中国教育不匹配的收入效应基于CHNS数据的实证-中国统计教育学会.PDFVIP

中国教育不匹配的收入效应 ：基于CHNS 数据的实证 山东工商学院 徐聪、刘菲、王宁 摘要：本文利用1997 年和2006 年CHNS 调查数据，依据ORU 模型考察中国教 育收益率的变化，将教育收益率分为教育匹配、教育过度和教育不足收益率。首先通 过增加控制变量，发现控制个体异质性后教育匹配与教育不匹配收益率的差距减小。 其次，分全样本、性别和城乡三个数据集，比较其教育匹配和教育不匹配收益率的 OLS 与分位数回归结果。实证结果表明中国 1997 年和2006 年教育匹配收益率为正， 教育过度收益率为正，但小于教育匹配收益率，教育不足收益率为负，且三者均随时 间增大。均值回归忽略了工资结构中一些重要的特点，分位数回归结果揭示了教育收 益率在整个收入条件分布中不均衡，在收入条件分布的低分位点上高，在高分位点上 低。工作经验的收益率随着时间而减少，但可以弥补教育不足带来的收入惩罚。1997 年和2006 年中国收入分配不平等且有增大的趋势，教育收益率在其中扮演重要的角 色。 关键词： 教育收益率；教育不匹配；收入效应；分位数回归 I The effect of education mismatch on income in China: empirical analysis from CHNS data Abstract: In order to study the return of schooling, which is divided in to the return of education match, overeducated and undereducated, empirical analysis is conducted in the paper by CHNS (Chinese Health and Nutrition Survey) data on 1997 and 2006. In order to study the evolution of returns to education in China with ORU model, which is divided in to the return of education match, overeducated and undereducated, empirical analysis is conducted in the paper by CHNS (Chinese Health and Nutrition Survey) data on 1997 and 2006. First, by adding the control variables, it turns out that the gap between the return to education match and mismatch decreases, after controlling individual heterogeneity. Secondly, this paper compares the return of education based on the result of OLS and quantile regression through full sample, gender sample and area sample. The results of estimation shows that on 1997 and 2006, the return of overeducated is positive, but lower than the return of required education and the return of undereducated is negative. Average effects (OLS) ignore important features of the income structure and