This dissertation contains two research papers. In Chapter 2, I estimate the effect of program interactions between Food Stamps and Head Start during childhood on labor market and health outcomes in adulthood. Using a nationally representative panel data set and exploiting variation in the exact timing of program introduction for Food Stamps and Head Start across counties during the 1960s and 1970s, I compare outcomes for children who were differentially exposed to each program during childhood by virtue of the county and year in which they were born. Estimates from event study and difference-in-differences models indicate that accessing Food Stamps during childhood improves labor market and health outcomes in adulthood, and Head Start availability at preschool age improves health outcomes in adulthood. However, I find no evidence that the long-run effect of accessing Food Stamps during childhood depends on the availability of Head Start at preschool age. In Chapter 3, I study the effect of Head Start on children's long-run outcomes. Jackson and Johnson (2019) exploit quasi-random variation in the timing of Head Start introduction across counties to estimate the effect of preschool-age Head Start spending on outcomes in adulthood. Their event study estimates imply that, for poor children who reside in high Head Start spending counties, preschool-age exposure to Head Start increases educational attainment. I perform a replication exercise using Jackson and Johnson's analytical sample, and I am unable to reproduce these event study estimates. Instead, I find no statistically significant effect of preschool-age exposure to Head Start on poor children's educational attainment, in either high or low Head Start spending counties.