Although the efficiency of the water sector has been studied extensively utilising data envelopment analysis (DEA), the literature tends to use the conventional DEA model to compute efficiency scores. However, conventional DEA input/output data may contain random errors, which may result in distorted efficiency frontiers due to statistical noise. Bias-correcting double-bootstrap DEA came into being because of this shortcoming in the conventional DEA approach. This study joins a growing number of studies using bootstrapping DEA to correct efficiency scores. Most importantly, little is known about the comparative performance of conventional DEA versus bias-corrected DEA. A conventional model is deterministic in nature and yields biased efficiency scores. To determine the bias-corrected efficiency scores of rural and urban water utilities in South Africa, this study uses a robust, non-parametric DEA model to generate them. The truncated double-bootstrap regression results give insight into the drivers of efficiency. We found that there are significant differences between the rankings and efficiency scores generated by the conventional DEA model compared to the double-bootstrap DEA model, for both urban and rural samples. The regression model found location and the ratio of metered to unmetered connections to be significant determinants of efficiency for both urban and rural water utilities. Non-revenue water is a significant explanatory variable for urban utilities only. The number of consuming units mattered for the rural utilities only.
by 