Are Frontier Efficiency Methods Adequate to Compare the Efficiency of Water Utilities for Regulatory Purposes?

Abstract

Frontier efficiency methods have been recurrently used in the water sector to assess the performance of water utilities. These methods are also used for yardstick regulation, with greater efficiency being sought by creating competition between the utilities, which can have an impact on decision-making processes, such as tariff setting. This study analyzes the adequacy and limitations of these methods for regulatory purposes, particularly how they deal with data uncertainty and their capacity to manage large number of variables. In order to achieve this, two representative methods—a nonparametric technique (data envelopment analysis) and an econometric one (stochastic frontier analysis)—are applied to an audited sample of 194 water utilities. Results will show that the results from the methods may not be considered conclusive in the water sector and their application should be carried out with considerable reservations.

Benchmarking the economic and environmental performance of water utilities: a comparison of frontier techniques

“…The authors found higher efficiency scores under the StoNED method than under SFA and DEA suggesting that the choice of estimation affected companies' reported efficiency. The choice of benchmarking technique, SFA versus DEA, was also researched by Estruch-Juan et al. (2020) for several water utilities in Spain.…”

“…The authors found higher efficiency scores under the StoNED method than under SFA and DEA suggesting that the choice of estimation affected companies' reported efficiency. The choice of benchmarking technique, SFA versus DEA, was also researched by Estruch-Juan et al (2020) for several water utilities in Spain. The authors concluded that DEA scores were sensitive to data uncertainty whereas the variance in SFA efficiency scores although it was less than DEA it was still significant.…”

“…The empirical application conducted in this study evidenced that the choice of the efficiency assessment method matters. DEA is a deterministic approach and is more sensitive to data variation than SFA (Estruch-Juan et al, 2020). By contrast, SFA splits inefficiency into real inefficiency and random effects.…”

“…However, there is no clear conclusion on which method is superior to evaluate the efficiency of water companies. The inconclusive results have raised concerns about the suitability of both methods for benchmarking regulation in the water industry (Estruch-Juan et al. , 2020).…”

“…Two are the main methodological approaches employed for the purposes of efficiency analysis such as data envelopment analysis (DEA) and stochastic frontier analysis (SFA) (e.g. Dassler et al, 2006;Bogetoft and Otto, 2011;Kuosmanen et al, 2013;Estruch-Juan et al, 2020). DEA is a non-parametric approach which does not assume a priori a functional form for the underlying technology (Charnes et al, 1978).…”