What Enables Climate Policy Innovation Under Authoritarian Regimes? Evidence from China’s Low-Carbon Pilot Cities

Climate governance depends not only on ambitious national targets but also on the capacity of local governments to innovate in policy design and implementation. This paper examines China’s Low-Carbon Pilot Cities program to explore how hierarchical governance shapes policy innovation and diffusion. Despite uniform national directives, pilot cities display significant variation in policy outcomes, underscoring the role of local leadership, intergovernmental relations, and institutional context. Drawing on policy data from 63 pilot cities (2010–2024), the study analyzes the conditions enabling genuine innovation, defined as locally tailored measures exceeding higher-level directives. Such innovations are identified through text-matching, BERT-based semantic similarity, and Large Language Model-assisted classification, and assessed with multilevel regression models at both city and province levels. Findings show that central government support initially stimulate innovation. Political will, particularly when City Party Secretaries take leadership roles, is a stronger driver than performance pressure. Over time, however, these effects diminish as procedures become institutionalized. Horizontal dynamics also matter, though less through competition than through administrative mobility, which facilitates the transfer of practices across cities. Overall, durable innovation in China’s experimental governance is driven more by commitment and institutional incentives than by coercion or tournament-style pressure. These effects are more pronounced in high-GDP cities, while much weaker in low-GDP contexts. The study contributes to broader debates on climate governance, showing that local agency within authoritarian settings is shaped by motivation, learning, and institutional design, with important parallels to democratic systems.