Do Universal Free Lunch Programs Boost Academic Performance?

Across the United States, more and more districts are tossing the old eligibility checklist aside and serving a free lunch to every child, regardless of household income. Seattle’s board made the move this summer, joining a wave that has more than tripled since the 2016‑2017 school year, according to federal data.

Funding comes from a patchwork of sources. Some places—Seattle, New York State, for example—pay out of local coffers, while the bulk of districts lean on the National School Lunch Program, a $17‑plus‑billion federal effort that fed roughly 4.9 billion meals last fiscal year.

Proponents cheer the policy as a low‑cost, high‑impact intervention. “It’s the best thing since free sliced bread,” says Amy Ellen Schwartz, a public‑policy professor at the University of Delaware. In a working paper on New York City middle schools, Schwartz and a colleague reported that universal meals nudged math scores up by about 0.08 standard deviations for non‑poor students—roughly three percentile points—and English scores by 0.06 standard deviations. Poorer students also benefited, though the gains were a touch smaller.

Other researchers echo the optimism but warn against sweeping generalizations. Orgul Ozturk, an economics professor in South Carolina, found that elementary schools adopting free meals saw math test scores climb about 0.06 standard deviations. “The effects are heterogeneous,” she notes, “but they’re positive even when modest, especially for a program that isn’t terribly expensive.”

Not everyone is convinced. A recent synthesis of 42 studies concluded that, in high‑income economies, school‑meal programs have only minimal impact on behavior, health, or educational outcomes. Critics like Chris Edwards of the Cato Institute argue that federal funding for school food is an outdated entitlement and should be handed back to the states.

To cut through the noise, Scientific American dug into two public datasets: the Stanford Education Data Archive’s standardized‑test estimates and the USDA’s records on which districts qualify for free meals under the Community Eligibility Provision. Looking at three pre‑COVID years (2016‑2019) and two post‑COVID years (2022‑2024), the analysis found that districts with universal meals tended to score about 1.6 grade levels lower in reading and 1.66 lower in math. When demographic factors—poverty rates, racial composition, urban versus rural setting—were accounted for, the gap narrowed to roughly two‑thirds of a grade level.

Even more striking, a matched‑pair comparison of 1,452 districts—one with universal meals, one without, but otherwise similar—showed the free‑meal districts lagging by about 1.1 grade levels in both math and reading during the 2023‑2024 year. Yet when the researchers examined year‑to‑year changes within the same districts, they found no clear link between expanding free‑meal coverage and shifts in test scores.

The bottom line? The data reveal an association, not a cause‑and‑effect story. Experts caution that gaps in the Stanford data, the lack of student‑level detail, and variations in how districts roll out meals could be clouding the picture. What’s clear, though, is that universal free lunch delivers undeniable social and nutritional benefits—even if the academic payoff remains modest and uneven.