Unlocking productivity growth through digital transformation’s dual matching-learning mechanism: Evidence from Chinese listed enterprises’ TFP analysis

This study systematically investigates how digital transformation (DT) enhances enterprises’ total factor productivity (TFP) through a novel “matching-learning dual-mechanism” framework, bridging gaps in the existing literature. Unlike prior research focusing on singular mechanisms like transaction cost reductions or knowledge spillovers, this study integrates static resource allocation efficiency (matching) and dynamic technology spillovers (learning) to reveal their synergistic impact on TFP. Using a multidimensional DT index derived from textual analysis of Chinese A-share enterprises’ annual reports (2007–2019), this study captures micro-level digital behaviors across infrastructure, technologies, platforms, and industries, thereby overcoming traditional ICT-centric limits. Subsequently, this study empirically examines DT’s influential mechanism on enterprises’ TFP using data from Chinese listed enterprises on the A-share market between 2007 and 2019. The major findings are as follows: (1) DT significantly improves enterprises’ TFP, a conclusion that remains robust after a series of rigorous robustness and endogeneity tests. (2) Mechanism analysis further reveals that DT enhances TFP by mitigating resource misallocation within industries. Additionally, it facilitates the “learning by doing” effect, amplifying technological innovation spillovers and thereby further boosting TFP. Finally, this study suggests that governments should intensify the construction and investment in digital infrastructure, advance institutional reforms, and support digital technological R&D practices.