Text measurement analysis of green and low-carbon industrial policies through machine learning: insights from the Chengdu–Chongqing economic circle

Green and low-carbon industrial policies serve as a key guideline for China’s pursuit of its dual-carbon goals. Green and low-carbon industrial policies, therefore, play a crucial role in advancing green technology and fostering the sustainable transformation of traditional industries. This study examines the Chengdu–Chongqing economic circle as a case study, employing a machine learning model and the policy modelling consistency index to assess the intrinsic value, evolutionary characteristics, and policy effectiveness of green and low-carbon industrial policies within the region. Findings reveal the following: (1) Three categories of policy objectives and eight types of policy instruments were identified; (2) Although the number of policy objectives is similar, both Chengdu and Chongqing primarily rely on demand-side policy instruments; (3) Over the three developmental stages of green and low-carbon industrial policy, policy priorities have slowly shifted from theoretical guidance to implementation control; (4) Although the policy effectiveness in Chengdu is slightly lower than in Chongqing, it exhibits greater stability. Enhancing policy level could substantially improve effectiveness; (5) Notably, whilst demand-side instruments are the most frequently employed in the Chengdu–Chongqing economic circle, our analysis of three policy centres indicates a reliance on both supply-side and environmental instruments, a pattern that may be influenced by China’s unique policy landscape. This study not only provides policy recommendations for policymakers but also expands the field of industrial policy evaluation in the carbon-neutral era.