The Supply Chain Innovation and Application Pilot Policy and Corporate Green Innovation: Evidence from Chinese A-Share Listed Firms

This study reassesses whether China’s 2018 Supply Chain Innovation and Application Pilot Policy changed corporate green innovation. The policy designated both 55 pilot cities and 266 pilot enterprises. To resolve ambiguity in treatment assignment, firms headquartered in pilot cities constitute the primary treatment group, while official pilot-enterprise status is analyzed separately. The updated panel contains 33,473 firm-year observations for 4265 A-share listed firms from 2011 to 2023; the main specification links covariates at year t to patent applications at t + 1, so outcomes extend through 2024. Firm and year fixed effects are included, and inference is clustered by city in the pilot-city design. Pilot-city exposure is associated with a 0.050 increase in log(1 + green invention patents) (p = 0.037), equivalent to a 5.16% change in one plus patent counts, or approximately 0.271 additional patents at the post-policy control mean. The corresponding pilot-city coefficient for utility model patents is −0.041 (p = 0.060). A joint event-study test does not reject equal pre-policy trends (p = 0.653), although the invention effect attenuates in the latest outcome cohort. Official pilot-enterprise exposure is associated with positive coefficients for both patent types. Robustness is mixed: the pilot-city invention result is directionally consistent under negative-binomial estimation and entropy balancing, but is not significant under PPML, industry-by-year fixed effects, or propensity-score matching. Exploratory channel tests do not support talent structure, lagged technology novelty, or information transparency as candidate pathways. Pilot-city exposure is associated with a higher share of cited patents but not with significantly lower environmental protection costs. The evidence therefore supports a narrow conclusion: the pilot-city estimates are consistent with a modest reallocation toward invention-oriented and more frequently cited green innovation, but the average causal magnitude and causal channels remain uncertain.