就业选择效应与公共部门和非公共部门工资差距——基于两部门模型视角的研究

Sector Selection and Wage Difference Between Public and Private Sectors:A Research based on Two-Sector model

本文在两部门模型视角下研究了公共部门和非公共部门工资差距问题,尤其关注就业选择效应。分析表明,劳动者对部门的选择是基于就业价值的比较,这允许模型的均衡结果偏离完全竞争市场假设下的“一价定律”,因为可以存在工资差异。基于2013年和2018年中国家庭收入调查(CHIP)数据,本文使用Copula方法来估计转换回归模型,以控制就业选择效应和放松联合正态性的限制性假设。研究结果显示,首先,公共部门由2013年的正向工资溢价(6.0%)转变为2018年的负向工资溢价(-0.1%),再次出现“工资惩罚”现象;两部门之间的工资差距存在异质性,处理组的平均处理效应在两个年份分别为37.5%和10.6%,非处理组的平均处理效应则分别为-11.5%和4.7%,这说明实际选择公共部门的个体更可能获得比在反事实情况下更高的工资;其异质性还反映在教育、性别和合同类型等特征上。

This paper studies the wage difference between the public and private sectors based on a two-sector model, with particular attention to the sector selection. Workers’ choice of sectors is based on a comparison of the employment value, which allows the equilibrium results of the model to deviate from the "law of one price" under the assumption of perfectly competitive markets, because wage differences can exist. Based on the 2013 and 2018 Chinese Household Income Project(CHIP) data, we use the Copula method to estimate the switching regression model, so as to consider the employment selection and relaxing the restrictive assumptions of joint normality. First, the public sector changed from a positive wage premium(6.0%) in 2013 to a negative wage premium(-0.1%) in 2018, which means the phenomenon of "wage penalty" still exists in the public sector. Second, there is heterogeneity in the wage difference between the two sectors. The average treatment effect of the treated group in two years are 37.5% and 10.6%, while the average treatment effect of the untreated group is-11.5% and 4.7%, indicating that individuals who actually choose public sector can earn higher wages than in the counterfactual case. Third, this heterogeneity is also founded in characteristics such as education, gender and contract types.