Does centralisation clean up? quasi-experimental evidence from China’s environmental reforms
利用中国省级环境监管集权改革作为准自然实验，研究发现中央集权通过减少精英俘获、纠正数据操纵和内部化跨辖区外部性，显著降低了颗粒物浓度，对污染严重和腐败深的地区效果更明显。
Abstract This study examines whether centralisation of environmental regulatory authority improves air quality by addressing governance failures in decentralised systems. Exploiting the staggered provincial implementation of China’s environmental regulatory centralisation reform as a quasi-natural experiment, we employ a difference-in-differences framework with granular grassroots-level data to identify causal effects. Results demonstrate that centralisation substantially reduces particulatematter concentrations through three mechanisms: reducing elite capture by insulating decisions from local networks and corruption, correcting incentive-driven data manipulation as evidenced by convergence between satellite and official measurements, and internalising cross-jurisdictional externalities by aligning regulatory scope with pollution diffusion. Heterogeneity analysis reveals that pollution reductions are concentrated in regions with greater pollution severity and deeper corruption, whereas differences in economic development and industrial structure play a comparatively modest role. These findings advance institutional economics by providing causal evidence that governance structure reforms addressing elite capture and principal-agent problems can generate marked environmental improvements, with implications for regulatory design in developing economies facing weak local institutional capacity.