Abstract: By making use of the data of A-share listed companies in Shanghai and Shenzhen stock exchanges from 2003 to 2018 and employing the stochastic frontier analysis method (SFA) to estimate and decompose the total factor productivity (TFP) of listed companies, this paper investigates the impact of R&D subsidies on corporate overall TFP and the four components that constitute TFP, i.e., the frontier technological progress, the technical efficiency, the scale efficiency and the allocation efficiency. The findings indicate that R&D subsidies cannot significantly promote the overall TFP growth. In terms of the internal structure of TFP, R&D subsidies have significantly promoted the frontier technological progress and significantly inhibited the improvement of allocation efficiency, but no significant impact on technical efficiency and scale efficiency. Further analysis in groups according to the nature of enterprise ownership, the financing constraints, the industry characteristics and the level of intellectual property protection reveals that R&D subsidies have significantly promoted the frontier technological progress of non-state-owned enterprises, enterprises with a higher degree of financing constraints, enterprises in non-high-tech industries, enterprises in fiercely competitive markets, and enterprises in areas with lower levels of intellectual property protection. However, the allocation efficiency of enterprises with lower levels of financing constraints, enterprises in non-high-tech industries and enterprises with lower level of market competition is significantly reduced. The conclusion of this paper has some enlightenment for the improvement of the R&D subsidy policy and the improvement of corporate total factor productivity.

Key words: R&D subsidies, total factor productivity, structural decomposition