Calculating China's Industrial TFP At The Prefectural Level ——Based On An Improved Solow Residual Method With Spatial Temporal Econometrical Models Embedded

Calculating China's industrial total factor productivity(TFP)at the prefecture level comprehensively and accurately,is not only the inevitable requirement for China's industrialization to enter the new development stage of "improving quality and efficiency",But also the practical needs for TFP improvement at the industrial level.At present,the academic research on the measurement of China's industrial TFP at the prefecture level is not accurate enough because there is lack of in-depth investigation on the spillover effects in the estimation of empirical production function.Based on the economic development thoughts of structuralism and classicism,combined with the literature reviews on recent researches on TFP,industrial TFP,and spatial-temporal econometric models,this paper promotes an improved Solow Residual Method with spatial-temporal global and local models embedded,and innovates the global and local methods of calculating industrial TFP.And then,this paper calculates China's industrial TFP of 280 prefecture cities during2003-2019 taking the total industrial output value and industrial added value individually as the output index,and industrial employment,industrial capital stock and industrial electricity consumption as the input indexes.Also,this paper compares the calculating accuracy of China's industrial TFP at the prefecture level based on above two different methods through the analysis of statistical robustness.Finally,based on analysis of basic statistics and kernel density estimation,this paper systematically explains the development properties and the spatial-temporal evolution laws of China's industrial TFP in prefecture cities.This paper consists of seven chapters.Chapter 1 is the introduction,it mainly explains the background,research value,technical route,framework arrangement and research methods of this paper.Chapter 2 is the theoretical basis and literature review.This chapter introduces some theoretical basis including economic development theory and industrialization theory,and also reviews the academic history and recent research progress of TFP,industrial TFP,and spatial-temporal econometric models.Chapter 3 is the innovation of TFP calculating methods.it points out the possible defects of traditional Solow residual method,and innovates global and local methods of calculating TFP in a perspective of embedding spatial-temporal global and local econometric models.Chapter 4 is the application of the innovation of TFP calculating methods into the calculation of China's industrial TFP in prefecture cities.The data processing details are explained in this chapter,including definition of prefecture cities involved,selection of variables in the estimation of empirical production function,and determination of instead indices of the variables.the interpolation details of missing value are also explained in this chapter.Chapter 5 calculates China's industrial TFP of 280 prefecture cities during 2003-2019,and its fixed base growth rate,and year by year growth rate separately using the best model chosen from the global or local calculating method.It also selects calculating results from the local method as a better one from the robustness diagnose through statistical analysis.Chapter 7 summarizes the research conclusions in the full text.Possible shortcomings and further research directions of this paper are also discussed in this chapter.Main conclusions of this paper include four points.First,this paper summarizes different stages,modes and driving forces of the development of industrialization.The main development stages of industrialization include pre industrialized society,early,middle and late industrialization,and post industrialization.The main modes include the British model driven by enterprise market,the German model driven by government market,the American model driven by rule of law and company,the former Soviet model driven by heavy industry,the Latin American model dominated by import substitution,and the East Asian model dominated by export.The main driving forces include capital,labor and technology.Innovation,institutional arrangement,infrastructure construction,urbanization and trade development will also have an impact on industrial development.Second,the improved Solow residual method with spatial-temporal models embedded can accurately calculate the industrial TFP at the prefecture level.Industrial TFP is the extension of TFP in the industrial dimension,thus Its calculating methods are basically consistent with those of TFP,which mainly include data envelopment analysis,stochastic frontier analysis and Solow residual method based on empirical production function estimation.Relatively,Solow residual method based on empirical production function estimation is more popular than other methods to calculate TFP.However,there are still some defects in Solow residual method,such as missing the spillover effects in empirical production function settings,and few studies concentrating on the TFP calculating at the industrial level in prefecture cities.Thus,embedding spatial-temporal global and local models into the empirical production function to calculate industrial TFP at the prefecture level,can solve the two defects and promote the development of related researches.Third,the results calculated by the local method are more accurate and robust.Generally speaking,industrial TFP at the prefecture level can be calculated both by the global models with general nesting spatial model and its degradation models embedded,and the local models with PGTWR models embedded.In the global calculation,the optimal models for empirical production function are separately spatial Dubin model with individual fixed effects and spatial autoregressive model with mixed effects.In the local calculation,the optimal spatial and temporal bandwidth are 46 and 17 respectively.From the calculation results of China's industrial TFP at prefecture level based on the optimal models,the results of the local analysis are more scientific because the results from the perspective of total industrial output value are less different from those of industrial added value.Fourth,China's industrial TFP at prefecture level shows an overall development trend of “first increasing,then decreasing,and then slowly recovering”,in which the time nodes of decreasing and slowly recovering are in 2009 and 2015 respectively.Meanwhile,the development characteristics of different prefecture cities are different when different outputs indices and different regional economic sectors are considered.In detail,the calculating results from the perspective of total industrial output value is higher than those of industrial added value.As for the development trend of prefecture industrial TFP in different regional sectors,it is poor and also has an obvious decline trend in in Northeast China,and best in eastern China,while the development trend of central and Western China is between the East and Northeast China.Moreover,the average level of industrial TFP at prefecture level in Southern China is higher and better than that in Northern China.