Date of Award:


Document Type:


Degree Name:

Doctor of Philosophy (PhD)


Applied Economics

Committee Chair(s)

W. Cris Lewis


W. Cris Lewis


Donald L. Snyder


Basudeb Biswas


John E. Keith


Drew Dahl


Efficiency in public education is a significant issue in the United States. Nationwide, real expenditure per student increased 8% per year between 1960 and 1993, but output as measured by standardized test scores has not increased and in some cases (i.e., the verbal SAT [Scholastic Achievement Test] score) has declined. One explanation is that resources are not being utilized efficiently either in the technical or allocative sense. Also, the issue is important because substantial savings are possible by consolidation of schools and/or districts.

This dissertation explores efficiency by measuring technical efficiency at the school district level from four perspectives. The first essay (Chapter 2) explores whether the cost efficient production units are the bigger schools or the bigger districts, using the concept of a standard education cost function (the dual of neoclassical production function). The empirical analysis uses panel data from Utah school districts and estimates the cost and expenditure functions using the covariance and error component models after making corrections for heteroscedasticity and autocorrelation. The evidence indicates scale economies associated with school size but not district size.

In the second essay (chapter 3), technical efficiency of individual school districts is measured using an educational production function and stochastic frontier methodology. The empirical analysis shows substantial variation in efficiency among school districts.

An extension of the second essay (Appendix B) estimates technical efficiency and total factor productivity using school districts as multi-output producing units. A deterministic nonparametric approach, known as data envelopment analysis (DEA), is applied to a panel data. The empirical results provide provide pure technical efficiency disaggregating the components of scale, congestion, and technical innovation.

Most studies of technical efficiency using a stochastic production function are estimated using cross-section data. Technical inefficiency effects are assumed (1) to be a function of the district-specific variables (i.e., dropout rate, socioeconomic status of the students, etc.) and time, and (2) to be independently distributed as truncated normal with constant variance and with means dependent on the firm-specific variables and time. The empirical results suggest that technical inefficiency increased over time for Utah school districts, and that inefficiency effects are stochastic.



Included in

Economics Commons