Estimating Nonlinear Intergenerational Mobility in China with an Analysis of Influencing Factors

This paper integrates income inequality and intergenerational mobility (IGM) under a unified framework, providing a theoretical basis for the nonlinear measurement of IGM. Also, it proposes using a nonparametric quantile regression gradient boosting tree model to estimate the China's nonlinear IGM elasticity and to study the interactive effects of macro-micro determinants. The empirical results show that using the nonlinear IGM elasticity, IGM should be significantly enhanced, with the maximum increase reaching 12.92%. Furthermore, the fitting of the nonlinear quantile regression gradient boosting tree model is superior to that of the linear quantile regression model. The nonlinear characteristics are particularly significant for the population with an annual income between 30,000 yuan and 150,000 yuan. China's intergenerational income elasticity lies between 0.1861 and 0.7026, and parental income evidently affects offspring income with a "strong two-tailed and weak middle" pattern. Moreover, there are income and regional heterogeneities in the characteristics of IGM, and there are significant differences in the income transmission process and the degree of nonlinearity among different regions. Finally, this paper explores IGM traits of income spectrum extreme groups and the "poverty trap" in China. The results show that the most important influencing factors for achieving income class crossing are economic growth, industrial optimization, intergenerational educational mobility, and wealth capital investment.