作　　者： (刘欢）;

机构地区： 武汉大学社会保障研究中心,湖北武汉430072

出　　处： 《经济经纬》 2017年第5期43-48,共6页

摘　　要： 基于中国居民收入调查数据库(CHIP)2013年的数据,利用分位数回归(QR)模型,研究贫困与非贫困对比条件下农村家庭人均纯收入决定模型。结果显示:健康、教育、社会保险、非农就业特征与家庭贫困呈显著相关性;对差距缩小的系数差检验结果显示,养老与其他医疗保险、本地和异地外出务工时间比变量的系数差呈显著负向。在稳健性回归检验中,引入低收入与非低收入分位数的回归结果与贫困、非贫困家庭各变量的显著性结果无差异,但各变量的系数差检验的显著性水平差异有较大变化,本地非农就业时间比对低收入家庭收入变化的作用更显著。 Based on the survey data from Chinese residents＇ income survey database（ CHIP） in 2013,this paper uses the Quantile Regression（ QR） model to study the model of net income per capita of rural households under the contrast of poverty and non-poverty.The results show a significant correlation of health,education,social security,non-agricultural employment to household poverty. The coefficient difference test showed that the difference between the pension and other medical insurance,time ratio of local and urban working was significantly negative. In the robust regression test,the regression results of the low-income and non-low-income quantiles show that there is no significant difference between the variables of poor and non-poor families,but the difference in significance level of coefficients shows great changes in coefficient difference test,which shows the effect of local non-agricultural employment time ratio is more significant for low-income family income.

关 键 词： 农村贫困 人力资本 社会保险 外出务工