A sample of one hundred and seventy‐seven New South Wales’ local governments is used to compare the data envelopment analysis (DEA) and stochastic frontier approaches to cost efficiency measurement. The results suggest that the choice of reference technology implies different levels of efficiency and rankings for identical samples and thereby has a dramatic influence on the policy implications drawn from frontier cost efficiency measures. Notwithstanding these comments, DEA and stochastic frontiers should be thought of as complementary tools in the analysis of local public sector efficiency. In the first instance, the frontier adheres closely to the notion of best‐practice efficiency, whereas in the second it refers to an absolute measure of efficiency. Clear advantages thereby accrue to DEA in identifying benchmark local governments and peer groups for comparison, while the stochastic frontier permits the opportunity for direct comparison with other institutional milieus.
No takes yet. Share an insight, caveat, or question.
Andrew C. Worthington (2000) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: