Distribution Characteristics,Heterogeneity And Spatial Effects Of Educational Returns

Education is a critical task for a century,and the Chinese government highly values the development of education.The return on investment in education,which measures the size of the benefits of education investment,is an important research area related to national development.Existing literature on the return on investment in education has mostly focused on cross-sectional or longitudinal studies.This article systematically studies the heterogeneity and distribution characteristics of the return on investment in education based on previous research,and explores its spatial effects as a research direction.The focus of this study is on the distribution characteristics,heterogeneity,and spatial effects of the return on investment in education in 31 provinces and cities in China.The CGSS data from 2018 were used for empirical analysis.This study selected four control variables,namely gender,political affiliation,household registration,and work ownership,and used the OLS method to study the heterogeneity of the return on investment in education.Robustness tests were conducted using methods such as variable substitution,sample size reduction,and the Heckman two-step method.To fully depict the relationship between the dependent variable and independent variables,quantile regression was used.The 31 provinces and cities were divided into seven regions,including East China,South China,North China,and Central China,to study the distribution characteristics.In terms of spatial econometrics,this study first conducted spatial autocorrelation tests for global and local spatial autocorrelation,followed by diagnosing the spatial effects,and finally determining the SAR model as the final regression model.The research findings are as follows: First,using an extended Mincer income equation,the returns to education are found to be 15.5% in China,with returns to education of 14.8%for state-owned workers and 16.9% for non-state-owned workers.The returns to education for party members are 17.1%,while the returns to education for non-party members are15.2%.These results indicate that there is heterogeneity in returns to education in China.Second,in the robustness test,the return to education obtained by the substitution variable method was 15.1%,and the return to education obtained by the narrowed sample method was 15.6%.After addressing the sample selection bias using the Heckman two-step and maximum likelihood estimation methods,the returns to education are estimated to be 18.3%and 16.7%,respectively.Thirdly,dividing 31 provinces into seven regions,estimates were made at the 0.2,0.4,0.6,and 0.8 quantiles,revealing that the northwest,southwest,and north China regions have higher returns to education,and the returns to education decrease gradually as the quantiles increase.Fourth,the Moran’s I index for the returns to education is 0.230,and the spatial autoregressive coefficient of the SAR model is 0.296,indicating positive spatial autocorrelation in income levels.