LAUSR

Abstract Since urban eco-efficiency is known to be influenced by the contemporaneous efficiency level in the vicinity, one must allow for possible spatio-temporal correlations between such efficiencies. This article assessed the urban eco-efficiency of 282 prefecture-level cities in China from 2004 to 2016 using a super-efficiency model based on the non-radial directional distance function (Super-NDDF) and then conducted neighborhood correlation tests using the spatio-temporal Moran’s I index (STMI). After a variety of parameter identification and econometric tests, the spatio-temporal Durbin model (STDM) and partial differential method were used to investigate the impact of the spatio-temporal neighborhood effect on urban eco-efficiency. The results showed that compared with the traditional spatial econometric model, the spatial coefficient ρ of the STDM increased by 75%, indicating that the time lag and accumulation of neighborhood effects could not be ignored. After the setting of the spatio-temporal weight matrix in the time dimension was changed, the diminishing marginal neighborhood effect on urban eco-efficiency was found as the temporal window increases and the attenuation boundary appeared within approximately five years. Additionally, in contrast to the statistical insignificance of the neighborhood effect on the strict spatial dimension, environmental regulation had a significant positive impact on the eco-efficiency of neighboring cities when the neighborhood effect was considered in a spatio-temporal context, suggesting a time lag in the neighboring demonstration effect of environmental regulation. The obvious increase in the ratio of indirect effects to direct effects indicated that the spatio-temporal neighborhood effect exacerbated the inertia of neighborhood dependence evolution and strengthened the positive spatial externalities of industrial structure upgrading and technological progress.