BENCHMARKING ANALYSIS IN ELECTRICITY DISTRIBUTION *

separate working paper with William Greene. As competition is introduced into electricity generation, incentive regulation of distribution utilities has become more common. Most of the regulation schemes used in practice are based on benchmarking that is, measuring a company’s efficiency against a reference performance. In benchmarking applications the regulator is generally interested in obtaining a measure of firms ’ efficiency in order to reward (or punish) companies accordingly. The reliability of efficiency scores is therefore crucial for an effective implementation of the incentive mechanism. There are a wide variety of methods to measure efficiency. These methods can be classified into two main categories: non-parametric or linear programming methods such as data envelopment analysis, and parametric or econometric methods such as stochastic frontier analysis. A main problem faced by regulators is the choice of the benchmarking method (parametric and non-parametric) and within each method the choice among several legitimate models, especially as different models usually produce different results. Part of this discrepancy i