Association for Public Policy Analysis & Management

School meal programs have been often promoted as an important policy tool for improving the health and nutrition of impoverished children in both developed and developing countries. However, their value in aiding academic outcomes has often been debated. I exploit a recent change to the U.S. National School Lunch Program (NSLP) and School Breakfast Program (SBP) to estimate the effect of an expansion of the school meals programs on students' academic performance.

The Community Eligibility Provision (CEP), included as part of the Healthy, Hunger-Free Kids Act of 2010, allows participating schools to provide free meals to all of their students, regardless of family income, as long as at least 40 percent of the students are income eligible. Constructing a unique data set by combining eight years of school-level math and reading achievement data from the Department of Education with the school CEP participation data, I estimate intent-to-treat (ITT) and treatment-on-treated (TOT) effects for schools who are just above the 40 percent income eligibility cutoff compared to schools who are just below the cutoff and therefore ineligible.

I find the expanded school meals program leads to a significant increase of several percentage points in the average math and reading proficiency levels. These findings thus make an important contribution to the literature, which in the past has primarily focused on policy changes at the local or regional level with ambiguous results.