Association for Public Policy Analysis & Management

Initiated as part of the Healthy, Hunger-Free Kids Act of 2010, the Community Eligibility Provision (CEP) allows schools and school districts with high rates of poverty to make breakfasts and lunches available at no cost to all students. Under the CEP, districts can use readily-available administrative data to discern eligibility at the school-level, eschewing burdensome and potentially stigmatizing household-level eligibility checks that were the hallmark of earlier programs (USDA FNS, 2016). Under the CEP, all students can get free meals in schools where 40% or more of the student body is eligible. Initiated in a handful of pilot states beginning in 2011, with additional states added each year and nationwide implementation during the 2014-2015 school year, the CEP has significantly increased participation in the school meals programs in participating schools.

Using state administrative data on participation in the CEP matched with child-level data from the Early Childhood Longitudinal Study – Kindergarten Cohort, class of 2010-2011 (ECLS), this study examines whether school-level participation in the CEP is associated with food insecurity, child academic performance, and child behavioral outcomes. The roll-out and nature of the CEP allow for the use of two quasi-experimental methods to assess program impacts. For one, the 40% eligibility cutoff creates a natural discontinuity, enabling comparisons of outcomes for children in schools on either side of the cutoff. Second, the staged roll-out of the CEP and selective participation among schools in pilot states makes it possible to compare year-to-year changes in child outcomes in schools that enrolled in the CEP to similar changes in schools that did not

Accordingly, this paper uses fuzzy regression discontinuity and difference-in-difference analyses to estimate the impacts of the CEP on a range of child outcomes. We focus particular attention on subgroups of free and reduced lunch-eligible children (for whom the CEP is likely to be particularly meaningful) and children living in rural areas, who tend to have significantly lower rates of participation in food and nutrition programs.