Re-examine the air pollution control effect of China's pollutant emissions trading system: Evidence from a multi-period causal inference approach

The literature ignores the characteristics of the gradual implementation of the pollutant emissions trading system (PETS) in China and lacks studies investigating whether the PETS meets the quasi-natural experimental conditions. These may lead to biased results when using the single-period difference-in-difference (DID) model to investigate the pollution control effect of PETS. Therefore, this study first investigates whether PETS satisfies the quasi-natural experimental conditions by examining the parallel trend hypothesis, randomness grouping hypothesis, and SUTVA hypothesis. Second, this study identifies the timing of PETS document issuance in different regions. This study adopts a multi-period DID model to investigate the haze pollution control effect of PETS, then employs the single-difference model, the counterfactual method, and the explained variable substitution method to test the robustness. Finally, this article further explores the persistence and heterogeneity of the haze pollution control effect of PETS. The results showed that PETS met the quasi-natural experimental conditions, and PETS effectively reduced the intensity of haze pollution by about 1.6%. The haze pollution control effect of PETS was significant between 2007 and 2011, and insignificant between 2012 and 2016. Overall, the control effect diminished over time. Second, the haze pollution control effect of PETS was heterogeneous, and the pollution control effect was greater in the pilot sites with severe haze pollution. These findings lead to a better understanding of the implementation of the PETS in China, and the conclusions are beneficial to promoting the PETS and haze pollution control.