Suchergebnisse

Data Envelopment Analysis (DEA) is often used by regulators to create a pseudo-competitive environment for sectors with natural monopolies. In addition to develop a theoretically well-behaved model, regulators need to take into account several other factors, such as the political agenda and the historical context of the regulation. This sometimes results in some unconventional approaches, which furthermore are not easily changed. In this paper, we discuss the model used for DEA-based benchmark regulation of the Danish water sector. More specifically, we look at the characteristics of the method the regulator uses to take into account differences in the companies' environmental conditions. We show how the approach currently used to control for differences in environmental conditions seemingly does not sufficiently control for the actual differences as intended since second stage analysis still reveals significant correlations between the efficiency scores and these external factors. To explain this, we reconsider the second stage analysis, using permutation-based approaches and also accounting for the fact that only those companies that in the DEA assign weights to those output measures adjusted for environmental conditions, will benefit from the adjustments.

Data Envelopment Analysis (DEA) is often used by regulators to create a pseudo-competitive environment for sectors with natural monopolies. In addition to develop a theoretically well-behaved model, regulators need to take into account several other factors, such as the political agenda and the historical context of the regulation. This sometimes results in some unconventional approaches, which furthermore are not easily changed. In this paper, we discuss the model used for DEA-based benchmark regulation of the Danish water sector. More specifically, we look at the characteristics of the method the regulator uses to take into account differences in the companies' environmental conditions. We show how the approach currently used to control for differences in environmental conditions seemingly does not sufficiently control for the actual differences as intended since second stage analysis still reveals significant correlations between the efficiency scores and these external factors. To explain this, we reconsider the second stage analysis, using permutation-based approaches and also accounting for the fact that only those companies that in the DEA assign weights to those output measures adjusted for environmental conditions, will benefit from the adjustments.