The Effect of School and District Size on Education Costs in British Columbia

Abstract

IN T1IIS PAPER we attem ipt to determ-iine the extent to whiclh the cost per studenit of elemiietntary and secondary educationi in British Colutmbia depends on school and district size. We avoid two serious drawbacks of previous school cost curve studies-the aggreg.ation of various cotmiponents of operating costs, and use of district rather than individual school data.2 We conclude that average cost curves are not U-shaped, and hence there is no optimual size school, at least within the range of our sample. We measure the output of the educational system at the elementary and secondary levels by the number of students (N) enrolled at these levels.3 We study the behavior of operating costs within a school as a function of school size, anid choose a form that allows for the possibility of a U-shaped average cost curve as a function of school enrolment. A problem with this procedure is that it requires some sort of standardization of costs for the differing quality of educational services provided. That is, what we are really interested in estimating is the extent which school size affects the cost of proving some standard program at the elemetntary and highl school levels. Consequently we attempt to abstract from cost differences that arise for reasons other than scale, such as those due to a different quLality teaching staff, and the wide variety of special courses offered at somiie higlh schools.4