Double negative effects on the wages of ethnic minority women: evidence from China

ABSTRACT

Using data from the Chinese Household Income Project surveys of 2002, 2013, and 2018 and employing decomposition methods, this study is the first to examine the impact of the double negative effect on the wages of ethnic minority women in urban China. Three key conclusions emerge. First, the results indicate the presence of a double negative effect, with the gender wage gap effect being more pronounced than the ethnicity wage gap effect. This suggests that discrimination against women is the primary factor contributing to the formation of the wage gaps. Second, the decomposition results on wage distribution suggest that the return effect on both ethnicity and gender wage gaps is greater for the high-wage group than for the low- and medium-wage groups, suggesting workplace discrimination may be much more severe for the high-wage group. Third, the wage gaps and determinants of their formation differ by heterogenous ethnic minority groups.

KEYWORDS:

• Double negative effects
• Han-minority wage gap
• gender wage gap
• ethnic minority women

Authors’ contributions

The author (XM) contributed to idea generation, conducted empirical analysis, and prepared the manuscript. The author read and approved the final manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability of statement

The dataset used in this study, Chinese Household Income Project surveys of 2002, 2013, and 2018, is publicly available (http://www/ciidbun.org.chip/index.asp?lang=CN). The dataset and materials constructed by the author are available upon request.

Notes

1 For the latest empirical studies on the ethnic minority-majority wage gap, see Bağçe et al. (2021), Cantalini et al. (2022), Christl et al. (2020), and Dostie et al. (2021).

2 Weichselbaumer and Winter-Ebmer (2005) conducts a meta-analysis based on 262 studies on the gender wage gap worldwide. For the current empirical studies on the gender wage gap in other countries, see Blau and Kahn (2017) for the U.S., Biewen et al. (2020) for Germany, and Masso et al. (2022) for Estonia.

3 Iwasaki and Ma (2020) conducted a meta-analysis based on 199 studies in China. They found that a gender wage gap existed from the 1990s to 2010s in China.

4 Some empirical studies have explored the differences in education (Gustafsson & Sai, 2015; Hannum, 2002), occupation (Hannum & Xie, 1998; Hannum & Ho, 2015; Wu & He, 2018; Li, 2021), and labor force participation (Maurer-Fazio et al., 2007; Cherng et al., 2019) between the ethnic minorities and Han in China.

5 Based on the CHIP data spanning 2002–2018, the proportion of non-working individuals to the total individuals aged 16–59 years is 38.22% for women, 19.45% for men, 28.68% for the Han, and 31.48% for total ethnic minorities.

6 This study used age and unearned income as the identification variables for the exclusion restriction. The reasons are as follows. First, based on the new classical economic theory on individual labor supply, participation in work is mainly determined by an individual's income constraint and work (leisure) preference. The unearned income may influence the income constraint; therefore, it may affect the likelihood of participation in work, especially for women (Heckman, 1979). Several studies have used unearned income in the labor supply function (e.g., Heckman, 1979; Heckman & MaCurdy, 1980; Schulz, 1980; Ma, 2022b). Second, in China, the mandatory retirement age in the public sector (government organization, SOEs) differs by gender. It is 50 years for female workers, 55 years for female cadres, and 60 years for male workers and cadres. The early mandatory retirement system is also enforced for special groups such as workers in high-risk occupations in the public sector. Therefore, we can use age as an identification variable in the labor supply function. Some studies (e.g., Ma, 2022b) have used the age variable in the first-step estimation of the Heckman two-step method.

7 The decomposition results based on Equations (6) and (7) are similar. Due to limited space, I only report the results based on Equation (6) in this study. The other results are available upon request.

8 The results using the samples aged 16 to 55 years were approximately similar to the results using the samples aged 16 to 59 years. These results are available upon request.

9 Abnormal values are those not in the range of ‘mean value ± three times SD.’

10 As CHIP 2002 has no detailed information on ethnicity groups, this study only used CHIPs 2013 and 2018 in these analyses. Based on the CHIP survey data used in this study, the proportion of total ethnic minorities is 4.4%, the largest ethnic minority is the Man ethnicity (0.97%), and the second largest is the Hui ethnicity (0.82%). This study used 239 Man samples and 203 Hui samples.

Additional information

Funding

This work was supported by the Japan Society for the Promotion of Science [grant number 20H01512 and 20H01489] and the project research fund from the Joint Usage and Research Center, Institute of Economic Research, Hitotsubashi University [grant number IERPK2323].

Notes on contributors

Xinxin Ma

Dr. Xinxin MA is a professor at the Faculty of Economics, Hosei University. Her current research focuses on labor market institutional transition and income inequality in China. Her recent articles have been published in journals such as Journal of Asian Economics, China Economic Review, Review of Development Economics, Emerging Markets Review, and Economic Systems.