Improving nutritional intake can improve children's academic achievement and longterm health outcomes. This paper evaluates the extent to which, and for which groups of students, schoolwide meals affect reading and math performance. Leveraging crossstate variation in the timing of eligibility for schoolwide free meals under the Community Eligibility Provision, I find universal access to free meals substantially increases breakfast and lunch participation. Math performance increases approximately 0.02 standard deviations in districts with the largest shares of students becoming eligible for free meals. Scaling by the share of newly-eligible students, this effect implies gaining access to free meals increases math scores by 0.07 standard deviations. Test score improvements are concentrated among Hispanic and white students -groups with relatively low eligibility rates under the traditional school meals program.

No abstract

and the Association for Education Finance and Policy for helpful comments. Any errors are our own. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research. NBER working papers are circulated for discussion and comment purposes. They have not been peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.

Government agencies around the world struggle to retain frontline workers, as high job demands and low job resources contribute to persistently high rates of employee burnout. Although four decades of research have documented the predictors and potential costs of frontline worker burnout, we have limited causal evidence on strategies that reduce it. In this article, we report on a multi-city field experiment (n=536) aimed at increasing perceived social support and affirming belonging among 911 dispatchers. We find that a six-week intervention that prompts dispatchers to share advice anonymously and asynchronously with their peers in other cities reduces burnout by 8 points (0.4 SD) and cuts resignations by more than half (3.4 percentage points) four months after the intervention ended. We provide supporting evidence that the intervention operates by increasing perceived social support and belonging in an online laboratory experiment (n=497). These findings suggest that low-cost belonging affirmation techniques can reduce frontline worker burnout and help agencies retain workers, saving a mid-sized city at least $400,000 in personnel costs.