Purpose
– The purpose of this paper is to present a yardstick efficiency comparison of 269 Indonesian municipal water utilities (MWUs) and measures the impact of exogenous environmental variables on efficiency scores.
Design/methodology/approach
– Two-stage Stackelberg leader-follower data envelopment analysis (DEA) and artificial neural networks (ANN) were employed.
Findings
– Given that serviceability was treated as the leader and profitability as the follower, the first and second stage DEA scores were 55 and 32 percent (0 percent = totally inefficient, 100 percent = perfectly efficient), respectively. This indicates sizeable opportunities for improvement, with 39 percent of the total sample facing serious problems in both first- and second-stage efficiencies. When profitability instead leads serviceability, this results in more decreased efficiency. The size of the population served was the most important exogenous environmental variable affecting DEA efficiency scores in both the first and second stages.
Research limitations/implications
– The present study was limited by the overly restrictive assumption that all MWUs operate at a constant-return-to-scale.
Practical implications
– These research findings will enable better management of the MWUs in question, allowing their current level of performance to be objectively compared with that of their peers, both in terms of scale and area of operation. These findings will also help the government prioritize assistance measures for MWUs that are suffering from acute performance gaps, and to devise a strategic national plan to revitalize Indonesia’s water sector.
Originality/value
– This paper enriches the body of knowledge by filling in knowledge gaps relating to benchmarking in Indonesia’s water industry, as well as in the application of ensemble two-stage DEA and ANN, which are still rare in the literature.
An Examination of COVID-19 Mitigation Efficiency among 23 Countries
