Repeated Modifications in Policy Diffusion: Evidence From River Management Regulations in China's Cities

ABSTRACT Why do local governments repeatedly modify policies after their initial adoption? While policy diffusion research has focused on adoption, far less is known about why and how policies are revised over time. This study analyzes a city–year dataset of 293 Chinese prefecture‐level cities from 1986 to 2022, tracking both the adoption and subsequent modifications of River Management Regulations (RMR). A two‐stage modeling strategy combining a Probit model and a Prentice–Williams–Peterson Cox model is employed to analyze repeated modifications. The findings reveal three patterns. First, policy modifications at the provincial level and amendments to superior laws significantly increase the likelihood of local RMR revisions, whereas central‐level modifications do not exert a direct effect. Second, while neighboring jurisdictions' policy modifications do not significantly affect local revisions, adoption by neighboring late adopters generates a positive effect, indicating a form of “reverse diffusion.” Third, policies issued by local legislative branches are more likely to be modified than those issued by executive agencies, highlighting the role of institutional structure and procedural constraints. By shifting attention from adoption to repeated modification, this study extends policy diffusion theory and contributes to a broader understanding of post‐adoption policy dynamics in hierarchical and decentralized political systems.