Do neighboring prefectures matter in promoting eco-efficiency? Empirical evidence from China

This study investigates the spatial externality of eco-efficiency using a novel panel dataset of 191 prefectural-level cities in China covering the period 2003-2013. We apply an extended data envelopment analysis (DEA) model, while simultaneously considering metafrontier, undesirable outputs, and super-efficiency slack-based measure (Meta-US-SBM) to estimate eco-efficiency. Then, we use a two-regime spatial Durbin model to examine the spatial externality of eco-efficiency. The empirical results indicate that: 1) there are significant spatiotemporal disparities in eco-efficiency and, on average, the eco-efficiency of the eastern region is relatively higher than that of the central and western regions; 2) the kernel density estimations of eco-efficiency of different regions reveal left-skewed distributions; 3) the estimates of the two-regime spatial Durbin model indicate the presence of emulation of higher eco-efficiency by the prefectural-level cities, while the sensitivity analysis indicates that the conclusions are robust for different spatial weighting matrix specifications. The policy implications presented are based on the empirical results.