Evaluation and Spatial–Temporal Difference Analysis of Urban Water Resource Utilization Efficiency Based on Two-Stage DEA Model

In the present study, a two-stage data envelopment analysis (DEA) model and spatial econometric method were employed to evaluate and analyze the utilization efficiency of urban water resources and spatial–temporal differences in cities of China. The traditional DEA model was enhanced by adopting the Shannon entropy in the first stage. After selecting variables based on the previous step and the Bayes information criterion (BIC), redundant variables were removed. In the meanwhile, a comprehensive efficiency score (CES) was generated to rank the efficiency. Finally, spatial econometric analysis was applied to explore the spatial–temporal differences of urban water resource utilization efficiency. Results demonstrate that: 1) according to the calculations and analysis, communities should concentrate on increasing investment in equipment and technology that can help enhance water consumption efficiency, while overlooking some minor aspects, such as per capita gross domestic product (PCGDP); 2) most cities have poor water resource utilization efficiency (low CES). However, both input and output have much room for the improvement; 3) Lhasa, Beijing, Haikou, and Shanghai have high CES, indicating that the utilization efficiency of water resources is not entirely dependent on economic development; and 4) through performing the Lagrange multiplier (LM) test, the spatial error model (SEM) test is passed at the significant level of 5%. Moreover, the water resource utilization efficiency of a city may be enhanced with the economic development in neighboring cities.