【作者】 夏万军；

【导师】 纪宏；

【作者基本信息】 首都经济贸易大学 ， 数量经济学， 2008， 博士

【摘要】 经济增长收敛性问题是20世纪80年代以来经济增长理论研究最为热门的论题之一。这个论题的核心,在于考察不同经济体之间或一国内部不同地区之间经济增长速度的差异,寻找这种差异产生的原因。对于中国这样一个区域众多、经济差距明显且时有扩大的国家而言,这一论题显得尤为现实。改革开放以来,我国经济差距的缩小或扩大,其中比较重要的影响因素就是区域经济增长速度的差异。近年来很多学者研究了中国地区间经济差距问题,其中也不乏从收敛角度的研究,但是鲜有涉及基于技术扩散视角的研究。本文拟在已有研究的基础上,从新的角度来研究技术扩散与我国区域增长收敛性问题。在理论上,根据新古典经济增长理论和新增长理论两个学派关于经济增长收敛性的相关内容,研究技术扩散对我国区域经济增长差异的影响,扩宽了收敛性的研究范围。在方法运用方面,通过对技术扩散的不同处理,构建出了新的分析我国区域经济增长收敛性的模型。并在此基础上采用中国省际数据对其进行实证检验。本文在对我国区域经济增长差异进行历史研究的基础上,提出了技术扩散与经济结构相关假说,并在此基础上通过将技术进步分解成(1)反映全国范围内技术进步和劳动效率因素(如总体制度变迁等)影响的常数项p;(2)地区特别技术进步因子(q_i);(3)技术扩散;(4)随机扰动项。利用空间计量经济学的方法,提出了一个基于技术扩散的区域经济收敛理论框架,并建立了一个扩展的收敛方程。我们的扩展收敛方程表明,如果技术扩散是产业内部扩散的,经济结构的不同应被看成一个附加的因素,收敛是条件的。由于将技术进步看作是产业内部转移的,只要经济结构不同,各个地区也将收敛于不同的稳态长期劳动生产率的增长率。基于以上理论分析,本文从机制和产出两个方面进行了收敛性检验。实证检验揭示:1981～2005年,我国区域经济存在弱条件β-收敛和“俱乐部收敛”的证据,但通过对不同五年计划的考察发现各个时段的收敛速度波动较大。同时,通过实证的考察,技术扩散对经济增长收敛性有显著影响。但对于我国的不同时段和不同区域的影响显著性不同。证实我们的扩展收敛模型能够用于对我国区域经济收敛的分析考察。最后,我们在Dowrick和Rogers分析框架的基础上,构建了一个新的能综合分析新古典收敛机制和新增长收敛机制的框架。实证结构表明,改革开放以来,特别是1981～2005年期间,在我国区域经济增长收敛性既有新古典收敛机制的作用,同时也有新增长收敛机制的作用。但是在不同时期,不同的收敛机制在发挥作用。1990年以前,资本积累对收敛的作用显著有效,新古典收敛机制存在,但1990年以后,新古典机制的作用并不明显。1986～1995及2001～2005年期间,技术扩散对收敛的影响显著有效,新增长收敛机制存在。但在其它时段,新增长收敛机制不存在。更多还原

【Abstract】 Economic growth Convergence theory has become one of the most popular subjects in economic growth theories since the 1980s. The core of the subject is to study the difference of economic growth between different economies or different regions in a county and look for the the causes of these differences. For China, a large number of regions and the economic gap between different regions expanding significantly, the topic appear to be particularly realistic. The economic gap has become expand or narrow can be firstly attributed to the regional different rate of economic growth. Although there are a lot of research on China’s regional economic disparities, quite a few research analyzed using convergence theory, there seldom involved in the technology-diffusion-based research perspective. On the basis of existed research, this paper studies technology diffusion and regional economic growth convergence in China from a new perspective. In the theoretical side, according to the convergence research in the Neo-Classical Theory of Economic Growth and New Growth Theory, this paper studies influence of technical diffusion on the diversity of regional economic growth in China, Which widenes the scope of the convergence study. In the method, this paper constructs a new model to study regional economic growth convergence and estimated by means of Panel Data procedures and data on the Chinese regions over the 1981-2005 periods.Based on the historical studies of difference in Chinese regional economies growth, economic structure correlation hypothesis of technical diffusion is proposed. Technical progress is broken down into four parts: (1) constant term p is here let to reflect the effects of "country-wide" technical progress and of all the factors that influence the effectiveness of labour in all regions contemporaneously; (2) region-specific component of technical progress; (3) technical diffusion; (4) random shock. Using the "spatial econometric perspective" approach, this paper proposes a theoreticl framework and derives an "extended" convergence equation. The implication of our model is that, if technological diffusion is of an intra-sector type, economies’ structural differences need to be taken account of as an additional factor on which convergence is conditional. In other words, anumber of economies, identical in all other respects, will still not be subject toβ-convergence, unless they are as structurally similar as to share the same technology.Based on the theoretical analysis, this paper estimates the convergence from two aspects of the convergence mechanisms and outputs. Empirical studies indicate that China’s regional economic growth exist weak conditional convergence and the "convergence club" evidence over the 1981-2005 periods, but the convergence rate in different Five-year Plan fluctuates. The technical diffusion has the remarkable influence to the economic growth convergence, but it is significantly difference in different region at different time. The results bring empirical support to our approach.Finally, this paper constructes a new technical-diffusion-based framework to generalized analyses Neo-classic Growth Convergence Mechanism and the New Growth Convergence Mechanism based on the Dowrick and Rogers’ framework. The results indicate that since reform and opening up, especially during the period from 1981 to 2005, there are effects both of Neo-classic Growth Convergence Mechanism and New Growth Convergence Mechanism mechanism. However, at different times, the convergence of different mechanisms plays a role. Before 1990, capital accumulation has significant effect, the Neo-classical Mechanism of convergence exists, but after 1990, the effect of the Neo-classical Mechanism is not obvious. From 1986 to 1995 and from 2001 to 2005, the technical diffusion has significant effect. However, in other periods, New Growth Mechanism of convergence does not exist.更多还原