to decrease by 0.23 units and the probability of being obese decreases by 2.58 percentage points or by about 10%.