Regional Characteristics Of China's Economic Structure Transformation And Its Impact On TFP

China's economy has experienced rapidly growth during the past three decades.According to the experience of developed countries,economic growth is usually accompanied by economic structure transformation.Due to the regional differences in China's economic development,the long-standing structural contradictions are gradually displayed.Indeed,economic endowment varies with economic development opportunities of different regions or provinces.It causes the difference between the provincial economic structure transformation paths.Undoubtedly,it is significance of policy value to study the regional differences of the economic transformation path,to excavate the common points of the economic structure transformation and to formulate industrial policies and regional economic development strategies.Based on the "Kuznets facts",the main work done in this paper as follows:Firstly,we take the ratio of the three industries output and the proportion of labor force in 31 provinces as indicators,extract the absolute level,fluctuation level and trend level of panel data to reflect the similar characteristics of economic structure transformation,and use the clustering evaluation index to determine the optimal clustering.Through cluster analysis of panel data,the 31 provinces are divided into 5 categories according to the ratio of the three industries output;and they are divided into 3 categories based on the proportion of labor force.Secondly,in order to study the spatial correlation of economic structure transformation,the geographical factors which are represented by the latitude and longitude of provincial capitals are introduced to measure the spatial characteristics of the provinces.The spatial clustering based on the three industries output and the proportion of labor force divided 31 provinces into 5 categories and 6 categories.Thirdly,on the basis of existing literature,we introduce research level RD,marketization index MI,the opening degree FDI,human capital level PL,infrastructure level STRL,urbanization level UL as the control variable model,the ratio of the tertiary industry output value TR is taken as the independent variable and TFP is the dependent variable in each province.On the basis of spatial clustering(5 kinds),the influence factors on TFP were studied by classification,and the influence factors of TFP in east,middle and western regions were further analyzed.The results of this study show that China's economic structure has the following regional characteristics:First of all,Shanghai and Beijing performed best.The two countries are in the leading position in economic development,which are important gateways for China and international integration.The second,the results of panel clustering and spatial clustering are different as well as similar.Both clustering results show that the higher the economic structure level,the higher the level of employment structure.The regions with the same economic structure level show the state of aggregation in the geographical space,which is related to industrial development and agglomeration,and is also related to the heterogeneity of regional economic endowments.Thirdly,the economic structure is unbalanced and the economic structure of different regions is obviously different in China.The uncoordinated problem of east to west economic structure gradually increase:The development of the service industry in the central and western regions lags behind,and the industrial development in the central part of the country remains the first place,and the western region is too heavily dependent on agriculture for its economic development.The results also show that the regional characteristics of economic structural transformation have the following conclusions about the influence of TFP:First,from the perspective of categories,"Kuznets facts”have been verified in Beijing,Shanghai,Shandong,Jiangsu and other parts of the region,However,in Hebei,Liaoning,Fujian,Hainan and other parts of the region,there is the so-called"reverse Kuznets" phenomenon.For another,from the view of the sub-region,the transformation of the economic structure in the East and central regions conform to the "Kuznets fact",while the western region shows the phenomenon of "reverse Kuznets".Tertiary,the results of different classes and different regions indicate that the promotion of PL has no significant effect on the improvement of TFP.Increasing RD would inhibit the improvement of TFP in the central and western provinces.Finally,among other factors affecting TFP in the region,FDI has significant positive influence to TFP in the region and MI almost inhibits TFP growth in all regions,the rise of STRL and UL isn't significant to promote regional TFP,even to inhibition in some areas,such as the central region.As you can see,the main innovations of this paper are as follows:In the first place,the clustering evaluation index is constructed by using the class distance and inter class distance to determine the optimal clustering according to the means of K-means clustering method in this paper.By mining the feature values of the panel data,we chose the mean,standard error and time trend as the clustering indexes.What's more,we use the latitude and longitude coordinates of provincial capitals to measure spatial attributes.By means of spatial clustering and visualization,we complete the spatial clustering of 31 provinces and cities in China,which describes the regional characteristics of the economic structure transformation in all provinces of China.In this paper,we have solved a problem that the existing research using principal component analysis to analyze the economic structure problem,which can't depict the regional characteristic of economic structure transformation.In the second place,we take TR as the independent variable and TFP is the dependent variable in each province;RD,MI,FDI,PL,STRL and UL are introduced as the control variables to establish fixed effect panel data model.Based on the spatial clustering results,we have discussed the influence of regional economic structure transformation on TFP from the perspective of different classes and different regions.Therefore,we have solved another problem that the existing research using shift-share method to analyze the economic structure problem,which can't describe the difference in effect of regional economic structure on TFP.