The Polarization of Chinese Household Wealth Distribution:Measurement and Analysis Based on Relative Distribution Method
Based on CFPS data from 2010 to 2020,this paper analyzes the changing trend of household wealth polarization by the recently-extended relative distribution method;meanwhile,it analyzes the internal causes of the change of wealth distribution by the quantile regression and Oaxaca-Blinder decomposition method.The results show that the polarization of household wealth in China has an expanding trend,where the polarization of wealth is mainly from the middle level to the lower end,being a serious polarization of the lower level.The phased analysis found that the polarization of wealth was mainly from 2010 to 2018,where the polarization of wealth significantly weakened after 2018.The analysis of micro causes of wealth polarization shows that the improvement of endowment structure of low wealth groups can effectively suppress the lower polarization,while the alleviation of upper polarization mainly comes from the effect of coefficient effect.Since 2018,rising household incomes,a declining rural population,and a general increase in the level of education of the population have weakened the downward polarization of wealth,while the rural-urban wealth gap and returns to education have suppressed the upward polarization of wealth.The decomposition within urban and rural areas reveals that the educational returns is not significant in rural households,while health is an important factor of rural wealth polarization.The policy implication of this paper is that wealth adjustment policies should pay attention to the vulnerability of middle-class families,increase wealth accumulation channels for low-and middle-income families,and promote the growth and stability of the middle class.To restrain the downward polarization of wealth,we should emphasize the optimization of endowment structure of low-wealth groups,while reducing upper polarization should focus on regulating wealth disparities and returns.
国家科技期刊平台
NSTL
万方数据
