技术溢出的空间计量和阈值回归分析
发布时间：2018-07-22 12:27
【摘要】： 技术的溢出有多种途径,按来源分有国际的技术溢出和国内的地区间技术溢出,按渠道分可以依赖干中学、看中学、竞争中学、RD中学或人员交流,不同的行业有不同的渠道特点和溢出强度。技术是否能够通过这些途径溢出依赖于吸收方的技术水平或人力资本水平,当他们的技术吸收能力满足一定的门槛条件,才能产生正的技术溢出效应。技术溢出由于人员流动、信息传播、产品贸易等原因存在空间相关,且技术溢出存在与地理距离相关的传播成本,这正是技术产业高度聚集的原因。技术溢出的研究虽然很多,但是以上领域的研究存在很多空白和分析方法上的不足。我们利用阈值回归来确定技术溢出的门槛条件,利用并改进空间计量方法,分析技术的空间相关性分析估计技术溢出强度随地理距离变化的模式,这些新方法的应用是已有研究所未做的。 准确衡量技术是分析技术溢出的基础。本文估算并比较了技术的三种主要衡量形式:索洛剩余法TFP、DEA Malmquist指数法TFP和专利数据。通过分解经济增长投入因素的贡献,发现技术对经济增长的驱动能力越来越强,1997年前主要表现为技术效率提高,1997年后纯技术进步的作用显著而稳定。技术和经济活动都存在局部集聚性,技术的集聚度高于经济;两者的集聚度随时间增强,地理分布高度一致。技术的全局空间相关性低于经济正好解释了技术的强集聚性。技术先进地区和经济发达地区都偏于沿海。 阈值回归表明,以受高等教育劳动力占总劳动力百分比表示的人力资本水平存在两个门槛:4.85%和10.99%.当人力资本水平达到4.85%的门槛时,FDI对东道国技术进步负挤出效应减半;当人力资本水平超过10.99%的符号改变门槛时, FDI的负挤出效应变为正溢出效应。从整体上看,我国人力资本水平低于符号改变门槛,但这并不意味着我国不能从FDI技术溢出中获益。由于存在地区差异和显著的省际技术溢出效应,处于门槛之下的相对落后地区可依赖超越门槛的发达地区,由发达地区吸收外部先进技术,而后通过省际技术溢出将技术扩散到相对落后地区。 在内部地区间存在经济和技术差异的发展中国家,省际技术溢出和外资技术溢出同等重要。本文的研究发现来自国内的省际技术溢出比来自FDI的技术溢出贡献更大。省际技术溢出强度和技术差距的关系是U型曲线,技术门槛存在但处于值域的上界之外,因此技术差距越小则越容易发生技术溢出。干中学和RD是内部获取技术的稳健渠道。RD在吸收FDI技术溢出方面发挥重要的作用。中西部的省际技术溢出效应强于东部,落后地区同等投入的干中学和RD投入可以获得比先进地区更大的产出,说明存在省际技术追赶的可能性。在发达地区,技术进步的主要来源是干中学、本土RD以及依赖RD吸收的FDI技术溢出。 随地理距离快速下降的技术溢出效应是导致技术和经济局部集聚的原因,基于空间计量方法和Romer研发模型的分析结果表明,在一到两个省的范围或800公里内为技术的密集溢出区,此范围可用于考虑技术影响力的经济圈划分;800公里以上为快速下降区,技术溢出效应强度减半的距离为1250公里。远距离时快速递减的扩散效应和相对稳定的缪尔达尔回流效应是东西部发展不均问题的原因之一。RD资本存量与创新产出之间存在理想的正线形关系;空间外部性主要通过误差冲击的空间传递来实现;RD外部性、人力资本流动以及市场的竞争和合作是导致创新溢出的主要原因。强化以上因素作用以及促进信息高速公路、学术信息库和技术交易市场的建设将有助于区域的平衡发展。 外贸和外资的技术溢出在不同行业存在差异。在不考虑吸收因素的交叉作用时,外资对技术创新产生负的挤出效应,其挤出效应超出了其本身作为知识生产者的正面作用;内资是否发挥作用的关键在于行业市场是否充分竞争和是否重视研发的投入;外资技术溢出效应显著的部门是适宜“看中学”的行业和因为技术差距较大而未与外资直接竞争的企业;出口贸易和干中学拉动创新的关键在于生产的产品是否有较高的技术含量。 区域的专利数据由于人口流动和知识传播的原因,存在着空间相关。采用空间面板模型提高了模型的拟合度和系数的正确性。人口的流动和流动导致的人口的增量对专利创新有重要的影响,人口流入较多的地区专利创新多;经济增长对创新的影响是正面稳定的且具有一定的局部性;高校教育有正面影响且空间相关性较大,但其作用尚待发展。
[Abstract]:There are many ways of technology spillover. According to the source, there are international technology spillovers and domestic technology spillovers. According to the channel, they can depend on the middle school, the middle school, the competition middle school, the RD middle school or the personnel exchange. The different industry has different channel characteristics and spillover intensity. Technical spillovers can produce positive technical spillover effects. Technology spillovers have spatial correlation due to personnel flow, information dissemination, product trade and other reasons, and technology spillovers are related to the ground distance, which is the high technology industry. There are a lot of research on technology spillovers, but there are many gaps in the field and the lack of analytical methods. We use threshold regression to determine the threshold conditions of technology spillovers, use and improve the spatial measurement method, analyze the spatial correlation analysis of technology to estimate the technological spillover intensity with the geographical distance. The application of these new methods has not been done by existing research.
Accurate measurement technology is the basis of analysis of technology spillovers. This paper estimates and compares three main measures of Technology: Solow residual method TFP, DEA Malmquist index TFP and patent data. By decomposition of the contribution of economic growth input factors, it is found that technology is becoming more and more powerful for economic growth. The main performance of technology before 1997 is technology. Efficiency is improved and the effect of pure technological progress is remarkable and stable after 1997. There are local agglomeration in both technology and economic activities. The agglomeration degree of technology is higher than that of economy. The concentration degree of both of them is enhanced with time and the geographical distribution is highly consistent. The global spatial correlation of technology is lower than economy to explain the strong agglomeration of technology. The economically developed areas are all off the coast.
The threshold regression shows that there are two barriers to human capital represented by the percentage of the higher education labor force in the total labor force: 4.85% and 10.99%., when the human capital level reaches the threshold of 4.85%, the negative extrusion effect of FDI on the host country's technological progress is halved; and the negative extrusion of FDI when the level of human capital exceeds the level of 10.99%. As a whole, the level of human capital in China is lower than the threshold of symbol change, but it does not mean that China can not benefit from the FDI technology spillover. Because of the regional differences and significant inter provincial technology spillovers, the relatively backward areas under the threshold can depend on the developed areas beyond the threshold. The region absorbs advanced technology, and then spreads technology to the relatively backward areas through inter provincial technology spillovers.
The inter provincial technology spillovers and foreign technology spillovers are equally important in the developing countries with internal regional economic and technical differences. The study found that the inter provincial spillover from the country is more contributing to the technology spillover from the FDI. The relationship between the inter provincial technology spillover and the technical gap is the U type curve, the technical threshold exists but is in the position of the technology gap. In addition to the upper boundary of the range, the smaller the technology gap, the more easy the technology spillover is, the dry middle school and the RD are the robust channels for internal acquisition technology,.RD plays an important role in absorbing FDI technology spillovers. The inter provincial technology spillover effect of the Midwest is stronger than the East, and the same input in the dry middle school and the RD input in the backward areas can be obtained more advanced. Greater regional output shows the possibility of inter provincial technology catching up. In developed areas, the main sources of technological progress are secondary schools, local RD and FDI technology spillovers that rely on the RD absorption.
The technological spillover effect of rapid decline with geographical distance is the cause of local agglomeration of technology and economy. The analysis based on the spatial measurement method and the Romer R & D model shows that a dense spillover area in the range of one to two provinces or 800 kilometers is a dense spillover area for technology, which can be used to consider the economic circle of technical influence; 800 km. The above is a rapid descent area. The distance of the technical spillover effect is 1250 km. The diffusion effect and the relatively stable Myrdal reflux effect are one of the reasons for the uneven development of the East and West. There is an ideal positive relationship between the.RD capital stock and the innovation output. RD externality, human capital flow and market competition and cooperation are the main causes of innovation spillover. Strengthening the role of the above factors and promoting the construction of the information highway, the academic information base and the technology trading market will contribute to the balanced development of the region.
The technological spillovers of foreign trade and foreign capital are different in different industries. In the absence of the cross effect of absorption factors, foreign investment has a negative extrude effect on technological innovation. The crowding out effect exceeds its own positive role as a knowledge producer; the key of whether the internal capital plays a role is whether the industry market is fully competitive and whether the market is heavy or not. In view of the investment in R & D, the sector which has a significant spillover effect of foreign capital is an industry suitable for "watching middle school" and the enterprises that have no direct competition with foreign investment because of the large gap in technology gap; the key to export and secondary school driving innovation lies in the high technical content of the products produced.
There are spatial correlations between the patent data of the region and the reasons for the spread of knowledge. The use of the spatial panel model improves the accuracy of the model's fitting and coefficient. The population increment caused by the flow and flow of the population has an important influence on the patent innovation, and the population flow into more areas has more patent innovation, and the economic growth is the same. The impact of innovation is positive and stable, and has a certain locality. College education has positive effects and spatial correlation, but its role remains to be developed.
【学位授予单位】：华中科技大学
【学位级别】：博士
【学位授予年份】：2008
【分类号】：F224;F062.4
本文编号：2137464
[Abstract]:There are many ways of technology spillover. According to the source, there are international technology spillovers and domestic technology spillovers. According to the channel, they can depend on the middle school, the middle school, the competition middle school, the RD middle school or the personnel exchange. The different industry has different channel characteristics and spillover intensity. Technical spillovers can produce positive technical spillover effects. Technology spillovers have spatial correlation due to personnel flow, information dissemination, product trade and other reasons, and technology spillovers are related to the ground distance, which is the high technology industry. There are a lot of research on technology spillovers, but there are many gaps in the field and the lack of analytical methods. We use threshold regression to determine the threshold conditions of technology spillovers, use and improve the spatial measurement method, analyze the spatial correlation analysis of technology to estimate the technological spillover intensity with the geographical distance. The application of these new methods has not been done by existing research.
Accurate measurement technology is the basis of analysis of technology spillovers. This paper estimates and compares three main measures of Technology: Solow residual method TFP, DEA Malmquist index TFP and patent data. By decomposition of the contribution of economic growth input factors, it is found that technology is becoming more and more powerful for economic growth. The main performance of technology before 1997 is technology. Efficiency is improved and the effect of pure technological progress is remarkable and stable after 1997. There are local agglomeration in both technology and economic activities. The agglomeration degree of technology is higher than that of economy. The concentration degree of both of them is enhanced with time and the geographical distribution is highly consistent. The global spatial correlation of technology is lower than economy to explain the strong agglomeration of technology. The economically developed areas are all off the coast.
The threshold regression shows that there are two barriers to human capital represented by the percentage of the higher education labor force in the total labor force: 4.85% and 10.99%., when the human capital level reaches the threshold of 4.85%, the negative extrusion effect of FDI on the host country's technological progress is halved; and the negative extrusion of FDI when the level of human capital exceeds the level of 10.99%. As a whole, the level of human capital in China is lower than the threshold of symbol change, but it does not mean that China can not benefit from the FDI technology spillover. Because of the regional differences and significant inter provincial technology spillovers, the relatively backward areas under the threshold can depend on the developed areas beyond the threshold. The region absorbs advanced technology, and then spreads technology to the relatively backward areas through inter provincial technology spillovers.
The inter provincial technology spillovers and foreign technology spillovers are equally important in the developing countries with internal regional economic and technical differences. The study found that the inter provincial spillover from the country is more contributing to the technology spillover from the FDI. The relationship between the inter provincial technology spillover and the technical gap is the U type curve, the technical threshold exists but is in the position of the technology gap. In addition to the upper boundary of the range, the smaller the technology gap, the more easy the technology spillover is, the dry middle school and the RD are the robust channels for internal acquisition technology,.RD plays an important role in absorbing FDI technology spillovers. The inter provincial technology spillover effect of the Midwest is stronger than the East, and the same input in the dry middle school and the RD input in the backward areas can be obtained more advanced. Greater regional output shows the possibility of inter provincial technology catching up. In developed areas, the main sources of technological progress are secondary schools, local RD and FDI technology spillovers that rely on the RD absorption.
The technological spillover effect of rapid decline with geographical distance is the cause of local agglomeration of technology and economy. The analysis based on the spatial measurement method and the Romer R & D model shows that a dense spillover area in the range of one to two provinces or 800 kilometers is a dense spillover area for technology, which can be used to consider the economic circle of technical influence; 800 km. The above is a rapid descent area. The distance of the technical spillover effect is 1250 km. The diffusion effect and the relatively stable Myrdal reflux effect are one of the reasons for the uneven development of the East and West. There is an ideal positive relationship between the.RD capital stock and the innovation output. RD externality, human capital flow and market competition and cooperation are the main causes of innovation spillover. Strengthening the role of the above factors and promoting the construction of the information highway, the academic information base and the technology trading market will contribute to the balanced development of the region.
The technological spillovers of foreign trade and foreign capital are different in different industries. In the absence of the cross effect of absorption factors, foreign investment has a negative extrude effect on technological innovation. The crowding out effect exceeds its own positive role as a knowledge producer; the key of whether the internal capital plays a role is whether the industry market is fully competitive and whether the market is heavy or not. In view of the investment in R & D, the sector which has a significant spillover effect of foreign capital is an industry suitable for "watching middle school" and the enterprises that have no direct competition with foreign investment because of the large gap in technology gap; the key to export and secondary school driving innovation lies in the high technical content of the products produced.
There are spatial correlations between the patent data of the region and the reasons for the spread of knowledge. The use of the spatial panel model improves the accuracy of the model's fitting and coefficient. The population increment caused by the flow and flow of the population has an important influence on the patent innovation, and the population flow into more areas has more patent innovation, and the economic growth is the same. The impact of innovation is positive and stable, and has a certain locality. College education has positive effects and spatial correlation, but its role remains to be developed.
【学位授予单位】：华中科技大学
【学位级别】：博士
【学位授予年份】：2008
【分类号】：F224;F062.4
【引证文献】
相关博士学位论文 前3条
1 张晓宁;产业集聚的技术溢出研究[D];西北大学;2011年
2 张浩然;中国城市经济的空间集聚和外溢: 理论分析与经验证据[D];吉林大学;2012年
3 吴梅;广东工业产业中知识溢出效应的空间经济计量实证研究[D];华南理工大学;2012年
相关硕士学位论文 前1条
1 唐国新;我国区域金融差异及其空间计量分析[D];复旦大学;2011年
，本文编号：2137464
本文链接：https://www.wllwen.com/jingjilunwen/jingjililun/2137464.html
