Science Research Management ›› 2021, Vol. 42 ›› Issue (7): 50-59.

Previous Articles     Next Articles

Are government R&D incentives conducive to improving regional innovation efficiency?

Liu Xielin, Zhu Langmei, Yang Boxu   

  1. School of Economics and Management, University of Chinese Academy of Sciences, Beijing 100190, China
  • Received:2020-11-09 Revised:2021-05-25 Online:2021-07-20 Published:2021-07-19

Abstract:     Under the innovation-driven development strategy, the externality of R&D hinders the enthusiasm and initiative of organizational innovation, which leads to the market failure of R&D. For this reason, as with other countries, the Chinese government has adopted a series of innovative policies such as government R&D subsidies and R&D tax credits to try to solve the problem of market failure in R&D. So, how effective is the implementation of these innovation policies? Existing studies have interpreted the impact of different types of innovation policies on R&D input or output. But we know that innovation is the process of converting inputs into outputs, which involves the concept of efficiency. Therefore, only considering innovation input or output unilaterally cannot fully explain the implementation effects of innovation policies. Meanwhile, different innovation processes correspond to different outputs, which generally divide into technical outputs and commercialized outputs, and there are differences in the innovation efficiency measured by different innovation outputs. Therefore, it is necessary to investigate the implementation effect of innovation policy from the perspective of innovation efficiency subdivision. From this, our paper intends to divide innovation efficiency into technological innovation efficiency and commercial innovation efficiency according to the technology output and commercialization output of innovation and respectively explores the influence of government R&D subsidy and R&D tax credits on the two kinds of efficiency. In addition, considering the differences in the development-driven methods of various regions, the regional development-driven methods are divided into innovation-driven and investment-driven by comparing the intensity of R&D investment in each region with the average R&D investment intensity of the country. Then it explores the moderating effect of regional development driving methods on the relationship between different R&D incentive measures and regional innovation efficiency.
     To this end, this paper takes the panel data of 30 regions in China from 2007 to 2017 as samples, takes the regional R&D capital stock and labor force level as the innovation input, and takes the number of invention patent applications and new product sales revenue as the innovation output respectively. A stochastic frontier model in the form of a trans-logarithmic production function is used to test the hypotheses. Furthermore, we consider the moderating effect of the driving mode of regional development.
     The results show that government R&D subsidies have a positive impact on regional technological innovation efficiency, while have a negative impact on regional commercial innovation efficiency; R&D tax credits harm the regional technological innovation efficiency, while has no significant effect on the regional commercial innovation efficiency. Further, government R&D subsidies have no significant difference in the efficiency of technological innovation in regions with different development driving modes, and all of them have a positive impact. But there has a more strong negative effect on the efficiency of business innovation in innovation-driven regions. Meanwhile, compared with investment-driven regions, the deduction of R&D expenses has a more significant negative impact on technological innovation efficiency and positive effects on business innovation efficiency in innovation-driven regions.
    Possible innovation points of this paper are as follows: Firstly, this paper discusses the implementation effect of government R&D incentive measures from the perspective of efficiency, which complements the bias of unilateral consideration of innovation input or output research; Secondly, according to different outputs corresponding to different innovation processes, this article subdivides regional innovation efficiency into technological innovation efficiency and commercial innovation efficiency. Thirdly, according to the comparison between the R&D input intensity of each region and the average R&D input intensity of the whole country, a new regional classification method is proposed.
     The main policy implications of this study are as follows: Firstly, innovation policy factors are affecting regional innovation efficiency, such as government R&D subsidies and R&D tax credits. Secondly, regional differences in development-driven modes lead to differences in the implementation effect of R&D incentives, and policy makers need to take measures according to local conditions. Government R&D subsidies apply to nationwide promotion, while additional deduction of R&D expenses is not always valid, which is more suitable for innovation-driven areas. Finally, the research on the impact of R&D incentive measures on different regional innovation efficiencies is helpful to clarify the valid incentive objects corresponding to different types of R&D subsidies, to facilitate policymakers to take targeted schemes. Among them, government R&D subsidies are more inclined to subsidize technology R&D projects with long-term, high-risk, high-tech, and high social returns. On the contrary, the R&D expense plus deduction prefers short-term, low-risk, and low-tech commercial new product R&D projects.

Key words: R&D subsidy, R&D tax credit, innovation efficiency, innovation-driven, investment-driven