Provincial drivers of green total factor productivity in China: Policy insights from panel data analysis

China's transition to high-quality development requires decoupling economic growth from environmental degradation. This study investigates the drivers of Green Total Factor Productivity (GTFP) across 30 Chinese provinces from 2005 to 2020, specifically examining the non-linear transmission mechanisms of technological change. Employing a dynamic panel System GMM estimator to control for endogeneity and persistence, we uncover a robust U-shaped relationship between green technology innovation (GTI) and GTFP. This finding indicates that provinces must overcome an initial “innovation trap” characterized by high adjustment costs before innovation yields net productivity gains (threshold GTI ≈ 5.077). Furthermore, our results validate the Porter Hypothesis, showing that environmental regulation significantly boosts GTFP, whereas traditional infrastructure expansion and high energy intensity currently exert negative structural drags. These insights challenge linear growth models and inform a differentiated policy framework: we advocate for “bridge financing” to support lagging regions through the innovation valley and a strategic pivot from physical road density to digital and green infrastructure investments. • Dynamic analysis of Green Total Factor Productivity (GTFP) across 30 Chinese provinces (2005 – 2020). • System GMM estimation reveals a U-shaped relationship between green innovation and GTFP. • Identifies a critical innovation threshold (GTI ≈ 5.077) separating adjustment costs from productivity gains. • Environmental regulation acts as a positive driver, validating the Porter Hypothesis in China. • High energy intensity and traditional infrastructure expansion negatively impact green productivity.