Gender Gap in Earnings in China: A Cross Sectional Study

© 2021 Published by VŠB-TU Ostrava. All rights reserved. ER-CEREI, Volume 24: 69–79 (2021). ISSN 1212-3951 (Print), 1805-9481 (Online) Gender Gap in Earnings in China: A Cross Sectional Study Yajun MENG a,b * a Department of National Economy, Faculty of Economics, VSB - Technical University of Ostrava, Sokolská třída 33, 702 00 Ostrava, Czech Republic. b Faculty of Economics and Trade, Hebei GEO University, Shijiazhuang, China. Abstract The privatization of the Chinese economy and the mobility of the labour force from rural areas to urban areas are increasing the pressure of gender wage inequality in China. This paper analyses the gender wage gap based on microdata from the Chinese General Social Survey conducted in 2015. The methodologies employed in this paper include the Mincer earnings function (1974) and the Blinder–Oaxaca decomposition (1973). The empirical results reveal that the gender wage gap was over 20% in China in 2014. Education has a significantly positive influence on wages, but the rate of return of education on wages differs for male and female groups. The effects of education on wages in various regions in China are different. Education contributes to narrowing the difference in wages between female and male workers in West China. Keywords Education, gender wage gap, Chinese labor market, mincer earning function. JEL Classification: J16, J21, J24, J31 * [email protected] @vsb.cz (corresponding author) 70 Ekonomická revue – Central European Review of Economic Issues 24, 2021 Gender Gap in Earnings in China: A Cross Sectional Study Yajun MENG 1. Introduction After 1978, the mobility of the labour force caused internal migration from rural areas to urban areas in China. The development of the Chinese labour market in recent years shows broken segmentation caused by the earlier household registration system (Hukou ). Although the Hukou still has effects on workers’ individual welfare and their employment opportunities in some cities, this factor is now less important in the Chinese labour market (Weng, 2016). Before 1978, state-owned enterprises dominated the national economy in China. The Government uniformly arranged college graduates’ employment. Female and male workers had equal opportunities for employment, and there was little difference in their wages, which were decided by the government rules instead of the individ-ual characteristics of human capital (Wang, 2005). With the reform to a market-oriented economy, non-state-owned enterprises and businesses with multiple properties created a huge demand for labour in China. By the end of 2017, there were more than 27 million private enterprises and 65 million individual and private businesses in China in total. These private enterprises and businesses account for 90% of all enterprises in China. Non-state-owned entities provide more than 80% of all jobs (Wuhan University, 2018). The decisions to hire or fire workers and set the wages are the duties of companies (Cai et al., 2009). In addition, women’s participation in the labour force in China is much higher than the global level; see Table 1. Table 1 Female labour force participation rate in 2012–2017 (%) Year 2012 2013 2014 2015 2016 2017 China 63.3 63.1 62.8 62.4 62 61.5 World 48.9 48.93 48.85 48.79 48.88 48.68 Source: World Bank (2018) With the improvement of the social living and economic level in China, more and more educated workers are entering the labour market. The number of graduates with college and junior college degrees is continually increasing, as shown in Table 2. Table 2 College and junior college graduates in China in 2008–2015 (million) Year 2008 2009 2010 2011 Graduates 5.12 5.31 5.75 6.08 Year 2012 2013 2014 2015 Graduates 6.25 6.38 6.59 6.80 Source: National Bureau of Statistics of China, 2018 With the development of its modern labour mar-ket, the income differential between male and female workers in China became a social phenomenon. The difference in monthly income between men and women was 22% in 2017, and the gender wage gap was over 30% in the early years (Jiang, 2018). The aim of this paper is to study the gender wage gap using a rich and new longitudinal data set. For that purpose, it analyses the gender differential of wages in West, Central, and East China and the contribution of education to individual wages. The Mincer earnings function and Blinder–Oaxaca de-composition are used to test the effects of relevant factors on wages and to identify the differences in wages between men and women. Our contribution is to analyse the effects of various levels of education on personal wages and to study the differences between West China, Central China, and East China with the newest available data. The remainder of this paper is organized as follows. Section 2 summarizes the empirical papers that relate to this topic in China. Section 3 introduces the data used for this empirical analysis. Section 4 out-lines the empirical models. Section 5 provides an analysis of the empirical results. Lastly, conclusions and some policy recommendations are provided in Section 6. 2. Literature review This section presents some studies that are relevant to this topic. The work of G. Becker (1964) and J. Mincer (1958) starts the discussion on the relationship between human capital, such as education, and personal earnings. Empirical studies from China show that the return of education on earnings has increased since the 1990s and the capability of individual education has improved personal wages significantly. A high education level has an increasing marginal return for wages. However, the effect of education does not dominate the wages of workers, and other factors, such as the industry, the workplace, and so on, play important roles. The expanding scale of college education does not decrease the return of high educa-tion on personal earnings but, Y. Meng – Gender Gap in Earnings in China: A Cross Sectional Study Wage differential of female and male employee in China 71 in contrast, increases it. The characteristics of human capital make different contributions to the difference in men’s and women’s wages, and the gender wage gap in China keeps changing over time (Fang and Huang, 2017; Park and Qu, 2013; Wang, 2005; Yang and Wang, 2015). The gender difference in wages in China became a new social and economic problem after its market-oriented reform. The mobility of labour from state-owned enterprises to non-state-owned companies and the transfer of workers from the secondary sector to the tertiary sector have changed the structure of the labour market. The gender wage gap gradually wid-ened from 1995 to 2007 in Chinese urban areas, and the differential of men’s and women’s wages declined after 2007 and has picked up again since 2014 (Song et al., 2017). Chen (2011) states that the gender wage gap increased by 10% from 1989 to 2009. The largest gender wage gap is in the group of employees who are over 40 years old and have less education in the non-state-owned sector (Zhang, 2004). Women with high education have a higher rate of educational return on their personal wages than men, but women have lower wages than men (Liu, 2008; Peng, 2011). The level of education is negatively related to the degree of discrimination (Liu, 2008). The difference in wages between workers who have more education and workers who have less education is greater for women than for men. The gender wage gap is caused by a variety of factors. The economic transition has created differences in wages between the sectors and the regions in China. The sectoral features determine the wage level and the difference in wages of workers. The gender wage gap in China is mainly caused by the inter-sectoral factor (Peng, 2011; Song et al., 2017). Many small stateowned enterprises became bankrupt due to the reform of enterprises in the mid-1990s. Workers moved from state-owned enterprises to private businesses. The privatization of the Chinese economy increased the gender wage gap since more and more women were employed by private enterprises. This increased the discrimination in the labour market (Chen, 2011). Meanwhile, migrants moved from rural China to urban China. They worked in various industries and different enterprises instead of being concentrated in the agriculture sector. The gender wage gap is more obvious in the private sector (Chen, 2011). However, the social insurance and the employment protection for migrants in informal employment are not enough (Cai et al., 2009). A new regulation, The Employment Contract Law, was implemented in China in 2007. The enforcement of this law allowed the workers who bore the worst working conditions to require higher payments. This improved the equality of workers’ earnings in the labour market (Cai et al., 2009). The labour market policy, such as The Minimum Wage Policy introduced in 2007, positively affected the gender difference in wages (Song et al., 2017). Wang (2005) studies the gender wage gap in China with employment data from five big cities. The decomposition results show that the difference in wages between the sectors dominates the gender wage gap. More than 80% of the gender wage gap in an industry cannot be explained. The discrimination against women is significant. A similar result is report-ed by Chen in 2011. Li and Dong (2008) employ the characteristics of enterprises as the explanatory variable to study the gender wage gap. Their results show that the return of education on wages decreases significantly when adding relevant factors. The scale of the enterprise, the privatization of the company’s property, and the outside competitive environment are important in determining the gender wage gap. The previous studies also test the wage gap between informal employment and formal employment. Hou (2013) states that the characteristics of workers can explain one-third of the wage gap between formal employment and informal employment in cities. Women who work in the informal sector face more discrimination in the Chinese labour market. Park and Qu (2013) investigate the microdata of six Chinese cities and state that the wage difference in the return to education between the formal sector and the informal sector has increased. Workers in the informal sector have a lower return of education on wages than workers in the formal sector. Furthermore, the educa-tional return increases with quantiles for informal employment. Some studies state that the declining fertility of the young generation in urban China and the increasing number of single individuals will narrow the gender wage gap. Unmarried workers have a smaller gender wage gap than married workers who have one or more children (Song et al., 2017). Zhang et al. (2006) analyse the wage gap between different areas with spatial econometrics and indicate that the economic transition, the local regulation, the local education, a change in the company’s ownership, and capital investment affect the wage gap in the different areas. 3. Data and methodology 3.1 Data This section introduces the microdata and methodology used in this paper. It employs microdata from the Chinese General Social Survey (henceforth CGSS) in 72 Ekonomická revue – Central European Review of Economic Issues 24, 2021 the year 2015. This project has been run by the National Survey Research Center at Renmin Universi-ty of China since 2003 and is organized with 48 other universities. It consists of a survey conducted with a representative sample of 10,000 Chinese households in mainland China. It collects original data on the level of social, family, and individual residences, including basic information and some specific re-sources necessary to study the different fields in each survey systematically. The CGSS focuses on the change in relationship between the social structure and the quality of life in China. It is the main data-base for social and related research for academic institutions and the government. The CGSS conducted in 2015 provides the most recently disclosed data. It covers 23 provinces and four municipalities in China and contains interviews with about 11,000 randomly chosen individuals and more than 1,000 variables. This paper selects the sample of workers who are 18 to 59 years old with full-time employment. It excludes people who serve in the army and farmers. The purpose of this paper is to study the wage gap between men and women with a full-time job. It therefore excludes part-time workers and unemployed workers. In 2014, the average minimum wage in China was about 1,050 CNY per month, so it selects individuals who receive more than 12,500 CNY per year. Thus, the total number of observations is 2,359 (men account for 57.4% and women for 42.6%). In this paper, the workplaces of the observed workers are in East China, Central China, and West China according to the definition from the China Health Statistics Yearbook. East China is more developed than Central China and West China. Of the top 10 provinces or municipalities regarding per capita income in China in 2014, seven of them belong to East China. The observed factors that have an impact on personal wages include demographic factors and em-ployment factors. The demographic factors are working experience, education, and workplace. The second group consists of job-related variables. Education is an important factor in the human capital model and is a measurable ability. According to Xie and Hannum (1996), the duration of the Chinese educational degree is determined by the time spent to achieve the certificate. The duration of different education levels ranges from 3 years to 19 years in this analysis. Students complete primary school education after 6 years (3 years for lower education); middle school after 9 years; high school after 12 years; junior middle technical school after 11 years; special secondary 1 Edu0–Primary school and less education; Edu1 –Junior middle school; Edu2–High school and technical secondary school after 13 years; junior college after 15 years; college or university after 16 years; and postgraduate education after 19 years. The educational level consists of five categories, each representing different levels of educational attainment. These five categories are primary school and less, junior middle school, high school and tech-nical or secondary school, college and junior college, and postgraduate education. The enterprise’s ownership also causes a difference in wages. In the paper, employees are separated into five categories according to the ownership of the company: workers who are self-employed, workers who work in a private enterprise, staff who work in a state-owned or collective enterprise, employees who are hired by foreign enterprises or enterprises from Hong Kong, Macao, and Taiwan, and staff who work in a government or institutional organization. The descriptive statistics of the male and female workers selected are provided in Table 3. About 60% of those workers have the urban Hukou in a city. The proportion of workers who have education from a college and junior college is much larger than other groups regarding the education level for women. The shares of men who have middle school education, high school and technical secondary school educa-tion, and college and junior college education are approximate, and each accounts for about 30% of all male workers. The higher proportion of women with higher education proves that women need more education to be employed in a similar position to men. Women have a higher mean value of years of educa-tion than men. Table 3 Descriptive statistics and distribution of observed workers by gender (%) Variables Male Female Difference Experience (years) 21.45 (11.14) 19.38 (10.34) 2.07 Education (years) 12.00 (3.51) 12.34 (3.83) -0.34 Mean log hourly wage 3.09 (0.63) 2.91 (0.57) 0.18 West China 14.69% 15.74% -1.05% Middle China 28.93% 24.20% 4,73% East China 56.38% 60.06% -3.68% Edu01 7.52% 10.26% -2.74% Edu1 29.45% 24.70% 4.75% Edu2 29.37% 23.80% 5.57% school; Edu3–College and the junior college degree; Edu4– Master and higher degree. Y. Meng – Gender Gap in Earnings in China: A Cross Sectional Study Wage differential of female and male employee in China 73 Edu3 31.37% 38.55% -7.18% Edu4 2.29% 2.79% -0.50% Hukou (Urban) 64.94% 58.97% 5.97% P12 15.86% 15.44% 0.42% P2 46.86% 46.51% 0.35% P3 15.06% 11.75% 3.31% P4 2.44% 2.79% -0.55% P5 19.78% 23.51% -3.73% Observations 1,355 1,004 2,359 Source: author’s calculation, standard deviations are given in brackets Considering the individual employment, 46% of the workers work in private enterprises, while about 16% of the observed workers are self-employed. The proportion of respondents who work in the govern-ment or institutions is about 24% for women and about 20% for men. The share of men who work in state-owned and collective enterprises is 4% larger than that of women. Employment in the government or institutions is more attractive to women, while men prefer competitive work. The average number of working years for men and women is 21 and 19, respectively. The average annual wage is about 55,900 CNY for men and 44,700 CNY for women. Women’s average wage is 24% lower than men’s. Figure 1 shows the distribution of log hourly wages of the observed workers by gender. Women have a left-oriented distribution of wages relative to men. Figure 1 Density distribution of log wage by gender Source: Author calculated Without a doubt, the average wage of workers in East China is much higher than that in Central China and West China. The wage distribution of workers in Central China and West China is similar; see Figure 2. 2 P1– Individual and private business; P2–Private enterprise; P3–State-own and collective enterprise; P4–Foreign and Figure 2 Density distribution of log wage by region Source: Author calculated The employees who work in private companies in West China have the lowest average wage, which is one-third of the highest average wage of workers who are employed by foreign fund companies in East China. Education improves the individual income significantly. Workers with higher education have greater equality in the wage distribution; see Figure 3. The education at the primary school level and lower and junior middle school education show little differ-ence in the effects on individual wages. Figure 3 Density distribution of log wage by level of education Source: Author calculated 3.2 Methodology The empirical analysis starts with the standard Mincer earnings function to express the relationship between the individual wages and the personal capability of human capital, such as education and working expe-rience (Mincer, 1974). It is given by: 𝐿𝑛(𝑤𝑎𝑔𝑒)= 𝛽0+𝛽1𝐸𝑑𝑢 +𝛽2𝐸𝑥𝑝 +𝛽3𝐸𝑥𝑝2+𝜇, (1) Hong Kong, Macao and Taiwan fund enterprise; P5–Government and institutions. 0.2 .4 .6 .8 kdensity lnwage 2 3 4 5 6 Male Female 0.2 .4 .6 .8 kdensity lnwage 2 3 4 5 6 West Middle East 0.2 .4 .6 .8 kdensity lnwage 2 3 4 5 6 Edu0 Edu1 Edu2 Edu3 Edu4 74 Ekonomická revue – Central European Review of Economic Issues 24, 2021 where Ln(wage) is the natural logarithm of the gross hourly wages, Edu stands for the average years of schooling, Exp represents the labour market experience, and 𝐸𝑥𝑝2 is a variable of squared years of work experience describing the decline of wages as the worker ages. Years of education is a continuous variable. In this paper, the years of education of observed workers are calculated according to their educational level, as mentioned above. Here, 𝛽𝑖 are the unknown coefficients to be estimated and 𝜇 is the error term, which contains other factors that influence a worker’s wage. This study estimates the model using ordinary least squares (OLS). The expanded Mincer earnings function is used to test the impact of gender on individual wages, shown as 𝐿𝑛(𝑤𝑎𝑔𝑒)= 𝛽0+𝛽1𝐹𝑒𝑚𝑎𝑙𝑒 +𝛽2𝐸𝑑𝑢 +𝛽3𝐸𝑥𝑝 + 𝛽4𝐸𝑥𝑝2+ 𝜇, (2) where 𝐹𝑒𝑚𝑎𝑙𝑒 is the dummy variable of gender that equals one for women and zero for men. Moreover, the linear regression to estimate the logarithm wage of individual i in group A is 𝐿𝑛(𝑤𝑎𝑔𝑒𝐴𝑖) = 𝑋𝐴𝑖𝛽𝐴𝑖 +𝜇𝐴𝑖, 𝐸(𝜇𝐴) = 0, (3) where Χ is a vector of individual characteristics, β is a vector of coefficients to be estimated, and μ is an error term. The model is rearranged to include demographic factors such as the type of hukou, the workplace of the worker, and the type of education. Another dummy variable represents the different business properties. The age of the observed workers is not included in this paper. 3 In addition, we present the results of the Blinder– Oaxaca decomposition (Blinder, 1973; Oaxaca, 1973) based on the Mincer earnings regression to identify the individual effects of observable factors on the gender wage differential. The difference in wages between men and women is shown as the difference in the linear prediction at the group-specific means. Let 𝛽 󰆹𝑚 and 𝛽 󰆹𝑓 be the regression estimates of the coefficient of men and women, respectively; then, the mean difference in log wages between men and women is decomposed as 3 We omit the age of worker due to its collinearity with variable of working experience in this paper. 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 = 𝐿𝑛𝑊𝑚−𝐿𝑛𝑊𝑓 =(𝑋𝑚 ′−𝑋𝑓 ′)𝛽 󰆹𝑚 +𝑋𝑚 ′(𝛽 󰆹𝑚−𝛽 󰆹𝑓) +(𝑋𝑚 ′−𝑋𝑓 ′)(𝛽 󰆹𝑚−𝛽 󰆹𝑓). (4) This equation decomposes the gender wage gap into three parts. (𝑋𝑚 ′−𝑋𝑓 ′)𝛽 󰆹𝑚 is the part of the wage gap that can be explained by the differential of the observed characteristics of individual workers (the endowments). 𝑋𝑚 ′(𝛽 󰆹𝑚−𝛽 󰆹𝑓) represents the difference in wages caused by the differential of the coefficients between the two groups. (𝑋𝑚 ′−𝑋𝑓 ′)(𝛽 󰆹𝑚−𝛽 󰆹𝑓) is the interaction that presents the difference in coefficients and endowments simultaneously (Guo et al., 2011; Jann, 2008). To express the twofold decomposition of the wage differential, equation (4) is used: 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 = 𝐸𝑥𝑝𝑙𝑎𝑖𝑛𝑒𝑑 𝑃𝑎𝑟𝑡 + 𝑈𝑛𝑒𝑥𝑝𝑙𝑎𝑖𝑛𝑒𝑑 𝑃𝑎𝑟𝑡. (5) The explained part shows the gender difference in wages caused by the endowment, and the unexplained part represents the gender difference in wages caused by the coefficient and the interaction. In this paper, the index problem 4 (Guo et al., 2011; Oaxaca, 1973) is not considered. Male workers form the reference group in this analysis. 4. Empirical results This section presents the empirical results. The results of the ordinary least squares (OLS) regression are displayed in Table 4. Overall, the model fits the data well. The coefficients have the expected signs and are significant at the conventional significance levels. A positive association between education and personal wages is found. In particular, the marginal return rate of one year of education is 7.2% for all workers. One year of education and working experience have more benefits for men. The gender difference in wages is significant, and male workers have wages that are 21% higher than those of female workers. This result is in line with the previous literature (for instance Zhang, 2004). Table 4 Estimates of Mincer earning regressions Model 1 Model 2 Female Male Total 4 The index problem is the choice of reference group in the model. If a different reference group is chosen by gender, it would lead to a different result of wage decomposition. Okomentoval(a): [ARA1]: Replenished? Y. Meng – Gender Gap in Earnings in China: A Cross Sectional Study Wage differential of female and male employee in China 75 Female -0.213** (9.23) Education (Years) 0.068** (13.31) 0.075** (15.45) 0.072** (20.45) Experience 0.016* (2.52) 0.024** (4.00) 0.021** (4.82) (ExpSq*)2 -0.037* (-2.43) -0.055** (- 4.21) -0.043** (-4.82) Intercept 1.942** (19.88) 2.001** (21.58) 1.856** (26.96) Observations 1,004 1,355 2,359 Adjusted R2 0.20 0.18 0.21 *p<0.05; **p<0.015 ExpSq*= Experience/100 Source: author’s calculation, standard deviations are given in brackets Education improves individual wages significantly across China. The contributions of education to the wages of workers in West China, Central China, and East China are different. An extra year of education makes the greatest positive contribution to the wages of workers in East China. Women’s personal wage benefits more from one year of education than men’s in West China; see Table 5. Elementary education has a positive effect on earnings for women in West China. Education at junior middle school shows significant and positive effects, increasing women’s wages by 30% in West China. Education at high school and technical sec-ondary school increases women’s wages by 55% in West China. In East China, individual education contributes more to men’s wages than to women’s wages. Although high education, including junior college, college, and postgraduate education, have different effects on workers’ wages in West China, Central China, and East China, it is the most efficient investment to improve individual wages; see Table 6. Table 7 displays the results of expanding the Mincer regression. It appears that Hukou did not affect individual wages significantly in 2014. The location of work is an unignorable variable, showing that male workers in East China have a 26% higher wage than women in West China. Women would earn a 17% higher wage if they moved from West China to East China, and men would have a 26% higher wage if they were employed in East China. Jobs in East China are attractive to workers. A positive association is consistently obtained between the level of education and the personal wages. This result is also in line with the previous literature (including citations). Education at junior college and 5 * significant at 1% and ** significant at 5%. college increases personal wages by 30% in Central China and by more than 50% in West China and East China when controlling all the explanatory variables. Investment in postgraduate education increases personal wages by more than 100% for men and women in China. Each successive educational stage can double personal wages when controlling the explanatory variables for men and women. Considering the effects of the different properties of enterprises on individual wages, both men and women working in the government and institutions and in stateowned and collective enterprises have significantly lower wages than workers who are self-employed (as the reference group). Having a job in the government or an institution reduces women’s wages by 39% and men’s wages by 29%. Regarding the difference in wages between groups produced by the ownership of enterprises, it shows a greater effect on the wages of female workers. The average difference in wages between the reference group and each of the other groups for women is 10% larger than that for men, respectively; see Table 7. Following the methodology presented above, the results of the twofold decomposition are presented in Table 8. Prediction-1 shows that the mean log hourly wage is 3.09 for men, and Prediction-2 indicates that the mean log hourly wage is 2.91 for women, yielding a wage gap of 0.18. Education, which is the observed characteristic of human capital in this paper, narrows the gender wage gap. The possible reason is that the observed female workers have a higher average number of years of personal educational attainment than male workers by 0.34 years. Although the share of employees with college or junior college education is larger for women than for men in this paper, the average wage is lower for women than for men. workers have a higher average number of years of personal educational attainment than male workers by 0.34 years. Although the share of employees with college or junior college education is larger for women than for men in this paper, the average wage is lower for women than for men. It seems that the difference in wages between men and women cannot be explained by the observed variables. The average compensation that men receive for the advantage of experience roughly equals the average compensation that women receive for the advantage of education. Unknown reasons dominate the differences in wages between male and female. It is hard to test the Okomentoval(a): [ARA2]: Try to divide the variable experience2 /100, to scaling the coefficients. 76 Ekonomická revue – Central European Review of Economic Issues 24, 2021 level of discrimination based on these limited variables of individual workers in this research. Table 5 Estimating return of one-year education on income Female Male West Middle East West Middle East Education (Years) 0.073** (4.78) 0.050** (4.95) 0.084** (11.42) 0.048** (3.74) 0.050** (4.95) 0.099** (14.15) Experience Yes Yes Yes Yes Yes Yes Expericence2 Yes Yes Yes Yes Yes Yes Employment Yes Yes Yes Yes Yes Yes Observations 158 243 603 199 392 764 Adjusted R2 0.23 0.12 0.30 0.15 0.12 0.26 * p<0.05; **p<0.01 Source: author’s calculation, standard deviations are given in brackets Table 6 Estimates of the return on wage by the level of education Female Male West Middle East West Middle East Primary school and less education Reference Reference Junior middle school 0.304* (2.02) -0.095 (0.97) -0.028 (0.33) 0.097 (0.70) -0.016 (0.15) 0.152 (1.75) High school and technical secondary school 0.554** (3.27) -0.010 (0.09) 0.242** (2.83) 0.356* (2.44) 0.154 (1.39) 0.285** (3.22) College and junior college 0.823** (4.51) 0.361** (2.77) 0.646** (7.13) 0.468** (2.92) 0.403** (3.24) 0.838** (9.09) Master and higher degree 1.442** (3.37) 0.754** (2.67) 1.073** (7.67) 0.354 (0.90) 0.742** (3.14) 1.404** (9.37) Experience Yes Yes Yes Yes Yes Yes Expericence2 Yes Yes Yes Yes Yes Yes Employment factors Yes Yes Yes Yes Yes Yes Observations 158 243 603 199 392 764 Adjusted R2 0.25 0.16 0.30 0.16 0.13 0.27 * p<0.05; ** p<0.01 Source: author’s calculation, standard deviations are given in brackets Table 7 Estimating wage of the female and male employee in 2014 Female Male Female Male Demographic Experience 0.015** (2.61) 0.026** (4.33) 0.021** (3.37) 0.027** (4.51) Experience2 0.000* (2.33) -0.001** (4.52) -0.000** (3.28) -0.001** (4.79) Okomentoval(a): [ARA3]: What I understand is that you consider as dependent variable the log of wages, and then you can control for…. Education, and bla bla, It is a bit confusing, once you use the average years of schooling, or you just use dummy variables to capture the level of education reached by individuals (dummies) Okomentoval(a): [P4R3]: Yes , it if log of wageYears of education Okomentoval(a): [ARA5]: Footnote is missing, with p values, and additional information Okomentoval(a): [P6R5]: Yes as footnote7 Okomentoval(a): [ARA7]: In my view, we also need to see the coefficients. Okomentoval(a): [P8R7]: It will be too big table since aim of this talbe only to show the effects of typies of educaion Okomentoval(a): [ARA9]: Please speficy if this a p value, and what is between brackets, in all estimation tables Okomentoval(a): [P10R9]: Yes Y. Meng – Gender Gap in Earnings in China: A Cross Sectional Study Wage differential of female and male employee in China 77 Hukou (Rural) Reference category Hukou (Urban) 0.093 (2.00) 0.054 (1.41) 0.102* (2.23) 0.068 (1.77) West Reference category Middle -0.049 (0.99) -0.039 (0.87) -0.038 (0.77) -0.034 (0.74) East 0.177** (3.95) 0.264** (6.18) 0.172** (3.82) 0.263** (6.14) Education Years 0.074** (11.32) 0.077** (13.23) Primary school and less education Reference category Junior middle school 0.004 (0.08) 0.085 (1.47) High school and technical secondary school 0.204** (3.17) 0.236** (3.86) College and the junior college 0.589** (8.32) 0.644** (9.43) Master and higher degree 1.039** (8.37) 1.111** (7.91) Employment Individual and private business Reference category Private enterprise -0.315** (6.01) -0.194** (4.01) -0.305** (5.78) -0.188** (3.84) The State-owned and collective enterprise -0.302** (4.49) -0.185** (3.06) -0.289** (4.34) -0.164** (2.71) Foreign and Hong Kong, Macao and Taiwan fund enterprise -0.029 (0.22) 0.050 (0.42) -0.031 (0.24) -0.026 (0.22) Government and institutional organization -0.388** (6.04) -0.292** (5.06) -0.391** (6.18) -0.283** (4.91) Intercept 1.967** (16.27) 1.945** (17.53) 2.526** (24.38) 2.527** (25.94) Adjusted R2 0.270 0.250 0.276 0.255 Observations 1,004 1,335 1,004 1,335 * p <0.05, ** p <0.01 Source: author’s calculation, standard deviations are given in brackets