| Innovation is the fundamental driving force and important support for highquality economic development in China.The requirements for improving the innovation capability of enterprises in the 14 th Five-Year Plan clearly stated that the nation should further support innovative small,medium and micro enterprises to grow into an important birthplace of innovation,and promote the financing innovation of large,medium and small enterprises.Scientific and technological small and medium-sized enterprises(S&T SMEs)are small and medium-sized enterprises that specialize in scientific and technological research and development activities,create independent intellectual property rights and transform them into high-tech products or services,so as to achieve long-term sustainable development.They are the reserve force of high-tech enterprises and the important force and source of national technological innovation,and its healthy development is one of the important signs of high-quality economic development."Knowledge" is the core competitiveness of S&T SMEs.However,S&T SMEs have the defects of relatively small scale and insufficient innovation resources.It is an inevitable choice for enterprises to carry out R&D activities by relying on the specialized social division and cooperation of enterprise clusters.Specialized social division and cooperation are not only conducive to the formation of regional industry scale advantages,but also promote the exchange and dissemination of knowledge.At the same time,S&T SMEs are in the same market environment and R&D system.Leading by the R&D of large enterprises in the technology field,the R&D goals of S&T SMEs in the same technology field have some similarities.With the exchange and dissemination of knowledge among enterprises,knowledge spillovers are critical to the innovation and development of enterprises.Therefore,it is of practical significance to identify and estimate the knowledge spillover effects among S&T SMEs to accurately implement policies to promote enterprise innovation and development.MAR externality(the abbreviated as Marshall-Arrow-Romer externality),which is also called specialized externality,holds that there is a close logical relationship between knowledge spillover and enterprise agglomeration,and emphasizes that knowledge can only spread among enterprises with similar technology.From the perspective of MAR externalities,the most direct reason for knowledge spillovers among S&T SMEs is geographical proximity.In the enterprise network of the cluster area,knowledge spillover effects can occur in the form of mutual imitation of production and the flow of R&D factors among enterprises.Previous literatures show that with the development of communication technology and transportation technology,knowledge spillovers may break through spatial locality.Geographical proximity may not explain the phenomenon of knowledge spillovers,and the knowledge spillovers of MAR externalities also come from the proximity of knowledge technology.Previous literatures also shows that knowledge spillovers can promote the innovation activities of enterprises,and knowledge spillovers are not bidirectional symmetric.Therefore,there may be asymmetric knowledge spillovers between high-tech S&T SMEs and non-high-tech S&T SMEs.Therefore,it is necessary to reveal the mechanism and characteristics of MAR type knowledge spillovers of S&T SMEs from the dimensions of geographical space and technology space,and to explore the role of asymmetric knowledge spillovers between high and non-high enterprises in S&T SMEs on innovation outputs.There is a relative lack of literature to explore these problems,so it has certain theoretical value to solve these problems.Under the framework of MAR externality theory,this paper makes an in-depth exploration on the spillover effects of explicit and implicit knowledge between the technology fields of S&T SMEs in the Yangtze River Delta by using the warehousing data of S&T SMEs in 2018.The main research contents are as follows:Firstly,from the perspective of enterprise agglomeration and technology similarity,the mechanism of explicit and implicit knowledge spillovers is analyzed;Then,from the perspective of knowledge spillovers,this paper studies the impact of different types of knowledge spillovers between high and non-high enterprises on innovation outputs;Finally,according to the empirical results,this paper puts forward some policy suggestions to promote the benign knowledge spillover among S&T SMEs.In the use of research methods,in addition to theoretical mechanism analysis,this paper also comprehensively uses statistical and mathematical methods such as DO index,text similarity analysis,spatial economics analysis and quantile policy processing effect model.Specifically,DO index is used to describe the agglomeration mode of S&T SMEs in various fields in the Yangtze River Delta,and the spatial weight matrix is constructed according to the agglomeration mode.On this basis,the knowledge spillover effect between enterprises from the perspective of agglomeration is studied;This paper constructs the enterprise technology similarity matrix through TNIC text analysis,and studies the knowledge spillover effect between enterprises from the perspective of technology similarity;Different types of knowledge spillovers indicators are constructed through the technical similarity matrix,and various knowledge spillover indicators are introduced into the model.Through the quantile policy treatment effect model,the impact of high-tech enterprises’ recognition policies on the policy depth and policy breadth of R&D outputs is analyzed,and the fact that there are asymmetric knowledge spillover effects between high-tech enterprises and non-high-tech enterprises is obtained;Finally,the impact of asymmetric spillovers on innovation output is analyzed by econometric model.In addition,in order to avoid the deficiencies of using cross-sectional data to research knowledge spillovers,this paper constructs the R&D output index and estimates the R&D capital stock on the basis of statistical facts.This paper is divided into seven chapters.The first chapter briefly introduces the research background,research content and the structure of this paper,as well as the innovation,main contributions and shortcomings of this paper.The literature review in Chapter 2 combs the literature related to this study and reviews the existing research.Chapter 3 is the theoretical basis of modeling,estimation and analysis.Chapter 4 to Chapter 6 are the empirical part of this paper,Chapter 4 and Chapter 5 respectively study the knowledge spillover effects among S&T SMEs under agglomeration and technology similarity.Chapter 6 studies the impact of asymmetric knowledge spillover between high-tech enterprises and non-high-tech enterprises on innovation output from the perspective of knowledge spillover.Chapter 7 is the conclusion and policy recommendations of this paper.After theoretical and empirical analysis,the conclusions are as follows:First,from the perspective of agglomeration,there are different forms of knowledge spillovers among enterprises in the agglomeration area of the S&T SMEs in the Yangtze River Delta,and the agglomeration of various technology fields may not necessarily promote knowledge spillovers.The relationship between enterprise agglomeration and knowledge spillovers in different technology fields is different.only when the number of the S&T SMEs in the technology field in the region becomes large or have technological advantages,in the regional technology field becomes large or the technology has advantages,there will be explicit knowledge spillover effects among the clustered enterprises.When the S&T SMEs are clustered,R&D manpower investment has a positive significant tacit knowledge spillover effect.At the same time,due to the universal R&D incentive policy and the "passive" nature of R&D investment of the S&T SMEs,R&D capital investment of the S&T SMEs have crowding out effect,making R&D capital unable to drive significant positive knowledge spillovers among enterprises.Therefore,except the fields of biology and new medicine,the tacit knowledge spillover effect driven by R&D capital investment of enterprises in other technology fields is mostly positive but not significant,and the significant results are basically negative.Second,from the perspective of technological similarity,the S&T SMEs in the Yangtze River Delta also have the externality characteristics of MAR technology based on technological similarity.Enterprises with similar technologies in the field of technology are more prone to positive explicit knowledge spillovers and positive tacit knowledge spillovers brought about by human capital.To a large extent,there is no excessive concentration of explicit knowledge among the S&T SMEs in the Yangtze River Delta,and the "crowding out effect" of R&D capital investment makes the tacit knowledge spillover effect among enterprises with similar technology are negative.The knowledge spillover effects between the S&T SMEs in the Yangtze River Delta and the Pearl River Delta economic circles are similar,and there is a significant positive explicit knowledge spillover effects,and the explicit knowledge spillover effect in the Pearl River Delta is relatively more obvious.In contrast,the explicit knowledge spillover effect of the S&T SMEs in Beijing-Tianjin-Hebei economic circle is not obvious.The "crowding out effect" of R&D capital investment exists in the technical fields with high technical barriers,and the tacit knowledge spillover brought by R&D capital investment is negative;The tacit knowledge spillover caused by R&D human capital investment is significantly positive in the Yangtze River Delta and the Pearl River Delta economic circle as a whole,while the tacit knowledge spillover caused by R&D human capital in Beijing-Tianjin-Hebei economic circle is negative.Third,the innovation ability of the Yangtze River Delta S&T SMEs determines the role of the high-tech enterprises recognition policy in improving their R&D output.The high-tech enterprises recognition has a generally obvious promoting effect on the S&T SMEs with stronger innovation ability.The breadth of the hightech enterprises recognition policies in different technology fields is different.In addition,there is asymmetric knowledge spillovers between high and non-high-tech enterprises.Compared with non-high-tech enterprises of S&T SMEs,high-tech enterprises can benefit more from explicit knowledge spillovers and tacit knowledge spillovers brought by technological personnel.High-tech enterprises will "squeeze out" non-high-tech enterprises’ passive R&D investment,and vice versa.For S&T SMEs,the R&D output is related to their innovation ability,and the promotion effect of enterprises on high quantiles is more obvious;There is an obvious asymmetry between the knowledge spillovers of high-tech enterprises and non-high-tech enterprises.The reasons why high-tech enterprises have more R&D output than non-high-level enterprises is not only that high-tech enterprises can benefit more from explicit and tacit knowledge spillovers than non-high-tech enterprises(less "squeezed out"),but also high-tech enterprises will more "squeeze out" the "passive" R&D capital investment of non-high-tech enterprises.The innovation of this paper mainly includes the following three aspects:firstly,it is the first time to use large-scale micro enterprise data sets to study existence and mechanism of explicit knowledge spillovers and tacit knowledge spillovers among S&T SMEs from the perspective of enterprise agglomeration;Secondly,the text analysis method is used to construct the technology similarity matrix for the first time,and the knowledge spillover mechanisms of the S&T SMEs is studied from the perspective of technology similarity;Thirdly,this paper looks for the instrumental variables of high enterprise identification policy,and discusses the heterogeneous impact of high enterprise identification on R&D output of S&T SMEs.At the same time,this paper analyzes the impact of asymmetric knowledge spillovers between high-tech and non-high-tech enterprises on the R&D output of S&T SMEs under the high-tech enterprise recognition policy. |
PDF Full Text Request