This article examines whether the means-tested component of the National School Lunch Program changes beneficiaries' dietary patterns by taking advantage of variation across public school districts in the financing of and demand for lunch and nutrition programs. Using data on fifth grade public elementary school children in the Early Childhood Longitudinal Study-Kindergarten (2003-2004), we find significant increases in weekly rates of consumption amongst fully and partially subsidized children. Our estimates also suggest that the increase was for items known to be a rich source of vitamins and minerals that are essential for children's health and development. The effects are larger for fully subsidized children relative to partially subsidized children, which suggests the nominal price of school lunch is a binding constraint for certain children on the margin of eligibility for the subsidies. To the extent that children from low-income households experience undernourishment with greater frequency, policy discussion focusing exclusively on the link between obesity and program participation is overlooking positive effects on those who are directly subsidized.