ON THE MEASUREMENT OF TECHNICAL EFFICIENCY IN THE PUBLIC-SECTOR

Existing measures of technical inefficiency obtained through linear pr ogramming models in the public sector do not properly control for envi ronmental variables that affect production. It will be shown that the consequences of not controlling for these fixed factors are biased est imates of technical efficiency. This paper extends the mathematical pr ogramming approach to frontier estimation known as Data Envelopment An alysis to allow for environmental variables. This modified model will be then contrasted with the existing model that purportedly controls f or exogenous factors to measure public sector efficiency with simulate d data. The results provide evidence that the existing Data Envelopmen t Analysis model will overestimate the level of technical inefficiency and that the modified model developed in this paper does a better job controlling for exogenous factors. The modified model is also applied to analyze the technical efficiency of school districts.