Environmental protection tax and green total factor productivity in China’s livestock enterprises: causal inference based on double machine learning

Livestock enterprises are the core actors in the livestock industry, and the Environmental Protection Tax (EPT) plays a critical role in driving their green transformation. Addressing the "environment–efficiency " trade-off and unlocking the policy dividends of EPT remain urgent research priorities. Using data from Chinese publicly listed livestock firms spanning 2007–2024, this study treats the EPT policy as a quasi-natural experiment and employs a double machine learning framework to investigate the impact and underlying mechanisms of EPT on livestock enterprises' green total factor productivity (GTFP). The results reveal that EPT significantly enhances GTFP. Mechanism analysis demonstrates that EPT promotes GTFP through two channels: green innovation and green mergers and acquisitions, while corporate digital transformation exerts a significant positive moderating effect on this relationship. Heterogeneity analysis further indicates that EPT's positive effect on GTFP is particularly pronounced for livestock and poultry farming firms, feed production enterprises, privately owned companies, and firms located in China’s western regions. Finally, economic consequence analysis confirms that EPT implementation reduces livestock enterprises’ market value and intensifies industry competition. Collectively, these findings not only clarify the pathways through which EPT improves GTFP in China’s livestock sector but also offer critical policy insights for optimizing environmental tax design and advancing sustainable livestock development.