Has the emissions intensity of industrial sulphur dioxide converged? New evidence from China’s prefectural cities with dynamic spatial panel models

The convergence of industrial SO2 emissions intensity is an important inherent characteristic and prerequisite for achieving peak industrial SO2 emissions in China. In this study, China’s prefectural city panel data are employed to examine the convergence of industrial SO2 intensity across 279 prefecture-level cities during the period from 2005 to 2015. A typical fact of industrial SO2 intensity convergence is that it exhibits both temporal and spatial dependence; conventional methods for β convergence might suffer from an endogenous problem and cause biased estimates. In order to tackle the potential endogeneity, the appropriate dynamic spatial econometric models allowing for the spatial dependence and dynamics of SO2 emissions are employed to identify the β convergence of industrial SO2 intensity and the role of spatial effects. Various time intervals and spatial weight matrix are utilised to conduct robustness checks. Our results indicate the existence of spatial spillover and β convergence in industrial SO2 intensity across prefectural cities, suggesting that cities with higher industrial SO2 intensity would decrease rapidly. The convergence of the dynamic spatial panel model is also higher than both dynamic but non-spatial, and spatial but non-dynamic models, highlighting the existing characteristics of dynamics and spatial dependence. Our findings emphasise that the current state of industrial SO2 intensity, spatial spillover effects and socio-economic factors should be taken into consideration when designing reasonable allocations of SO2 intensity reduction.