The Impact of Sci-Tech Finance on Renewable Energy Technology Innovation: Causal Inference Based on Double Machine Learning

Abstract

Under the global innovation-driven development strategy, sci-tech finance has emerged as a critical mechanism for channeling financial resources into technological advancement. However, whether sci-tech finance can effectively promote renewable energy technology innovation still lacks systematic empirical evidence, particularly regarding the causal mechanisms and heterogeneous effects across different regional contexts. This study addresses this gap by employing a Double Machine Learning (DML) framework, which integrates machine learning algorithms with causal inference to mitigate high-dimensional confounding bias. Using balanced panel data covering 31 Chinese provinces from 2012 to 2022, the analysis reveals that sci-tech finance exerts a statistically significant and robust positive effect on renewable energy technology innovation, with a notably stronger impact observed in photovoltaic technologies compared to other renewable energy domains. Mechanism analysis demonstrates that the agglomeration of sci-tech talent and the level of industrial upgrading serve as the primary transmission channels through which sci-tech finance fosters innovation outcomes. Furthermore, heterogeneity analysis uncovers substantial variation in the treatment effect: the promotional impact of sci-tech finance is significantly more pronounced in eastern regions, areas characterized by strong intellectual property protection, and regions exhibiting high degrees of industrial agglomeration. These findings provide rigorous empirical support for policymakers seeking to optimize sci-tech finance policies, refine regional innovation strategies, and accelerate the transition toward sustainable energy systems through targeted financial interventions.