


































Advances in Politics and Economics 
ISSN 2576-1382 (Print) ISSN 2576-1390 (Online) 

Vol. 6, No. 2, 2023 

www.scholink.org/ojs/index.php/ape 

53 
 

Original Paper 

The Relationship between Age and Growth Rate of Wages and 

the Gender Wage Gap in the Financial Sector in China 

Kexin Chen
1*

 

1
 Department of Economics, Huazhong University of Science and Technology, Wuhan, China 

*
 Kexin Chen, E-mail: u202016385@hust.edu.cn 

 

Received: February 18, 2023       Accepted: March 20, 2023     Online Published: April 12, 2023 

doi:10.22158/ape.v6n2p53            URL: http://dx.doi.org/10.22158/ape.v6n2p53 

 

Abstract 

Using CGSS data, this paper explores the relationship between age and wage growth rates in China’s 

financial sector by developing a fixed effects model. I find that the wage growth rate declines slowly 

with age, although the absolute value of wages has been rising. After a brief comparison of the 

financial sector, the information technology sector, and agriculture, I find that the pattern of their wage 

growth rates is similar, i.e., the wage growth rate declines slowly with age. Then, through empirical 

tests, I find that a gender wage gap does exist in the finance industry and that the gender wage gap 

gradually increases with age. Based on this, I further investigate the relationship between the age of 

female workers and the wage growth rate, and the possible reasons why the wages of female workers 

are lower than those of male workers. I find that the wage growth rate of female workers also gradually 

decreases as their age increases. Moreover, having children has a significant negative effect on the 

wage growth rate, but the marital status does not. 

Keywords 

fixed effects model, wage growth rate, gender wage gap, women’s wage, Chinese financial sector 

 

1. Introduction 

With the accelerating development of science and economics, information technology and the financial 

industry have become two of the most popular industries in China. The number of listed companies in 

China’s financial industry was only 10 in 1990, but with the continuous development of reform and 

opening up, the financial industry developed rapidly, with 2063 listed companies in 2010 and 4154 in 

2020. As of 2017, the scale of China’s online assets was nearly $3.5 trillion, and the cumulative 

transaction volume of online payment, online crowdfunding and online lending reached $70 trillion. 

As we all know, in the information technology industry, the market prefers creative young people, and 



www.scholink.org/ojs/index.php/ape          Advances in Politics and Economics                     Vol. 6, No. 2, 2023 

54 
Published by SCHOLINK INC. 

they tend to be promoted and increased in salary fast. What about the financial industry? What is the 

relationship between age and the rate of a salary increase? It’s an interesting question. Besides, there 

have been more and more complaints about gender discrimination in the financial industry and the 

issue of gender discrimination in the financial services industry has captured a lot of headlines in recent 

years. According to Deloitte, as of June 2021, women held less than 6% of CEO positions within S&P 

500 companies across all industries and just 4% of CEO positions within S&P 500 financial services 

institutions. Is this true? If it is true, does gender discrimination against women get worse as they age? 

And what causes gender discrimination. 

In this paper, I explore the relationship between age and the growth rate of wages, that is, whether the 

increase in age is conducive to the increase of wages. Then compare the results with the computer 

industry and the education industry. On this basis, I focus on the gender wage gap in gender 

discrimination and further examine whether there is a difference in the wage growth between male and 

female groups, and what is the difference. Finally, I try to find the possible and reasonable reasons for 

the above research results. 

The main contributions and novelties of this article are as follows. It is well known that wages usually 

grow over the life cycle. But little attention has been paid to which age groups are seeing the fastest 

wage growth and why, especially in the financial industry. And there is a large literature on 

male–female wage differentials. However, most of them are based on wage level and very few focus on 

wage growth rate. Thus, my study differs from the majority of previous work in this area because I 

focus on wage growth in the financial industry rather than absolute wage levels and try to find 

reasonable reasons to explain it, although I will also discuss age in relation to wages. Additionally, 

while the most of Chinese research on wages is macro, such as based on provincial panel data, I plan to 

use a micro data set called chip. 

In conclusion, this study contributes to a more scientific understanding of the wage growth rate and the 

current situation of the gender wage gap in China’s financial industry, and further analysis the sources 

of the gender wage gap, to provide a reference for the formulation of public policies to promote the 

equality and reasonable employment and wage of men and women in the labor market. 

I have developed hypotheses for how the provision will impact the dependent variables based on 

literature and economic theory. Hypotheses are as follows: 

Hypothesis 1: As age increases, the rate of wage growth gradually falls. 

Hypothesis 2: There is a significant gender wage gap, and it becomes wider with age. 

Hypothesis 3: Children and marriage can negatively affect women’s wages. 

The remainder of this paper proceeds as follows. Section 2 is the literature review of the previous work, 

section 3 is the description of data, section 4 describes models and empirical strategies, section 5 shows 

the results of regression models, section 6 checks robustness and discusses limitations, and section 7 is 

the conclusion. 

 



www.scholink.org/ojs/index.php/ape          Advances in Politics and Economics                     Vol. 6, No. 2, 2023 

55 
Published by SCHOLINK INC. 

2. Review of Literature 

As for age and Wage Growth rate, most researchers focus on life cycle wage growth. David Lagakos 

(2018) researched how life cycle wage growth varies across countries, he found workers have more 

advantages in rich countries than in poor countries due to more accumulation of human capital or fewer 

search frictions. And there is a lot of research about factors contributing to the differences in life cycle 

wage growth. The canonical model of wage growth is the on-the-job model of human capital 

investment (e.g., Ben-Porath, 1967; Becker, 1964). Ronni Pavan (2011) built a search model to 

distinguish the relative importance of various factors for generating wage growth over the life cycle.  

On the premise of the economic growth model, Murphy (1989) studied the relationship between 

capitalists’ profits and workers’ wages, as well as capitalists’ savings rate and workers’ wages, and 

concluded that the higher the capital accumulation rate and profit rate, the lower the wage growth rate 

of workers. In order to show the importance of a certain factor in wage determination, Chewei Zhang 

and Xinxin Xue (2008) discussed the influence of ownership factors. Xiuyan Liu (2019) used market 

potential to explain wage differences. Zewen Yang and Quanfa Yang (2004) discussed the effect of 

foreign direct investment on wages. Yuan Zhang and Jianqi Chen (2018) discussed the wage effect of 

industry characteristics. 

There are also quite a few scholars who focus on the gender gap in the life cycle of wage growth. Men 

and women have significantly different life cycle labor market outcomes, as sizable literature has 

documented (e.g., Betrand et al. (2010), Goldin (2014), Adda et al. (2017)). Hill (1979) was one of the 

first to examine the effect of motherhood on wage levels. She initially finds a 7% motherhood wage 

penalty for White women, but after controlling for productivity characteristics it nearly disappears. 

Waldfogel’s (1998a, 1998b) findings suggest a motherhood wage penalty of 4.6% for the first child and 

12.6% for two or more children. There is also a large literature on the motherhood penalty. Additional 

papers to the ones discussed above include Waldfogel (1997), Lundberg and Rose (2000), Anderson et 

al. (2003), Gangl and Ziefle (2009), and Pal and Waldfogel (2014).  

Besides of penalty of being a mother, there are some researches that reflect that a penalty appears to be 

associated only with being a woman. This is true not only in countries such as the U.S., where job 

interruptions and the associated wage penalties for women are common (Bronson, 2015) but also in 

countries like China. Wang Meiyan (2005) adopted Brown’s decomposition method and found that only 

6.95% of the gender wage difference could be attributed to personal characteristics, while 93.05% was 

caused by discrimination. Yuhao Ge (2007) used the method of quantile decomposition to study the 

gender wage difference and found that women were at a disadvantage in terms of the distribution of 

years of experience and the rate of return on experience, while women were no worse than men in 

terms of the level of education and the rate of return on education. Shi Li et al. (2014) used the 

Oaxaca-Blinder decomposition method to decompose the dynamic changes in the gender wage gap by 

using the data of the Chinese Household Income Survey in 1995, 2002, and 2007, and found that 

during the period from 1995 to 2007. In particular, from 2002 to 2007, the gender wage gap in China’s 



www.scholink.org/ojs/index.php/ape          Advances in Politics and Economics                     Vol. 6, No. 2, 2023 

56 
Published by SCHOLINK INC. 

labor market widened significantly, with an increasing proportion of the unexplained part. 

 

3. Data 

3.1 Data Source 

In this paper, I use data from the Chinese General Social Survey (CGSS), which is panel data. The 

Chinese General Social Survey (CGSS) was initiated in 2003 and is the earliest national, 

comprehensive, and continuous academic Survey project in China. It is a micrograph of households and 

individuals, covering more than 10,000 households in 125 county-level units and 1000 village-level 

units, and followed every 1-2 years. 

So far, this survey has been conducted ten times. I choose data obtained from the last three surveys, 

2013, 2015, and 2018, including 92077 subjects. Because I study people who work in the financial 

industry, I reserve data on individuals who work in the financial sector, in addition to which I collect 

their personal information, including income, gender, year of birth, marital status, educational 

background, monthly working hours, and whether they have children. For comparison purposes, I also 

reserve the sample data from the information technology industry and the agriculture industry (the 

questionnaire used for CGSS data collection combines agriculture, forestry, livestock, and fisheries into 

one industry, which is referred to as agriculture in this paper). So my final sample data includes 9023 

data from the financial industry, 8015 data from the information technology industry, and 10,091 data 

from the education industry. 

3.2 Indicators and Notations 

The explanatory variable in this paper is the rate of wage growth. The rate of wage growth is an 

important measure of an economy’s level of development and people’s living standards. Considering 

that this paper studies the rate of wage growth rather than the absolute value of wages, the logarithm of 

wages is taken as the explanatory variable. Since only annual income is available in the original data, 

so I use annual income to approximate annual wage income. 

There are many factors affecting wages, after referring to a large amount of literature, this paper 

decides to select the following indicators as control variables: working hours (wh), which is the total 

number of working hours per month, including overtime hours, to measure the degree of individual 

hard work; education level (edu), which is expressed by the number of years of education, i.e., 6 years 

of elementary school, 9 years of middle school, 12 years of high school, 15 years of college, and 16 

years of undergraduate and bachelor’s degree is 16 years. 

Age is the central explanatory variable throughout the paper. When exploring whether there is a gender 

wage gap, the core explanatory variables are age and gender. And when exploring the causes of the 

gender wage gap, the core explanatory variables are marital status and children. In the original data, 

marriages are divided into six statuses (first marriage, remarriage, cohabitation, divorce, widowhood, 

and unmarried). To simplify the model, I consider first marriage, remarriage, cohabitation, divorce, and 

widowhood as married. The key mathematical notations used in this paper are listed in Table 1. 



www.scholink.org/ojs/index.php/ape          Advances in Politics and Economics                     Vol. 6, No. 2, 2023 

57 
Published by SCHOLINK INC. 

Table 1. Notations Used in This Paper 

Symbol Description 

ln_w ln(wage): Logarithm of annual income 

gen gender: Defined as a dummy variable, female=1, male=0 

age individual’s age 

mar marital status: Defined as a dummy variable, married=1, unmarried=0 

edu years of schooling 

wh Working hours : hours on working per month, including overtime 

child Defined as a dummy variable, have children=1, no children=0 

 

3.3 Preliminary Data Analysis 

 

Table 2. Summary Statistics 

Variable Obs Mean Std. Dev. Min Max 

gen 9023 0.5465632 0.4981033 0 1 

age 9023 43.34146 10.60275 23 80 

mar 9023 0.7572062 0.4290093 0 1 

edu 9023 14.36364 2.318509 6 18 

wh 9023 184.3692 88.59246 -1980 480 

wage 9023 71723.13 79392.35 -99 1000000 

child 9023 0.7028825 0.4572423 0 1 

 

 

Figure 1. Scatter Plot of Data 

Average retirement age 



www.scholink.org/ojs/index.php/ape          Advances in Politics and Economics                     Vol. 6, No. 2, 2023 

58 
Published by SCHOLINK INC. 

Note. The preliminary optimal fitting result is y = -0.0012x2 + 0.1171x + 8.2327. According to the 2012 

and 2014 “China Labor Force Dynamics Survey” conducted by the Social Science Research Center of 

Sun Yat-sen University, the average retirement age in China over the past decade has been around 53 

years. 

 

 

Figure 2. Wage Gap between Men and Women 

Note. The preliminary optimal fitting result of female data is y=-0.0009x
2
+0.0831x+8.9473. The 

preliminary optimal fitting result of male data is y = -0.0018x2 + 0.1717x + 7.1263. According to the 

2012 and 2014 “China Labor Force Dynamics Survey” conducted by the Social Science Research Center 

of Sun Yat-sen University, the average retirement age in China over the past decade has been around 53 

years. 

 

Table 2 displays observations, mean, standard deviation, median, minimum and maximum values of 

each control and outcome variable. 

The scatter plot and the preliminary fit results in Figure 1 shows that there is a nonlinear relationship 

between the rate of wage growth and age. 

From Figure 2, we can see that in the financial industry, women have a slight gender advantage at the 

initial stage of employment, but men’s wages increase significantly faster than women’s and quickly 

surpass women’s wages, and there is a significant gender wage gap. Moreover, the gender wage gap 

widens and then decreases with age. 

 

 

 

Average retirement age 



www.scholink.org/ojs/index.php/ape          Advances in Politics and Economics                     Vol. 6, No. 2, 2023 

59 
Published by SCHOLINK INC. 

4. Empirical Models and Strategies 

4.1 Panel Regression Models 

 
(1) 

 

(2) 

 

(3) 

I use 3 panel regression models to complete my study. Where ~ , ~ , ~  

represent the regression coefficients to be solved, v represents individual heterogeneity and  is white 

noise. 

Model (1) is used to explore how age affects wages. If the coefficient  and  are significantly 

non-zero, it indicates that there is a significant relationship between age and the rate of wage growth, 

laying the groundwork for subsequent in-depth research. Model (2) is used to investigate whether there 

is a gender wage gap and how the gender wage gap is related to age. If the coefficient β5 is significant, 

it indicates that there is a gender wage gap and there is a significant association with age, then the study 

can be continued to the next step. Model (3) builds on models (1) and (2) to further explore the 

relationship between age and wages for the female group and the possible causes of the gender wage 

gap. The female group data for model (3) is derived from the data of model (1) (2). 

4.2 Empirical Strategies 

4.2.1 Selecting Models by Using Hausman Test 

1) Establish the regression fix effect and random effect model, and conduct the Hausman test. If the 

original hypothesis of the random utility model is rejected, the fix effect model will be adopted. 

2) If the hausman test fails to reject the null hypothesis, the random effect regression is performed first, 

and then the Lagrange Multiplier (LM) test of Breusch & Pagan is performed. If the null hypothesis of 

Pols regression is rejected (that is, the individual heterogeneity is assumed to be zero), the random 

effect regression will be used; otherwise, the Pols regression will be used. In order to avoid the 

correlation between heterogeneous panels and sequences, the robust regression of modified covariance 

matrix was adopted in the regression. 

After conducting Hausmann tests on these three models, I finally build three fixed effect panel models. 

 

5. Results 

The results reported in Table 3 show that most of the regression coefficients pass the significance test at 

the 95% confidence level, indicating the correctness of the data and model selection in this paper. 

 

 



www.scholink.org/ojs/index.php/ape          Advances in Politics and Economics                     Vol. 6, No. 2, 2023 

60 
Published by SCHOLINK INC. 

Table 3. Result of Regression 

Variables Model (1) Model (2) Model (3) 

coefficient P value(P>z) coefficient P value(P>z) coefficient P value(P>z) 

age 0.104 0.000 0.019  0.000  0.094 0.012 

age
2 

-0.001 0.000 \ \ -0.001 0.028 

edu 0.141 0.000 0.144  0.000  0.152 0.000 

wh 0.0006 0.033 0.001  0.088  0.000 0.192 

gen \ \ -0.012 0.000 \ \ 

gen*age \ \ -0.003  0.004  \ \ 

child \ \ \ \ -0.223 0.010 

mar \ \ \ \ -0.107 0.279 

_con 6.092 0.000 7.960  0.000  6.415 0.000 

 

5.1 Model 1  

From the result of model 1, we can see that age affects the wage growth rate. Specifically, each unit 

increase in age will result in a wage change of (10.45-0.18 age) %, which means that as age grows, 

wages grow and the relationship between age increase and the growth rate of wages depends on the age 

of the individual. Furthermore, α2=-0.001 indicates that with the increase of age, the wage increases 

but increase slower and slower. When a person reaches the age of 58, his salary reaches the highest 

level of his career, but he is likely to be retired by this time. Remember that the average retirement age 

in China is 53, as I mentioned in the notes of Figure 1 and Figure 2. By the way, for the control 

variables, the coefficients of education level and working hours are both positive, indicating that they 

both favor wage growth. In summary, salary increases are fastest when you first enter the industry. As 

age increases, wages keep increasing, but the rate of increase decreases by a slight margin. The 

regression results are exactly in line with my hypothesis. Comparing the financial industry with the IT 

industry and agriculture, the results are shown in the Figure below. 



www.scholink.org/ojs/index.php/ape          Advances in Politics and Economics                     Vol. 6, No. 2, 2023 

61 
Published by SCHOLINK INC. 

 

Figure 3. Comparison Chart of the Three Sectors 

Note. The figure shows the best-fit curves for the three industries. The horizontal coordinate is age, the 

vertical coordinate is ln_w, and the marked point is the highest. agr represents agriculture, IT represents 

the information technology industry, and finance represents the financial industry. All data are from 

CGSS. 

 

The following findings can be obtained from the graph. 

1) In terms of the absolute value of wages, the IT industry is always ahead of the other two industries, 

regardless of the age stage. The wage level of the financial industry is similar to that of the IT industry, 

but the wage level of agriculture is far behind that of the financial industry and the IT industry. 

2) From the viewpoint of the growth rate of wages, the growth rate of wages all gradually become 

slower in the rising stage of wages. 

3) It is worth mentioning that if we take the average retirement age of 53 years as the reference, the 

wage in the financial industry is increasing throughout the career, although the wage growth rate is 

slowing down, while the wage in the IT industry and agriculture is increasing first and then decreasing. 

Wages in agriculture start to decline before the age of 40, and the rate of decline is obvious and 

declines faster and faster. 

I think the following reasons can explain these phenomena. 

1) The difference in the total output value of the industry causes the difference in the absolute value of 

wages. In other words, the prosperity and development rate of the industry are closely related to the 

wages of the workers in the industry. Since the reform and opening up, the center of gravity of the 

agr IT finance 



www.scholink.org/ojs/index.php/ape          Advances in Politics and Economics                     Vol. 6, No. 2, 2023 

62 
Published by SCHOLINK INC. 

entire Chinese economy has changed. At first manufacturing and utilities were strong, then the financial 

sector emerged, while technology-related industries have been the top performers. The financial sector 

has grown rapidly through the demand for real estate credit and national infrastructure investment 

credit, while the IT sector has grown through the rise of the Internet industry at the end of the last 

century and the beginning of the century, and so its income levels have risen rapidly. In addition, 

China’s weak agricultural base and irrational agricultural industry structure have contributed to the 

generally low wages of Chinese farmers. According to the Chinese National Bureau of Statistics, in 

2020, the total agricultural output value will increase by 3.4%, while the IT and financial industries will 

both increase by about 13%. 

2) Young people are in the prime of their lives, both physically and mentally, and usually retain 

enthusiasm for their work, especially when they first enter the industry. Young people have more 

potential than middle-aged and older people, and there is more room for promotion and salary increases. 

So it’s easy to see why wages increase at the fastest rate when you’re young. 

3) The three types of workers in these industries represent three types of workers - experienced workers, 

brain workers, and manual workers. industries where experienced workers work, such as the financial 

industry, that have industry barriers. The industry barrier is experience. Experience is accumulated over 

time and the number of operations. With the passage of time, this industry’s barriers are higher and 

higher, eventually forming an extremely strong irreplaceable, as Warren Buffett often said moat. Of 

course, this does not mean that the industry does not require technical and mental labor, just that it pays 

more attention to experience. That’s why wages keep rising throughout a career. The work of brain 

workers is usually somewhat technical, but this technicality is easier to imitate. As they age, their 

mental and physical strength gradually declines and their skills become more and more replaceable, 

their income is likely to show a slow downward trend. The wages of manual laborers are often linked to 

the physical labor they put in, and as they age and their physical strength gradually declines, their 

wages are also likely to show a downward trend. 

Of course, in real life, with the growth of age, IT workers, such as programmers, their wages will not 

necessarily be reduced, but they are likely to face layoffs, forced to change careers, and other serious 

challenges. 

5.2 Model 2 

From the regression results of model (2), all coefficients pass the significance test. The coefficient of 

gen is -0.012 and the coefficient of the interaction term of gen and age is - 0.003, indicating that there is 

indeed a gender wage gap in the financial sector in China, where men’s wages are generally higher than 

women’s wages, and the wage gap slowly widens with a slight margin as age increases. 

After reading a lot of literature, I find some reasons that can contribute to the gender wage gap. 

1) Female pregnancy and childbirth bring pressure on business operations and affect business 

efficiency. 

China’s Regulations on Labor Protection for Female Workers clearly state that female workers are 



www.scholink.org/ojs/index.php/ape          Advances in Politics and Economics                     Vol. 6, No. 2, 2023 

63 
Published by SCHOLINK INC. 

entitled to maternity leave of no less than 98 days, and unfortunately, miscarriages are also entitled to 

15 or 42 days of maternity rest, depending on the circumstances. During this period, enterprises not 

only have to pay direct costs such as wages and filling job vacancies but also have to bear the indirect 

losses caused by the possible decline of work skills and lack of updated knowledge after returning to 

work after childbirth. In order to pursue higher efficiency goals, companies are reluctant to hire women 

or give them the same opportunities for promotion and salary increases as men. In addition, women’s 

pregnancy and breastfeeding will have a certain impact on the efficiency of work, and the double attack 

of family and work may also cause psychological pressure on female employees, which will have a 

certain impact on the effectiveness of enterprises and thus on their own wage increase rate. In recent 

years, with the legalization of three children, more and more women are choosing to have two or even 

three children. However, most women give birth to their second or third child at an age when they are 

in the prime promotion period of their career development. And the career interruption caused by the 

second or third childbirth will inevitably affect the staff planning of enterprises and thus they may miss 

the opportunity for promotion and salary increase.  

2) The stereotypical constraints of traditional gender concepts. 

Research shows that traditional gender concepts are one of the reasons why women are rejected in the 

workplace. The stereotype believes that men are more suitable to participate in social work in terms of 

physical strength, creativity, and adventurous spirit, while women are more suitable to engage in home 

production in terms of sensuality and carefulness, and married women will focus on their families and 

their work efficiency will be greatly reduced compared with that before marriage. In addition, because 

of the stereotype that women do not have the characteristics of managers such as “competitiveness and 

influence” when given the same opportunity for promotion, executives prefer men in leadership 

positions and women in other supporting positions. And this implicit selection bias makes women’s 

career paths more difficult. Li Lu (2016) and Luo Juan (2012) both point out that there is serious 

employment discrimination and gender discrimination in pay, and companies believe that male workers 

create more revenue than women, so they tend to pay men more for their work. 

These reasons make it far more difficult for women to get promotions and wage increases in the 

financial sector workplace than men. Over time, the wage gap with men has become larger and larger. 

So the gender wage gap slowly becomes larger as workers get older. 

In Model 3, I focus on verifying whether children and marital status are the cause of the gender wage 

gap. 

5.2 Model 3 

From the regression results of model (3), the regression coefficients of age, age*age, and child all pass 

the significance test at 95% confidence level, indicating that their coefficients are significantly not 0. 

This means that as age increases, the rate of wages increase for female workers also gradually 

decreases and having children will have a significant negative impact on their wages. As shown in the 

survey, the peak of women’s childbirth is concentrated in the age group of 23 to 40 years old, which is 



www.scholink.org/ojs/index.php/ape          Advances in Politics and Economics                     Vol. 6, No. 2, 2023 

64 
Published by SCHOLINK INC. 

also a critical period for women’s career development to enter the rising period. According to a survey 

conducted by Wisdom Associates, 33% of married women who are not pregnant say that their marital 

status significantly affects their chances of promotion in the workplace. Therefore, since the opening of 

China’s “three-child” policy, women are more worried about their future career development due to 

childbirth issues. This confirms part of hypothesis 3. This confirms part of hypothesis 3. However, the 

coefficient of mar is insignificant, indicating that marital status does not have an effect on wages, which 

contradicts my hypothesis 3. This shows that it is just a stereotype that married women will focus on 

the family and work efficiency will be greatly reduced compared to before marriage. 

 

6. Robustness and Limitations  

6.1 Robustness 

 

Table 4. Robustness Test Results 

Variables Model (1) Model (2) Model (3) 

coefficient P value(P>z) coefficient P value(P>z) coefficient P value(P>z) 

age 0.093  0.000  0.014  0.000  0.094  0.012  

age
2 

-0.001  0.000  \ \ -0.001  0.028  

edu 0.128  0.000  0.130  0.000  0.152  0.000  

wh 0.001  0.056  0.000  0.132  0.000  0.192  

gen*age \ \ -0.004  0.003  \ \ 

child \ \ \ \ -0.223  0.010  

mar \ \ \ \ 0.107  0.280  

_con 6.623 0.000 8.379 0.000 6.308  0.000  

 

In this section, I perform robustness tests by varying the econometric method. I use LSDV (Least 

Squares Dummy Variable method) to reconstruct three models that yield regression results as shown in 

the table below. 

In Model 1, the coefficients of age and age
2
 remain significant and the signs remain unchanged. In 

model 2, the sign and significance of the key variables don’t change, although the control variable wh 

changed from significant to insignificant. The results of model 3 also don’t change significantly. This 

indicates that the results of my study are robust. That’s fine as the purpose here is to check the 

robustness. 

6.2 Limitations 

In this paper, I have two major limitations. First, I use annual earnings to approximate annual wage , 

but annual earnings include wage income, dividends, social benefits, and so on, not just annual wage 

income. Second, I only verify the effect of having children on women’s wages, but I do not examine in 



www.scholink.org/ojs/index.php/ape          Advances in Politics and Economics                     Vol. 6, No. 2, 2023 

65 
Published by SCHOLINK INC. 

depth whether and how having different numbers of children affects women’s wages differently. 

 

7. Conclusion 

This paper uses the CGSS data to explore the relationship between age and the growth rate of wages in 

the Chinese financial sector by building fix effect models. I find that the rate of wage growth decreases 

slowly with age, although the absolute value of wages is always increasing. A simple comparison of the 

financial sector with the information technology sector and agriculture reveals a similar pattern 

regarding the growth rate of wages, with both wage growth rates declining slowly with age. Then I find 

through empirical testing that there is indeed some gender discrimination in the finance industry, which 

can be reflected by the gender wage gap. Moreover, the gender wage gap slowly becomes larger as age 

increases. Based on this, I further studied the relationship between age and the wage growth rate of 

female group and the possible causes of lower wages for female workers than men and found that the 

wage growth rate of female workers gradually decreases with age. Having children is indeed a serious 

penalty for the rate of wage growth, but a change in marital status does not have much effect, so having 

or not having children may be an important reason for the gender wage gap. 

In response to the gender discrimination phenomenon mentioned in this paper, I propose the following 

suggestions: First, improve the relevant laws in the labor market. Through government policy 

protection, legal guidance, and effective supervision, we can protect women’s reasonable and legitimate 

rights and interests, and ensure that they receive proper treatment in the labor market. Gradually reduce 

gender discrimination, and promote the effective use and reasonable allocation of labor resources. 

Second, the government and enterprises should pay attention to the opportunities for women to 

improve their abilities, and appropriately increase the human capital investment of female employees to 

improve their core competitiveness. Third, all industrial sectors, especially the non-state sector, must 

conscientiously implement relevant laws and regulations to effectively protect the legitimate rights and 

interests of the female labor force and ensure that women are treated appropriately in employment, 

salary, and promotion, so as to change the discriminatory mindset implicitly. 

 

References 

Anderson, D. J., Binder, M., & Krause, K. (2002). The motherhood wage penalty: Which mothers pay 

it and why? The American Economic Review, 92(2), 354-358. 

https://doi.org/10.1257/000282802320191606 

Anderson, D. J., Binder, M., & Krause, K. (2003). The motherhood wage penalty revisited: Experience, 

heterogeneity, work effort, and work-schedule flexibility. Industrial and Labor Relations Review, 

56(2), 273-294. https://doi.org/10.1177/001979390305600204 

Bronson, M. A. (2015). Degrees are forever: Marriage, Educational Investment, and Lifecycle Labor 

Decisions of Men and Women. Manuscript. 



www.scholink.org/ojs/index.php/ape          Advances in Politics and Economics                     Vol. 6, No. 2, 2023 

66 
Published by SCHOLINK INC. 

Zhang, C. E., & Xue, X. X. (2008). Wage Differences and Human Capital Contribution between the 

State-owned and Non-State-owned Sectors. Economic Research, 4, 2008. 

Gangl, M., & Ziefle, A. (2009). Motherhood, labor force behavior, and women’s careers: An empirical 

assessment of the wage penalty for motherhood in Britain, Germany, and the United States. 

Demography, 46(2), 341-369. https://doi.org/10.1353/dem.0.0056 

Lagakos, D. et al. (2018). Life cycle wage growth across countries. Journal of Political Economy, 

126(2), 797-849. https://doi.org/10.1086/696225 

Li, S. (2008). Market Competition or Gender Discrimination: The Trend of Increasing Gender 

Difference in Earnings and Its Causes. Sociological Research, 2008(2), 94-117. 

Liu, X. Y., Qin, M., & Li, S. L. (2019). Urban spatial structure and workers’ wage income. World 

Economy, 42(4), 123-148. 

Lundberg, S., & Rose, E. (2000). Parenthood and the earnings of married men and women. Labour 

Economics, 7(6), 689-710. https://doi.org/10.1016/S0927-5371(00)00020-8 

Murphy, K. M., Andrei, S., & Robert, V. (1989). Income distribution, market size, and industrialization. 

The Quarterly Journal of Economics, 104(3), 537-564. https://doi.org/10.2307/2937810 

Pavan, R. (2011). Career choice and wage growth. Journal of Labor Economics, 29(3), 549-587. 

https://doi.org/10.1086/659346 

Waldfogel, J. (1997). The effect of children on women’s wages. American Sociological Review, 62(2), 

209-217, 2657300. https://doi.org/10.2307/2657300 

Waldfogel, J. (1998a). The family gap for young women in the United States and Britain: Can 

maternity leave make a difference? Journal of Labor Economics, 16(3), 505-545. 

https://doi.org/10.1086/209897 

Waldfogel, J. (1998b). Understanding the “family gap” in pay for women with children. The Journal of 

Economic Perspectives, 12(1), 137-156. https://doi.org/10.1257/jep.12.1.137 

Zhang, Y., & Chen, J. Q. (2018). Opportunities and challenges of international labor cooperation under 

the “One Belt and One Road” initiative. International Trade, 2018(5), 37-43, 50. 

Ge, Y. H. (2007). The Effect of Sectoral Choice on the Gender Gap in Wages: 1988-2001. Journal of 

Economics (Quarterly), 6(2), 607-628. 

Yang, Z. W., & Yang, Q. F. (2004). The impact of FDI on the real wage level in China. World 

Economy, 27(12), 41-48. 

 


