研发补贴还是税收激励——政府干预对企业创新绩效和创新质量的实证分析

政府的税收优惠和研发补贴是政府支持企业创新的重要方式。但目前的学术研究并不能够回答“哪种支持方式更有利于企业创新”。本文以此为切入点，分析政府税收优惠和直接研发补贴对企业创新绩效和创新质量的影响。采用505家高新技术企业2011-2013年调研数据为样本，对假设进行了检验。结果表明政府的税收优惠政策有利于企业创新绩效和高质量创新产出的提升；相反政府的研发补贴对企业的创新绩效和创新质量均具有挤出作用；同时，对于技术能力比较强的企业，税收优惠对高质量创新产出的促进作用将越弱。本文的结论从理论上丰富了已有创新政策的研究，实践中为政府更有针对性的实施创新政策提供了有益参考。

R&D subsidies or tax incentives: An empirical analysis of government intervention on enterprise innovation performance and innovation quality

Nowadays, it becomes even more necessary for enterprises to obtain powerful technology ability to achieve innovation-driven transformation and upgrading. However, innovation is a high-input, high-risk, long-cycle activity with externalities, and enterprises innovation investment is often insufficiently motivated. Government intervention in innovation activities is an important supplementary mechanism to solve market failures and insufficient enterprise innovation. Supports from the government can help enterprises to reduce R&D costs, devote resources in R&D projects that will not be conducted without funding, and promote R&D investment in fundamental areas or focus on core competence technologies. Besides, government subsidies may have a “cost reduction effect” or “financing effect” on enterprises, alleviating the pressure of opportunity cost for enterprises which engaged in R&D activities, and diversifying the operational risks brought by financing. However, conclusions of existing studies about the effect of government intervention on innovation activities are contradictory. Some scholars believe that government intervention is conducive to promote R&D, while some believe that government intervention on innovation activities will distort innovative price, resulting in some fundamental effects. This article argues that the contradictory conclusion may be due to the fact that existing studies have not subdivide methods of government intervention on innovation activities. Therefore, the analysis of the heterogeneous effects of different types of innovation policies will be a critical opportunity to unlock the inconsistent conclusions of current studies.Based on the existing studies of innovation policy types, this article explores the impact of direct and indirect innovation policies on enterprise innovation performance. The direct innovation policies focus on the government R&D subsidies, while indirect innovation policies focus on tax incentives. R&D subsidies are innovative policies that are obtained directly from the government, it refers to the use of fiscal policies by the national and local governments to provide financial funds to specific groups, organizations and enterprises. Therefore, enterprises have very strong intentions to obtain government subsidies to conduct R&D activities. Tax incentives are indirect subsidies provided by the national government to give taxpayers various preferential tax treatments by adopting a tax system that is contrary to the structure of the current system. As an afterward incentive method, tax incentives have a relatively free mechanism. In our study, we further subdivide the dimensions of innovation performance and distinguish different levels of quality innovation. High-quality innovation is a substantial innovation, which is measured by the number of invention patents, while low-quality innovation is more like a strategic innovation, which is measured by the sum of appearance patents and utility model patents. We come up with the hypotheses that tax incentives are more conducive to innovation output, especially the high-quality innovation output; while R&D subsidies go against both high-quality and low-quality innovation output. At the same time, we also hold the idea that the moderating effect of enterprise technology ability on the two types of innovation policies.The data in this paper comes from the investigation of high-tech enterprises in Hebei Province by the Torch Center of the Ministry of Science and Technology, P.R. The dataset includes all the high-tech enterprises in Hebei province from 2011 to 2013, we finally have 505 high-tech enterprises panel dataset, excluding missing information samples in all the years. On this basis, we also manually collect the patent information of these 505 companies from 2012 to 2014. In order to ensure the order of causality, the patent data are one-year lags behind other company information. The hypothesis was verified using the panel Poisson model.From the perspective of the heterogeneous impact of different types of government innovation policies on enterprise innovation performance, this paper analyzes the impact of tax incentives and government R&D subsidies on enterprise innovation performance and innovation quality. The conclusion shows that tax incentives, as an afterward incentive measure, will affect the business process of enterprises and benefit the stable R&D investment of enterprises, presenting a positive effect on the innovation activities and high-quality innovation output. However, the effect of government R&D subsidies on enterprise innovation is completely opposite. As a prior incentive method, government R&D subsidies will generate an interference on enterprise innovation activities, distorting innovative price, inducing enterprises’ rent-seeking of innovation subsidies and dependency on the government, resulting in a negative effect on innovation performance and innovation quality. This article further discusses the differences in the effect of different policy types on high-quality and low-quality innovation in companies with different levels of technology ability. The results show that there is a significant interactive effect between enterprise technology ability and tax incentives. The stronger the technology ability, the weaker the effect of tax incentives on high-quality innovation output, enterprises are more likely to follow the existing technology pattern and less depend on the government policies. Therefore, the motivation for enterprises to obtain preferential government policies depends on the differences in enterprise technology ability, the stronger the enterprise technology ability, the smaller the effect of government policies.This paper mainly contributes to the area in the following aspects: First, under the background of enterprise transformation and upgrading, the innovation policy of the government is a compensation mechanism for the lack of intellectual property protection, but the effect of this government policy is uncertain. From the perspective of policy types, we subdivide the heterogeneous effects of different policies, and verify the “squeezing out” effect of direct subsidies and the “crowding in” effect of indirect tax incentives. All of our findings provide proof for understanding Chinese government innovation policies. Second, researches on interactive effects expand the theoretical boundaries in this area. Taking enterprise technology abilities as a starting point, we find the influencing factors and internal mechanism of policies, which provides a very useful reference for the government to find suitable methods to inspire innovation. Third, the impact of government subsidy policies on patent quality has been rarely explored in current situation. This paper draws on existing studies, distinguishes high-quality innovation output and low-quality innovation output, and finally finds the incentive factors of different innovation motives for enterprises from a policy-making perspective.