作　　者： ;

机构地区： 中山大学社会学与人类学学院

出　　处： 《社会学研究》 2012年第1期152-173,245,共22页

摘　　要： 以往关于户籍制度对收入的影响的诸多研究基本上遵循的是工资差异分解模型的思路。然而,将外来人口和本地市民的工资收入差异进行分解,无法克服样本选择性与遗失关键变量带来的偏误。本文基于中国家庭收入调查数据,采用新近发展的倾向得分匹配方法来控制样本选择偏误,并采用自抽样法(Bootstrap)法进行统计推断,以克服小样本偏误,试图回答户籍限制对流动人口收入的影响。研究结果发现:(1)户籍制度对流动人口收入影响非常显著,克服选择性偏误之后的负向效应要高于普通回归结果和代理变量回归结果;(2)户籍限制对倾向得分较低的流动人口的收入效应是正向的,对倾向得分较高(能力较高)的流动人口产生影响为负;(3)这一不利影响随着市场化进程推进变得更显著。 Traditional studies explaining the effect of Hukou types on wage disparity basically adopt Blinder-Oaxaca decomposition framework. However, the Blinder-Oaxaca decomposition method in studying the mean income differences between urban residents and migrant workers couldn't eliminate two types of selection bias in observational data: the first type is due to the heterogeneity of sample, and the second bias is due to the omission of key variables. Adopting the newly developed propensity score matching method to control for sample selection bias, and the Bootstrap method to overcome the small sample bias, this paper tries to explore the income effects of Hukou restriction on floating population based on China's household income survey data. The results show that: (1) the effect of Hukou on the income of migrants is very significant, and the negative effect with control for sample selection bias, is higher than those of ordinary regression results with proxy variables; (2) the effect of Hukou on low propensity scores migrants is positive,however,the effect on high propensity scores migrants is negative; (3) the higher the degree of marketization, the stronger the negative effects.

关 键 词： 户籍 流动人口 选择性偏误 倾向得分匹配法 模型

领　　域： [经济管理] [经济管理]