Adequacy of DEA as a regulatory tool in the water sector. The impact of data uncertainty.

Abstract

The regulation of water services shares many similarities with that of other utilities such as electricity or telecommunications. As a result, similar methods are often used by regulators to assess the efficiency of companies in those sectors. Data Envelopment Analysis (DEA) is one of those widely applied methods. This paper aims to determine the adequacy of DEA as a regulatory tool for urban water services, with a special focus on the quality of the available data. In order to obtain useful conclusions, two DEA simulations were performed with audited data from 194 water utilities, officially made available by the Portuguese water regulatory authority (ERSAR). Both simulations will demonstrate that the inherent inaccuracies found in some of the key data provided by water utilities represent a significant obstacle to obtain meaningful results with the DEA technique. This could represent a paradigm shift for some of the regulatory authorities currently using DEA or similar techniques, as the complexity of the method does not seem to be justified by a better analysis of the comparative performance of the different services.

Evaluating the Efficiency of Water Distribution Network Sectors Using the DEA-Weight Russell Directional Distance Model: The Case of the City of Valencia (Spain)

“…• Variable returns to scale (VRS) were assumed in some studies evaluating the efficiency/inefficiency of water supply services [3,5], the role of service quality in the efficiency [25], or regulatory aspects [26]; • Constant returns (CRS) were adopted to assess the sustainable efficiency [27] and to analyze the relationship between efficiency and management system [28] and between urban water use and wastewater decontamination systems [29]; • Variables and constant returns were used to understand the performance patterns in water utilities [6,30,31], to estimate potential savings in water distribution [32], to measure the impact of reforms in the sector [33], to assess the relevance or the type of ownership on efficiency [34,35], to study the setting of price limits [32], to incorporate qualitative indicators in water delivery [36,37], etc.…”

Operational Efficiency of Mexican Water Utilities: Results of a Double-Bootstrap Data Envelopment Analysis

“…Several studies in developing countries have pointed out that scarcity and availability of data is a major issue [19,21,22]. Furthermore, inaccurate data make the process even harder [18]. Mexico is not the exception of this trend.…”

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

“…The impact of data quality on the results of a DEA model were explored in detail by the authors in a previous paper [35], and a brief summary with the major findings is included below in order to compare them to SFA results.…”

“…A tolerance interval was established for each variable to account for uncertainty (as variables could actually take any value within this interval). Following Molinos-Senante et al [35], 81 (3 4 ) DEA scenarios were run with different combinations of values for variables. The three are the alternatives which correspond to the situations considered for each utility (DMU): favorable, unfavorable, and original.…”