This paper exploits a source of variation in the eligibility for federal nutrition programs to identify the program effects on food insecurity. Children are eligible for the WIC [Women, Infants and Children] program until the day before they turn 61 months old. The result is an age discontinuity in program participation at the 61-month cutoff. Using the Early Childhood Longitudinal Study Birth-cohort dataset, we find strong evidence of a sizeable increase in household food insecurity at the 61-month cutoff. Our findings are robust to different model specifications, datasets, and various bandwidth choices using various non-parametric estimations. A section on data sources is appended.