Association for Public Policy Analysis & Management

Students stand to gain a lot from eating school lunch, which has been linked to improved nutritional intake, health outcomes, and academic performance. Schools also stand to benefit as higher lunch participation can boost test scores and help schools avoid potential sanctions under accountability systems. Realizing these benefits, however, requires that students participate, yet nearly half did not in 2014. Popular anecdotes and accounts in the media often attribute low participation to the limited appeal of school meals. Reformulating school menus may therefore constitute an important policy lever available to districts for increasing lunch participation but whether and to what extent students are responsive to new menus is unclear. This study constitutes the first large-scale investigation of this relationship. Using detailed longitudinal student-level data and exploiting variation in the types of menus (e.g., in their variety, preparation, convenience, or sensory appeal) offered across nearly 900 New York City (NYC) schools, we examine the association between school menus and lunch participation among middle and high school students.

Specifically, we use student fixed-effects models leveraging variation in the menus students are exposed to over time while attending the same school together with a range of control variables to estimate the impact of a new menu on lunch participation in schools with Point of Service (POS) systems tracking lunch transactions. Our results show that the introduction of a new menu increases the share of students participating and how often they participate. Examining differential responsiveness by student characteristics reveals that new menus can help to close racial, gender, and socioeconomic gaps in the utilization of school lunch. In a series of extensions, we find suggestive evidence that menus may have a greater impact on participation in schools offering all students free meals and that the type of menu (e.g., “grab and go” or “food court”) served also matters. We find no evidence that menus relate to attendance or adverse weight outcomes.

This study presents the first rigorous evidence on the effect of menus on participation in school meal programs. Additionally, by examining an important component of the implementation of school lunch, it informs the efforts that many school districts are taking to reformulate their menus.