Does Sharing a School Cafeteria Cut Costs? Evidence from New York City Public Schools

• In Event: Examining the equity implications of school fiscal and physical resources and constraints

Abstract

In New York City, more than 60% of public schools share a building with at least one other school (Wolff & Rogers, 2014). While inter-district consolidation has been widely studied for its effects on reducing district-level expenditures (Duncombe & Yinger, 2007; Chin, 2023), little attention has been paid to the cost effectiveness of schools sharing buildings. For example, schools that share non-instructional facilities, such as cafeterias, may expect economies of scale in the provision and delivery of school food services.  

Shared school cafeterias may also moderate the extent to which school food policies, such as Universal Free Meals (UFM), affect spending on food services. A recent study finds that the adoption of Community Eligibility Provision, which makes school lunch universally free among schools in participating districts, increases per pupil food expenditures but decreases per meal costs (Rothbart et al., 2021). Yet, such effects may be smaller (or larger) in schools that share buildings and facilities. We add to the literature by answering the following questions: (1) Do co-located schools have higher or lower per pupil and per meal spending on food services than stand-alone schools, after controlling for observable characteristics? (2) Does co-location status moderate the relationship between UFM and food service expenditures? 

We analyze a balanced panel of 5,220 school-year observations of 1,044 New York City public schools in 2010-2014, drawing on school-level data from the New York State Annual School Report and the School Report Card (SRC) and the School Budget Expenditure Report (SBER). These data include information on school location, enrollment, grades served, student demographics, and line-item expenditures.

Employing both across-school (co-located vs. stand-alone schools) and within-school (before and after co-location) comparisons, we find that co-location is associated with higher per pupil spending on food services but no differences in per meal costs. Given that co-located schools have lower school meal participation, these results suggest that shared facilities may offset inefficiencies from smaller-scale food production and delivery by potentially distributing fixed costs across schools. Instructional spending, including teacher salaries, is not associated with co-location, but with the exception of spending on paraprofessionals; this is possibly due to paraprofessionals, rather than teachers, being more directly involved in coordinating shared non-instructional services such as food distribution and cafeteria supervision. We plan to further explore whether co-location moderates the effects of school food policies on expenditures. Results would offer new insights into whether shared school facilities can generate cost efficiencies in non-instructional services. 

Authors

• Jeehee Han, Texas A&M University
  Presenting Author

• Michah Weitzman Rothbart, Syracuse University
  Non-Presenting Co-Author