Incentive Effects of the R&D Expense Super Deduction Policy for High-tech Enterprises:Based on an Extended"Counterfactual"Model
During the 11th group study session of the Political Bureau of the 20th CPC Central Committee,General Sec-retary Xi Jinping pointed out that scientific and technological innovation can generate new industries,new models and new momentum,and is the core element for developing new quality productive forces.The development of new quality productive forces hinges on the simultaneous advancement of scientific and technological and industrial innovations,thereby cultivating new momentum and new advantages.Among the various technology incentive policies,tax incentives are common policy instruments,such as the R&D expense super deduction policy,which aims to promote the develop-ment of high-tech industries.The R&D expense super deduction policy was first issued in 1996,and at the beginning of the policy,it was limited to state-owned and collective industrial enterprises.Until 2003,there was a significant change in the policy,and the appli-cation of the R&D expense super deduction policy was extended to all kinds of enterprises,even private and foreign enter-prises.Therefore,we can treat the year 2003 as a cut-off point,and data before and after the implementation of the policy can be used to estimate the impact on the innovation of non-state-owned industries.The problem in the analysis is not ap-propriate for the traditional"counterfactual"framework.Our research object is the high-tech industry affected by the su-per deduction policy,commonly referred to as the treatment group.According to the"counterfactual"method,it is neces-sary to find the control group in traditional industries that are not affected by the policy.This paper is based on the panel data policy evaluation method proposed by Hsiao et al.(2012)and innovatively expands the policy intervention evalua-tion without a control group.It examines the impact of the R&D expense super deduction policy on R&D investment and innovation output.By utilizing variables related to R&D expense that are not affected by the R&D expense super deduc-tion policy,the study simulates the potential path of R&D expense in non-state-owned high-tech industries in the absence of policy intervention—a counterfactual scenario.By comparing the differences between the actual path and the potential path,this paper evaluates the effectiveness of the R&D expense super deduction policy at the industry level.Compared with previous literature,this paper has three major contributions.First,we introduce the"counterfactual framework"into the study of the effect of the R&D expense super deduction policy,expanding the treatment effect model of Hsiao et al.(2012),using the information from variables unrelated to the policy but related to R&D to estimate the po-tential path of the treatment group,and effectively estimating the width and depth effects of the policy.Second,given the model we used,we also employ the permutation test to test policy independent variables and provide statistical support.Third,utilizing available industry-level panel data at the meso level,this study demonstrates that the R&D expense super deduction policy exerts a significant breadth effect but a weaker depth effect on China's non-state-owned high-tech indus-tries.This research serves as a valuable supplement and extension to existing studies.This study finds that the expansion of the 2003 R&D expense super deduction policy to non-state-owned enterprises significantly boosted both the scale and the relative growth rate of their R&D investment.However,the increase in scale was more pronounced,while the relative growth rate was slightly weaker.This suggests that the super deduction policy has both a breadth effect and a depth effect on R&D investment in China's non-state-owned high-tech industries,al-though the depth effect still needs to be optimized.The policy also has a significant effect on innovation output that lagged two or three years.Based on the analysis of the growth rate of patent applications and patent inventions,we be-lieve that the super deduction policy has both significant width and depth effects on the patent output of non-state-owned high-tech industries,while the depth effect is relatively weaker.
万方数据
