Does union membership reduce gender earnings differentials? Evidence from employer–employee matched data in China

• Author: Chunbing Xing,Jing Liu,Yuhao Ge
• Published date: 01 February 2020
• Date: 01 February 2020
• DOI: http://doi.org/10.1111/1468-0106.12299

ORIGINAL MANUSCRIPT

Does union membership reduce gender earnings

differentials? Evidence from employer–employee

matched data in China

Jing Liu

1

| Chunbing Xing

2

| Yuhao Ge

3

1

School of Economics, Central University

of Finance and Economics, Beijing, China

2

School of Economics, Beijing Normal

University, Beijing, China

3

School of Labor and Human Resources,

Renmin University, Beijing, China

Correspondence

Yuhao Ge, School of Labor and Human

Resources, Renmin University, Beijing,

China, 100872.

Email: geyuhao@ruc.edu.cn

Funding information

National Natural Science Foundation of

China, Grant/Award Numbers: 71673317,

71773009, 71773124

Abstract

This paper examines whether union membership reduces

gender earnings differentials in the Chinese labour market

using an employer–employee matched data set. We have

three main findings. First, union membership helps reduce

the gender differentials in hourly wage and monthly allow-

ance, but not in monthly basic wage and yearly bonus.

Second, ensuring that female workers receive overtime

pay is one way by which union membership helps reduce

the gender earnings differentials. Third, controlling for

firm fixed effects reduces the effect of union membership

on the gender gap in hourly wage and monthly allowance,

which means that the unobserved firm characteristics

might impact the effect of union membership on gender

earnings differentials.

1|INTRODUCTION

Women earn significantly less than men in the Chinese labour market, and the gender earnings differ-

entials have enlarged significantly in recent years (see e.g. Li & Gustafsson, 2008; Wang & Cai,

2008; Zhang, Han, Liu, & Zhao, 2008; Dong & Zhang, 2009). Researchers have looked at many fac-

tors that contribute to the gender earnings differentials, including discrimination against female

workers, occupational segregation, differences in education levels and experiences, and increasing

return to skills (see Chi & Li, 2008; Dong, Macphail, Bowles, & Ho, 2004; Li & Dong, 2011; Li &

Gustafsson, 2008; Maurer-Fazio & Hughes, 2002; Meng, 1998; Meng & Miller, 1995; Rozelle,

Dong, Zhang, & Mason, 2002; Wang & Cai, 2008; Zhang et al., 2008; Zhang & Dong, 2008). Some

studies argue that gender earnings differentials are impacted by labour market regulations. For

Received: 23 April 2018 Revised: 26 October 2018 Accepted: 24 February 2019

DOI: 10.1111/1468-0106.12299

102 © 2019 John Wiley & Sons Australia, Ltd Pac Econ Rev. 2020;25:102–117.wileyonlinelibrary.com/journal/paer

example, Li and Ma (2015) showed that minimum wages would reduce gender earnings differentials

in the long run.

In recent years, the Chinese Government has encouraged the development of unions to manage

the labour relations and to reduce social conflicts. Both trade unions and union members have

increased significantly, and unions have been endowed with greater authority over worker protection,

such as the ability to negotiate employment contracts (Chen, 2012). Most existing studies on trade

unions in China are case studies, such as Clarke (2005), Clarke, Lee, and Li (2004), Friedman and

Lee (2010), Metcalf and Li (2006) and Zhang (2009). Lu, Tao, and Wang (2010) and Yao and Zhong

(2013) are two exceptions that offer quantitative analysis. Yao and Zhong (2013) show that unioniza-

tion is significantly associated with higher hourly wage and larger pension coverage and weakly

associated with lower monthly working hours. Lu et al. (2010) find that unions do not increase wages

but are helpful to guarantee other benefits, such as pensions, medical insurance and unemployment

insurance. So far, how unions in China affect female workers’welfare and gender earnings differen-

tials has rarely been studied.

Although there are some studies that have examined the effect of trade unions on gender earnings

differentials in other countries (Aidt & Tzannatos, 2002; Casale & Posel, 2011; Doiron & Riddell,

1994), no consensus has been reached. Some find that unions reduce gender earnings differentials

(Aidt & Tzannatos, 2002; Doiron & Riddell, 1994; Meng & Meurs, 2004) and some do not

(Duguet & Petit, 2007). There are also some studies that find larger gender earnings differentials in

the union sectors (Casale & Posel, 2011). Therefore, further evidence is needed to understand the

effect of unions on the gender earnings differentials.

In this paper, we use employer–employee matched data to examine the effect of union member-

ship on gender earnings differentials in China. Most previous studies rely on cross-sectional individ-

ual data without controlling for the unobserved employer characteristics. The omitted unobserved

employer characteristics might result in biased estimation of the union effect on gender earnings dif-

ferentials. Union firms may tend to pay female workers in better than nonunion firms due to some

unobserved characteristics. Employer–employee matched data allow us to control for the unobserved

firm-level heterogeneity.

We have three main findings in this paper. First, union membership is helpful to reduce the gen-

der differentials in hourly wage and monthly allowance but not in the monthly basic wage and yearly

bonus. Second, ensuring female workers receive overtime pay is one way by which trade unions help

reduce the gender earnings differentials. Third, the unobserved firm characteristics do impact the

effect of union membership on gender earnings differentials because such an effect decreases after

controlling for firm-level fixed effects.

This paper is organized as follows. Section 2 introduces the institutional background and develop-

ment of unions in China, emphasizing how unions may have different effects on male and female

workers. Section 3 describes the employer–employee matched data used in this paper. Section 4 dis-

cusses the empirical strategy used to identify the effect of unions on gender earning differentials.

Section 5 reports the empirical results and discusses the implications of these results. Section 6 con-

cludes the paper.

2|UNIONS IN CHINA AND THEIR ROLE IN HELPING

WOMEN

As a planned socialist economy, most enterprises were managed by the Chinese government before

the economic reforms in the 1980s, and it was the planning authority that determined workers’

LIU ET AL.103