Science Research Management

Science Research Management ›› 2021, Vol. 42 ›› Issue (4): 92-102.

Previous Articles     Next Articles

Convergence or decentralization: The scale "trap" of corporate innovation

Sun Wenhao1, Zhang Jie2   

  1. 1. School of Economics, Renmin University of China, Beijing 100872, China;  2. Institute of China Economic Reform and Development, Renmin University of China, Beijing 100872, China

  • Received:2019-12-18 Revised:2020-06-23 Online:2021-04-20 Published:2021-04-19

Abstract:  In order to speed up the flow of high-level labor factors and promote regional innovation-driven development, in recent years, major cities of China have participated in the "talents snatch war" to gather S&T talents through various preferential talent introduction policies. However, the existing literature that how the S&T talent scale influence the corporate innovation is not enough. Especially the problem that whether S&T talents converge to a few enterprises to build a large-scale S&T talent enterprise cluster or scatter among most enterprises to form a small-scale S&T talent enterprise cluster is unsolved. However, it may lead the "talents snatch war" city to gather S&T talents blindly if this problem remains unresolved, and then cause the S&T talent scale of enterprises to exceed the threshold, so as to reduce the corporate innovation performance. Therefore, this paper uses micro enterprise data of the national innovation survey enterprise database that comes from the National Bureau of Statistics of China over 2008-2014, which provides the information of various scientific and technological innovation activities of industrial enterprises, to answer whether there is a "scale threshold" between the S&T talent scale and corporate innovation performance, what is the value of " scale threshold", and which type of enterprises have obvious "scale threshold", etc.In order to answer the research question, this paper modifies the traditional static panel threshold model on the basis of predecessors, and explores the effect of the S&T talent scale on corporate innovation with more effective tool variables. Based on the empirical test results, we can conclude that as follows: (1) The scale of S&T talent has a triple threshold effect on the corporate innovation performance. The S&T talent scale with the high threshold range has a significant inhibitory effect on the corporate innovation performance. Empirical test shows that the "scale threshold" of S&T talent is roughly 341. (2) From the perspective of enterprise characteristics, compared with state-owned and foreign-funded enterprises, private enterprises have a "scale threshold" of roughly 317. When the scale of S&T talents in private enterprises exceeds the "scale threshold", the effect of S&T talents scale on corporate innovation performance is significantly negative. (3) From the characteristics of industrial organization: as for the "NN" enterprises with a large-scale S&T talent, the S&T talent scale has a significant inhibitory effect on the corporate innovation performance; as for the "LL" enterprises with a small-scale S&T talent, the S&T talent scale is beneficial to improve the corporate innovation performance. (4) From the perspective of government industrial policies: when there is a national high-tech zone in the city where the enterprise is located, the S&T talent scale has a significant promoting effect on the corporate innovation performance, and vice versa anti; as for the high-tech enterprises that obtain tax reduction, the S&T talent scale has a significant promoting effect on the corporate innovation performance, otherwise, it is not significant. Based on this, the conclusion of this paper can be summarized as follows: firstly, there is a "scale threshold" between the S&T talents and the corporate innovation performance, and exceeding the "scale threshold" is not beneficial to corporate innovation, what′s more, the "scale threshold" is roughly 341; secondly, the "scale threshold" is more obvious in private or "NN" type enterprises; finally, national high-tech zones or government tax reductions are beneficial to improve the value of "scale threshold". This paper provides the reference value for the government to improve the system and mechanism of market-oriented allocation of high-level labor factors, and suggests that "talents snatch war" cities should issue corresponding talent policies to encourage S&T talents to flow to the "LL" enterprises with small-scale scientific S&T talents, so as to avoid the "talents snatch war" city to fall into the S&T talent scale "gap". In addition, the government should speed up the upgrading process of provincial high-tech zones, and implement the "inclusive" tax reduction plan for high-tech enterprises to build a "center periphery" regional innovation model.This paper provides several marginal contributions. Firstly, on the research data level, based on the micro enterprise data of the national innovation survey enterprise database which comes from the National Bureau of Statistics of China over 2008-2014, this paper provides a new empirical data support for the triple threshold effect theory of "scale threshold" and "critical threshold". Secondly, in terms of research framework, based on the general equilibrium theory of ABBGH (2005), this paper constructs the innovation gap of enterprises, and further studies the effect of the S&T talent scale on corporate innovation performance with different innovation gaps. Finally, in terms of research methods, the existing literature has ignored the endogenous problems that may exist in the model. Based on this, this paper adjusts the traditional static panel threshold model, and studies the effect of the S&T talent scale on corporate innovation combined with more accurate tool variables, so as to solve the endogenous problems effectively that may exist in the model.

Key words: S&T talent scale, innovation performance, "scale threshold", panel threshold model