Abstract
This paper compares efficiency scores from the data envelopment analysis (DEA), stochastic frontier analysis (SFA), and stochastic non-parametric envelopment of data (StoNED) methods in the context of regulating South African water utilities. We estimate cost efficiency based on cross-sectional data from 102 South African water utilities in the period 2013/14. We compare the impact of methodological choices on the efficiency scores. For StoNED, we compare scores from two different estimation techniques (method of moments and pseudolikelihood). We also use the na & iuml;ve method of averaging (NMA) and compare its results to those generated by the other methods. Results show that the choice of method has an impact on efficiency estimates and monetary cost reduction targets. When water sector regulators in the developing countries intend to benchmark utilities based on each of these methods, heterogeneity and the operating environments of utilities should be considered before deciding on the estimation method.
