| Innovation is the primary driving force for development.In 2021,China’s R&D investment reached 2,786,4 billion yuan.This was mainly due to the improvement and implementation of incentive policies such as the additional deduction for enterprise R&D expenses and the reform of the management of research funds and project management.However,there are still problems of "weak quality and low efficiency" in strategic innovation,cheat and supplement innovation at the micro level and technological innovation at the macro level.Based on the above analysis,this paper takes the selective support policy of "high-tech enterprise identification"(hereinafter referred to as "Accreditation"),and verifies the existence of R&D classification shifting from the perspective of R&D input in "Accreditation",and analyzes its impact path between "Accreditation" and innovation efficiency.In this paper,the enterprises that obtained the "Accreditation" through the initial and reexamination of Shanghai and Shenzhen A-shares from 2008 to 2020 are taken as the research objects.First of all,in order to avoid the deviate problems caused by sample selection bias,use Propensity Score Matching(PSM)to find a control group with similar characteristics for identified businesses.Then,divides the innovation system into R&D investment stage and achievement transformation stage using two-stage Data Envelopment Analysis(DEA),and deconstructs the "black box" innovation system of high-tech enterprises from a system perspective.Finally,the Difference-in-Difference(DID)method is used to analyze the effect of "Accreditation" and empirically test the relationship among high-tech identification policy,R&D classification shifting and innovation efficiency.The results show that:(1)In the overall process of innovation,"Accreditation" is negatively correlated with comprehensive innovation efficiency.In R&D investment stage,"Accreditation" is negatively correlated with comprehensive innovation efficiency,and the existence of "fake high-tech" enterprises leads to policy implementation the deviation.In achievement transformation stage,there is no significant negative correlation.(2)"Accreditation" is positively correlated with classification shifting."Accreditation" makes enterprises have the behavior of R&D classification shifting.(3)In the overall process of innovation,R&D classification shifting has partial mediating effect on "Accreditation" and comprehensive innovation efficiency.Specifically,it is determined that the policy leads to influence enterprise innovation efficiency through promoting R&D classification shifting.In R&D investment stage,6.8% of the crowding out effect of "Accreditation" enterprises’ comprehensive innovation efficiency is indirectly caused by the change through R&D classification shifting.In achievement transformation stage,this mediating effect is not significant.(4)From the perspective of control right,identification has significant spillover to the innovation efficiency of non-state-owned enterprises.From the perspective of identification period,"Accreditation" has a great negative impact on enterprise innovation efficiency before identification.The empirical results of this paper show that: Before the identification and review process of high-tech enterprise in China,there is R&D classification shifting that transfers production costs to R&D investment.This paper analyzes the influence of micro-subject’s "catering" behavior in innovation,and provides reference for future reform and development. |
