社会杂志 ›› 2016, Vol. 36 ›› Issue (3): 64-85.

• 专题：教育社会学研究 • 上一篇    下一篇

中国高学历劳动者的教育匹配与收入回报

李骏   

  1. 上海社会科学院社会学研究所
  • 出版日期:2016-05-20 发布日期:2016-05-20
  • 通讯作者: 李骏 上海社会科学院社会学研究所 E-mail:lijun@sass.org.cn E-mail:lijun@sass.org.cn
  • 基金资助:

    本研究得到了上海市教育委员会和上海市教育发展基金会"曙光计划"项目的支持(项目编号13SG58)。本文数据来自中国人民大学中国调查与数据中心(NSRC)资助并组织实施的"首都大学生成长追踪调查"项目。该项目2009年和2010年的调查曾得到香港科技大学研究项目竞争基金(RPC07/08.HS02)、香港研究资助局优配基金(644510,项目负责人:吴晓刚)以及中国人民大学科学研究基金(2009030080、20100030415,项目负责人:冯仕政)的资助,研究设计也曾得到香港研究资助局优配基金(644510)的部分支持。

Education-Job Mismatches and Earnings among Chinese College Graduates

LI Jun   

  1. Institute of Sociology, Shanghai Academy of Social Sciences
  • Online:2016-05-20 Published:2016-05-20
  • Supported by:

    The research was sponsored by Shanghai Shuguang Program. Data analyzed in this paper were from Beijing College Students Panel Survey(BCSPS) conducted by National Survey Research Center (NSRC) at Renmin University of China. The data collection of BCSPS in 2009 and 2010 was also supported by HKUST Research Project Competition (RPC07/08.HS02),General Research Fund of Hong Kong Research Grants Council (644510),and Scientific Research Fund of Renmin University of China(2009030080,20100030415).

PDF

1062

摘要：

近十多年来,鉴于中国高等教育扩张对劳动力市场的影响,一些学者开始用经验数据研究劳动者与工作之间的教育匹配(尤其是过度教育)对收入回报的影响。但是,已有文献较少考虑过度教育的选择性问题,也很少对高学历劳动者进行专门研究。本文使用倾向值匹配的方法分析高学历劳动者的教育匹配与收入回报。研究发现,即使是在考虑了选择性之后,过度教育者的收入仍然显著低于适度教育者。同时,对家庭背景、学校出身、人力资本和劳动力市场四大影响因素的考察还发现,过度教育确实具有负向选择的特点,那些拥有较低人力资本和较差出身背景的人更容易发生过度教育。

关键词: 教育匹配, 过度教育, 选择性, 倾向值匹配

Abstract:

The Chinese higher education expansion since 1999 has stimulated a number of empirical studies on the impact of education-job mismatches on college graduates' income,especially among the overqualified. This paper aims to overcome the two limitations in the existing literature that often neglect the possibility of self-chosen over-education as well as the specific labor market for university graduates. Propensity Score Matching (PSM) is applied to analyze the data from the Beijing College Students Panel Survey,namely,the methods of greedy matching,optimal pair matching and optimal full matching,as well as the three predicting equations of propensity score. The analysis indicates a wage penalty associated with over-education that is consistent with the findings in other societies. The self-chosen over-education does not alter the pattern. The overeducated workers in this study earn 21% less than those whose education matches their job requirement. After controlling demographic characteristics and labor market factors in multivariate OLS regression analysis,the difference is down to 15%. The PSM estimates vary from 9% to 18%,depending on the different application of PSM methods and the specification of predicting equations of propensity score. Evidences in this study support the assignment theory,rather than the neo-classical human capital theory,implying that underutilization of education imposes a "productivity ceiling" or a limitation to the utilization of human capital. Furthermore,the study also finds a higher risk of over-education mismatches among graduates who come from less advantaged family background with less prestigious degrees and fewer human capital.

Key words: educational mismatch, over-education, self-selection, propensity score matching