Estimation And Application Of China's Regional Aggregate Production Function Model

The Cobb-Douglas production function model of mainstream economics is based on the "Kaldor facts" of structural stability and steady economic growth in developed countries.However,China is showing the characteristics of economic unsteady growth,and the problem of regional differences has become increasingly prominent,which is inconsistent with the "Kaldor fact." Based on China's special economic national conditions,it is necessary to construct a model of aggregate production function suitable for national conditions from a regional perspective.This article focuses on the factor output elasticity(factor income share),and studies its differences in time and region from both theoretical and empirical aspects.In theory,according to the literature,it is concluded that the differences in regional technical factors,different levels of market competition,and different levels of economic development are the main reasons for the differences in factor income shares among regions.In addition,researches on the estimation methods of production function models,panel data models and semi-parametric variable coefficient models show that it is feasible to construct and estimate China's regional aggregate production function models.In terms of empirical research,we use China's provincial panel data from 1993 to 2017 to estimate China's regional aggregate production function model.The first is the variable coefficient panel production function model.It is concluded that in general,my country's regional capital output elasticity shows an upward trend,labor output elasticity shows a downward trend,and the capital output elasticity of the eastern region is much higher than that of the central and western regions.This may be related to the regional economic development in our country.The second is the variable coefficient spatial production function model.It is obtained that the capital output elasticity of most provinces in eastern and central my country shows an inverted U-shaped trend,while the labor output elasticity shows a positive U-shaped trend.This may be related to the regional industrial structure of my country.The transition stage is related,and the labor output elasticity presents a characteristic that the western region is significantly higher than the eastern and central regions,which further illustrates the correlation between the regional factor output elasticity and the level of regional economic development.This article may have the following academic contributions: First,based on the fact that my country's economic growth is unstable and the characteristics of regional factor endowments,a variable coefficient panel production with output elasticity variable in time and region has been established.Function model;Secondly,based on the spatial correlation characteristics of my country's regional factor endowment structure,a variable coefficient spatial production function model is established in which output elasticity is not only variable in time and region,but also has spatial correlation.This is a beneficial exploration for constructing an aggregate production function model suitable for China's national conditions.Second,the Profile method and Back-fitting method commonly used in semi-parametric variable coefficient models are used to estimate the output elasticity of regional factors,which broadens the research thinking of production function model estimation.Third,this article applies the results of regional output elasticity estimation to the analysis of regional total factor productivity,regional total factor productivity growth rate and the overall dynamic evolution trend of factor output elasticity,which expands the application scope of the production function model.