Lexin Cai (pronounced Lex-in KAI)

I examine the impact of switching from means-tested to universal provision of free school meals. Using linked administrative data from Texas, I exploit the staggered rollout of the Community Eligibility Provision (CEP), which allows high-poverty schools to provide free breakfast and lunch to all students without requiring individual applications or income-based eligibility. CEP increases schoolwide average daily meal participation by 6.2 percentage points (10 percent), equivalent to 20 additional meals per student per year. Participation rises both among previously eligible students, consistent with reduced stigma and application burden, and among newly eligible students, whose increase is about twice as large. The marginal participants induced by CEP are more economically advantaged than students who participate under means testing, suggesting worse targeting on need. I find no statistically significant effects on academic, behavioral, or economic outcomes, either overall or by student income, in the short or long run. Specifically, the 95 percent confidence intervals rule out increases larger than 0.02 standard deviations in grades 3–8 test scores and 5 percent in high school graduation relative to the baseline mean. Applying the marginal value of public funds framework, I find that the costs of CEP exceed its benefits.

We examine the effectiveness of work requirements as a screening device in the Supplemental Nutrition Assistance Program (SNAP). Work requirements for “able-bodied adults without dependents” were suspended after the Great Recession and gradually reinstated across counties and states in the 2010s. Using linked administrative SNAP and employment data from five states and a triple-differences design, we find that work requirements reduce SNAP participation by 7 percent without increasing labor supply and disproportionately screen out low-income individuals. We develop a welfare framework to interpret these results and find that the social costs of work requirements exceed budget savings.

I investigate how universal free school meals affect the measurement of student poverty and state K-12 funding. About half of U.S. public schools participate in the Community Eligibility Provision (CEP), which provides free meals to all students and eliminates household applications. This could distort free or reduced-price lunch (FRPL) data that many states use to allocate additional aid to low-income students. Using administrative data from 31 states, I find that reported FRPL shares diverge from child poverty in opposite directions depending on how states count students in CEP schools. Relative to poverty, FRPL shares fall by 5 log points in states reporting actual counts and rise by 10 log points in states counting all students in CEP schools. Despite these differences in reported FRPL, I find little evidence of funding loss. In the 15 states that allocate poverty funding based on FRPL, I can rule out annual losses in state formula aid exceeding $105 per student.