【作者】 黄志岭；

【导师】 姚先国；

【作者基本信息】 浙江大学 ， 政治经济学， 2009， 博士

【摘要】 改革开放以来,在中国城镇劳动力市场上形成多种群体差异,本文着重关注性别和户籍差异,其中性别差异不单指城镇职工内部的性别差异,还包括城市外来人员内部的性别差异。在人力资本稟赋给定的条件下,造成群体间差异的主要原因便是劳动力市场上不同群体受到差别对待引起的,在某种程度上可归结为劳动力市场歧视。劳动力市场歧视不仅严重损害了受歧视群体的权益,而且也弱化市场机制配置人力资源的作用,影响一国经济的健康运行和社会的和谐稳定。目前国内学者对于上述两类群体差异的研究还不够全面,大多只关注群体间在工资这一个方面上的差异,忽视了群体间在就业、人力资本回报率、职业升迁和劳动参与等方面上的差异。而且研究的方法也比较陈旧,研究中大多采用传统Oaxaca-Blinder分解方法。与以往的研究相比,本文的贡献主要体现在研究的内容更加全面和方法的新颖性。除了研究群体间在工资方面的差异外,还对就业、人力资本回报率等方面的差异进行了研究,主要包括五个问题:①城镇职工性别间工资差异,②城镇职工性别间就业差异,③城镇职工性别间人力资本回报率差异,④城市外来人员性别间工资差异和⑤城乡户籍工资差异及其人力资本作用等问题。在方法上,本文采用Brown分解方法研究了性别间工资差异问题,获得了比以往更加可靠的结论和更为丰富的信息;通过将传统Oaxaca-Blinder分解方法扩展至Logit模型,从而分析城镇职工性别间在就业水平上的差异。本文采用2002、2003年城调队数据,2002年中国城镇住户调查数据,以及通过实地调查得到的2007年浙江省企业职工调查数据对上述五个问题展开了经验分析,主要的发现有:(1)不论是城镇职工还是城市外来人员性别间,均存在着较大程度的工资差异,女性在劳动力市场上相对处于不利的地位。采用Brown分解方法显示性别间工资差异主要来自于歧视等不可解释因素,其中职业内不可解释部分又占了绝大部分。通过对比两类性别差异,为回答市场化如何影响性别歧视提供了新的视角。(2)男性职工的就业率要高出女性6.53个百分点,通过回归分析发现:第一,总体上看,男性、年龄较小、高学历、身体健康、党员以及配偶处于就业状态的个体更容易就业;第二,从分性别的回归结果看,与家庭特征相关的变量对丈夫和妻子有着显著不同的影响;第三,采用扩展的Oaxaca-Blinder分解方法显示,性别间的就业率差异完全来自于劳动力市场结构因素(Labor Market Structual Issues),个体特征差异影响较小。(3)劳动力市场上存在对妇女歧视的同时,还存在女性的教育回报率高于男性的现象。对这一看似矛盾的现象,本文提出的假说是,教育程度的提高对女性来讲,除了与男性一样,具有增加生产力提高工资的作用外,还由于降低妇女在劳动力市场中受到的歧视程度增加工资。正是由于教育对妇女这一额外作用,使女性的教育回报率高于男性。本文采用2003年城调队数据验证了这一假说。(4)本文利用2007年浙江省企业职工调查数据所进行的关于人力资本对城乡户籍歧视影响的研究结果显示,工资差异的20%是由歧视造成的;城乡户籍歧视主要发生在高中组,其歧视程度要大大高于初中组。这一结果使那种认为教育程度的提高能够降低受歧视程度的假说在中国城乡户籍差异问题上得不到很好的验证。更多还原

【Abstract】 Since reform and opening up, there appeared differences between many groups in Chinese urban labor market, this article focuses on differences in gender and huji, and the gender difference not only about urban workers but also includes floating population. Given the conditions of human capital, the main reason for differences between groups is that different groups was subjected to differential treatment. To some extent this effects can be attributed to labor market discrimination. Labor market discrimination not only seriously damaged the interests of groups which was discriminated against, but also weaken the market mechanism in the allocation of human resources and influenced the health of a country’s economy as well as social harmony and stability.At present, the researches with regard to difference between these two groups are not comprehensive enough, most of which merely concerned about the wage differentials by gender and huji, overlooking groups difference in terms of employment, rate of return to human capital, career promotion and labor force participation and so on. In addition, the research methods are somewhat outdated, most of the researches used the traditional Oaxaca-Blinder decomposition method.Compared with previous studies, the main contribution of this paper is embodied in the contents of the study more comprehensive and the novelty of the method. In addition to the study in the wage differentials between groups, the differentials in employment, the return to human capital also have been studied. In a word, this thesis studys five issues of the differences between gender and huji in-depth, which are the gender differences in wages, employment, human capital returns of urban workers, the gender differences in wage in floating population and the wage gap between urban and rural workers. On method, this paper uses Brown decomposition which takes into account occupational differences to study the gender wage differentials, obtaining more reliable conclusions and much richer information than before, and uses an extension of the Oaxaca - Blinder decomposition technique to the logit models to study the gender differentials in employment between urban workers.Using 2002, 2003 Urban Household surveys (UHS) data, 2002 Chinese Household Income Project(CHIP) data and the latest enterprise employee data of Zhejiang, this paper implements the empirical analysis on the above five questions. The main findings of this paper are followings. (1) Regardless of urban workers or floating population, there exits large wage gap between gender, women at a relative disadvantage in the labor market. Using Brown decomposition, we find most part of the wage differential between gender is attributed to discrimination. (2) The employment rate of male workers is higher than female by 6.53 percentage points in urban workers. Through regression analysis we find these results. First, as a whole, being male, younger, better educated, healthy, communist party member, and spouse’s employment increases the probability of being employed. Second, family-related background characteristics have different impact on female and male. Third, using an extension of the Oaxaca - Blinder decomposition technique to the logit models, the results show that labor market structural issues explain almost all of the observed gender gap in employment, and the effect of the gender gap in individual endowments is very small. (3) Most studies show that there exists significant discrimination against female workers in the Chinese urban labor market. However, studies also show that the returns to schooling are higher for women than for men. Regarding such contradictory evidence, this paper assumes that women’s education can increase female wages not only by enhancing their productivity which works in the same way to men, but also by reducing the discrimination against women in the labor market. Such additional effect provided by the education to women makes female education returns higher than men. Using 2003 Chinese Urban Household surveys (UHS) data, we validate this hypothesis. (4) Using the latest enterprise employee data of Zhejiang to examine the impact of human capital on huji discrimination, we find that 80% of the total huji wage differentials can be interpreted by the differences of individual characteristics and 20% the effects of the discrimination, and the discrimination mainly occurred in the higher school worker group which is much higher than in the secondary worker group. This finding indicates the assumption that education can reduce the unexplained gap is not approved well in the question of Chinese huji wage differentials.更多还原