The implementation of farm to school programs and the use of local foods in US school meals have increased dramatically since the 1990s. However, supply constraints are often cited by school districts as an impediment to purchasing local foods. In this paper, we estimate the responsiveness of local food sourcing by schools in response to changes in local agricultural production. We test several hypotheses by merging data from the nationally administered 2015 Farm to School Census, which solicits information about local food sourcing from school districts, with Census of Agriculture data and other socio-economic data. We test whether local agricultural conditions influence the probability that a school district sources local foods, as well as the level of such purchases among the subset of school districts that are buying locally. We examine two types of local food purchases: local fluid milk purchases, which is the predominate food product that is locally sourced by schools, and local non-milk food purchases. We test the extent to which local purchases are influenced by local dairy production in the former case and local direct-to-consumer (DTC) agricultural production in the latter case. We find that the dairy and DTC agricultural production had a positive, although modest, impact on local milk and local non-milk purchases, respectively.
We find that county-level average income and the percentage of residents in poverty, when statistically significant, had positive and negative, respectively, impacts on local sourcing. Interpreting the coefficients on some of our other control variables involves greater nuance. For instance, while the percentage of students eligible for free and reduced-price meals decreases the predicted probability of a school making local non-milk purchases, it has a positive impact on the level of expenditures. We also find that the number of students in a school district has a positive impact on local food expenditures. However, while county-level population has a positive impact on local food expenditures, it has a negative impact on the predicted probability that a school district sources local non-milk products.