Size, setting and the cost of schools: Evidence from Ohio on the role of districts, and insights into school finance policy

The goal of this dissertation is to inform policy focused on the cost of education, the potential of district size as a cost-cutting tool for schools and districts, and the appropriate way to incorporate costs into state school aid formulas. Pursuing this goal has involved taking advantage of a rich four-year panel data to estimate traditional cost functions for Ohio districts and schools, assess the impact of size on cost, and calculate cost indices. This dissertation breaks new ground in several ways: Ohio has been the object of little research to date; panel data increases the precision of the results and reduces omitted-variable bias; school-level cost functions are estimated; costs across the whole spectrum of urban settings (rural, suburban and urban) are explored; and size is defined in a variety of ways, including the number of schools in a district.; District-level cost functions results are consistent with research in other states: costs generally increase as performance, input prices and cost factors increase, and the relationship between costs and enrollment indicates that there are economies of scale to be realized. When district are ranked based on cost indices calculated using different methods and on their residual costs, they move around in the rankings. Caution suggests using at least a few of the methods. District size may be manipulated to lower district costs. Options include manipulating enrollment, reorganizing large districts into smaller sub-districts and reducing the number of schools in the smallest districts.; Some school-level cost function results are similar to district-level results, while others differ, and the relationship between costs and size is complex, although there is weak evidence of economies of scale. If district size is chosen as a tool to curb district costs, this will not be without consequences for school costs. Policies tailored to groups of schools, depending on their urban setting, their size and the size of their home district, are warranted. Such tailoring may be unrealistic, potentially highly costly, and quite difficult to implement.