Can centralization of environmental regulations reduce firm emissions? Evidence from county-prefecture centralization reforms in China

• 成果类型: Article
• 署名作者: Cao, Xun; Wu, Mingqin
• 署名单位: Pennsylvania Commonwealth System of Higher Education (PCSHE); Pennsylvania State University; Pennsylvania State University - University Park; Pennsylvania Commonwealth System of Higher Education (PCSHE); Pennsylvania State University; Pennsylvania State University - University Park; Wuhan University
• 刊物名称: JOURNAL OF POLICY ANALYSIS AND MANAGEMENT
• ISSN/ISSBN: 1520-7122
• DOI: 10.1002/pam.22678
• 发表日期: 2025
• 页码: 964-996
• 关键词: natural experiment; decentralization; ENFORCEMENT; performance; IMPACT; TRADE; INFORMATION; COMPETITION; FEDERALISM; spillovers

摘要

Drawing on a natural experiment generated by prefecture-level centralization reforms in China in the early 2000s, we study whether a centralized regulatory system delivers better environmental outcomes in a developing country context. We examine the impact of centralization reforms on firm air pollution emissions using a difference-in-differences estimation strategy. We find that centralization reform reduces firm air pollution intensities in total waste air, SO2, and soot. This effect is robust when we control for contemporaneous environmental policy changes and SOE reforms and when we use alternative differences-in-differences estimations. We perform placebo tests to further demonstrate that the relationship is unlikely a function of a selection effect and omitted variables. Empirical tests on mechanisms reveal that pollution reduction is mainly due to increased pollution removals during the end-of-pipe treatment stage while there is little evidence that centralization increases firm scale of production, productivity, efficiency, and innovation efforts. We show that such firm responses are likely a function of increased regulatory enforcement brought by the reform. Finally, we test whether a centralization reform drives local firms away (i.e., a spillover effect) and we find no supporting evidence.