| A debate has emerged in recent literature about the way underlying economic structurewithin a geographic unit that may influence the diffusion of knowledge spillovers, andpromote industrial development and growth overtime. One view suggests that knowledgecomes from firms of the same industry, which means an increase in an industry'sconcentration would facilitate diffusion of knowledge spillovers and innovative activities. Wecall this effect MAR externality, which was asserted in the work of Arrow (1962), Marshall(1890), and Romer (1986,1990). The second view argues that the most important knowledgespillovers come from firms in diverse industries and the diversity and variety of industrieswithin a geographic region promote a greater return on the exchange of economic knowledgeacross firms. We call this effect the Jacobs externality,which was suggested by Jacobs (1969).Meanwhile there emerges a third point of view represented by Porter (1990) which holds thata highly competitive climate induces firms to innovate in order to remain competitive andmonopoly hurts innovation and growth. Since the beginning of the1990s, with thedevelopment of analysis of panal data, the model of panel data has become one of most usedtool in the relevant empirical studies in the field of knowledge spillovers.This study attempts to propose three-dimensional panel data SUR spatial econometricmodels as well as the corresponding economic computational evaluation technology, in whichboth dimension of time and spatial correlation among various regions are considered. On thisbasis, first of all, this paper makes an empirical study on GuangDong's industry knowledgespillover by using relevant data. Secondly, this paper will make up for the insufficientliterature in this field in China. Finally, from the perspective of knowledge spillover, thispaper will provide strategic policy proposals of industry development in GuangDong.Specifically, the study is divided into two steps:The first step is theoretical research, which mainly accomplishes the following two tasks:sorting out relevant economic theories and empirical study literature and summarizing theachievements and limitations of the existing research results; reviewing the literature ofmathematical economics and economic computation, establishing the three-dimensional paneldata SUR spatial econometric models and using GAUSS to do the programming.The second step is empirical study based successively on the whole industry samplemodel, industrial association model, industrial monopoly model and the high technologymodel, to which three dimensional panel data spatial analysis and classic three dimensionalpanel data analysis were applied respectively. The main empirical results include:Firstly, the econometric analysis shows clear evidence of the presence of MARexternality in the development of industry in GuangDong. It suggests that an increase in anindustry's clustering would facilitate diffusion of knowledge spillovers and innovativeactivities, and do benefit to the industry. MAR externality is highly associated with theregional economy development. MAR spillover has the dominant effect in Pearl River Deltaarea.Secondly, the econometric analysis displays that Jacobs externality is highly associatedwith the regional economy development, the industrial relationship and the level of monopolyin GuangDong. Jacobs spillover is evident in non-Delta area. In the industry of high industrialcorrelation Jacobs externality shows negative or unapparent, but the effect of Jacobsexternality is very notable in the industry of low industrial correlation. In the industry of lowmonopoly, Jacobs externality shows unapparent, however in the industry of high monopoly,the effect of Jacobs externality is very notable.Thirdly, the econometric analysis demonstrates that Porter externality is highlyassociated with the regional economy development and the industrial relationship inGuangDong. Porter spillover has the dominant effect in Pearl River Delta area and showsunapparent in non-Delta area. Meanwhile it is less evidentin the industry of high industrialcorrelation than that of low industrial correlation.Fourthly, the econometric analysis implies that the three effect of externality issignificantly different from the regions in GuangDong. MAR spillover and Porter spilloverhave the dominant effect in Delta area, while Jacobs spillover is more evident in non-Deltaarea.Fifthly, the econometric analysis manifests that MAR spillover and Porter spillover issignificant in the development of high-tech industry in GuangDong. However, Jacobsspillover is not significant.Lastly, the econometric analysis shows that the spatial correlation between regions playsa significant role in promoting the regional industrial development. The impact of introducingspatial correlation between regions into the empirical study is obvious in changing the resultof the three effects. MAR externality is increased and the Jacobs externality is significantlyreduced, while Porter externality remains at the same level.
|
PDF Full Text Request