Chapter I
Introduction

The shift from a centrally planned subsidized economy toward a market-oriented economy presumes a number of improvements in the functioning of the labor market. In this new phase, one of the many challenges faced by policy makers is to formulate adequate labour market policies. In an attempt to provide some insight into the functioning of the Vietnamese labour market in relation to gender issue, this thesis focuses on understanding gender wage gap in Vietnam.

Research on gender wage differentials is particularly important and timely for Vietnam. Currently, Vietnamese women account for 50.8 % of the population and 50.6 % of the social labor force (World Bank. 2000). However, it is supposed that more market oriented make women become one of vulnerable group that prevent them from getting well paid work. A natural question arise is “Does there exist gender wage gap in Vietnam? Are female employees being under paid?”

In order to answer the main research question, the following sub-questions will be examined:

-         Is there a gender wage differential and how much is it?

-         What factors contribute to the gap?

-         Is there gender discrimination against female in terms of payment in comparison with male wage structure and the pooled one?

-         What is the root of the problem and the government perspective on the problem?

-         What are solutions to improve the problem?

In addressing research question, qualitative and quantitative methods are used, including statistical and descriptive analysis, review of historical trends, and comparative methods. Besides, quantitative method is extensively used.

Data used in the analysis extracts from two sources. First, the data from the specialized survey on wage and employment in 2001 by the collaboration of ILO, CISLSA under MOLISA and GSO is used as the major one for descriptive and econometric analysis. The survey carried out in 505 Vietnamese enterprises of all three economic sectors by ownership (SOEs, domestic private enterprises and foreign invested enterprises), operating in 5 industry groups (manufacturing, mining and quarrying-electricity and water, trade, restaurants-hotels and construction). Second, data from Vietnam Living Standard Survey: 1997-1998 jointly developed by World Bank (WB) and General Statistics Office (GSO) is used to get the overall picture of gender wage employment in the labor market of the economy.

The thesis is structured in to five chapters. Chapter one is the introduction part. Chapter two gives a basic theoretical and methodological framework. Chapter three points out the characteristics of wage employment in Vietnam by historical review and descriptive statistic. Chapter four presents the econometric analysis, which include the regression result of Mincerian wage equation to find out the wage determinants and the Oaxaca as well as Newmark method to decompose the gender wage gap. Finally, chapter five will summarize the main findings; give policy implications as well as suggestions for further study.

 

Chapter ii
Theoretical review and analytical framework

I. Theoretical review

The causes of variation in wages among workers are complex and controversial. However, they can be divided in to two groups.

- Orthodox economist believe that a single wage rate would exist if all workers and jobs were homogeneous; market was perfectly competitive, labour mobility and migration were unimpeded. The wage structure will evidence no variability. The average wage rate will be the only wage rate in the economy. However, the Orthodox interpretation of the labour market and wage determination has long been a subject of controversy and debate. An alternative to the Orthodox theory is the Institutional view of the labour market.

- Institutional economist agree with Neoclassical view that if there is a perfect movement of labor, then wage differentials reflect the characteristics differences between the two groups of workers. However, in fact the presence of many barriers, such as strong unions and collective bargaining rendered orthodox wage theory unrealistic (McConnel & Brue. 1995).

In brief, wage differentials arise because jobs are heterogeneous, workers are heterogeneous, and market is imperfect.

Figure I.1:

Overview of the major contributing factors to wage differentials.

 

(Source: Mc Connel & Brue. 1995)

II. Methodological Framework
1. Selectivity bias - Hay's two stage

Hay’s two-stage procedure is the generalization of Heckman (1979) to overcome the bias problem.

- The first step is to estimate a multinomial logit model to get the probability of an individual i being in one sector j, called Pij. Then it is used to calculate the correction term, named l_ij:

l_{ij = (-1)}J+1                                       _{for j=                                                  (1)}

P_ij is the probability of the individual i in the sector j (Lui. 2002).

There has three sectors J=3                  in which            j = 0 if SOEs   

j = 1 if Domestic Private

j = 2 if Foreign Invested.

- In the second stage, this correction term is then included in the wages equation as an additional regressor to correct the situation of sample selection bias.

2. Conventional Mincerian equation of wage determination

The Mincerian wage is as follow:

_ij= a_ij + b_ijij                                        (2)         

 i: sex (male or female)

 j: sectors (SOEs, domestic private, foreign invested)

Where: W is vector of wage; X is a matrix of observations on explanatory variables, such as experiences, formal education…and other dummies variables on sex, position, sector, industry, region…

3. Econometric decomposition of gender wage gap
3.1. Oaxaca decomposition

The Oaxaca decomposition is:

_mj - _fj = (a_mj - a_fj) + b_mj(_mj - _fj) + _fj (b_mj - b_fj)                             (3)

The first term on the right hand side is the part of the wage gap that is attributable to differences in environment, the second term is different in factor- endowments; the third term is differences in returns to factor endowments.

Discrimination is equal to the combination of return gap and environment gap, after controlling for worker characteristics. The great advantage of equation (3) is that it can be used to reflect both Institutional and Neoclassical views. The Neoclassical view is reflected by the characteristic gap and the Institutional view is reflected by discrimination.

However, Oaxaca approach is subject to the index number problem. The decomposition results can be differed depending on which reference group or which non-discrimination wage structure is used.

3.2. Newmark decomposition

Newmark suppose that that under discrimination males are over paid but females are under paid. To avoid the index number problem, he proposes another way to decompose the gap in each sector-noted j:

_mj - _fj = (a_mj - a_fj) + b_j(_mj - _fj) + _mj (b_mj - b_j) + _fj (b_j - b_fj)    (4)

Where: bj is the non-discriminatory wage structure, estimated by using the weighted average of pooled wage structure.

The first term on the right hand side is environment gap, second term is the characteristic gap, the third and the forth capture the difference between the actual and pooled returns to characteristics of men and women respectively.

While Newmark decomposition is attractive, it is not clear that the pooled coefficient is a good estimator of the non-discriminatory wage structure.

III. Literature review

The importance of this issue is shown in various studies. Some remarkable ones have been done for United State (Francine D.Blau and Lawrence M.Kahn. 2000); Urban United State (Ronald Oaxaca. 1999); Sweeden (Mats Johansson, Katarina Katz and Hakan Nyman. 2000); Urban Bolivia (Trine Monsted. 2000); Africa (J Knight and R.Sabot. 1998); Latin America including Argentina, Brazil and Costa Rica (Luz. A. Saavedra. 2000); Vietnam (Liu. 2000). From these works, the gender wage gap is different from one country to another. These researches show that in all countries, male wage are higher than female wage. Education is the very important wage determinant. In developed countries such as United State, Sweden and so on unequal pay for equal work does not account for very much in male female differentials. Rather, it is the concentration of women in lower paying jobs that produce such large differentials. Different from developed countries, the difference in worker’s characteristics in developing countries is quite large. Even in Africa, the difference in worker’s characteristics between male and female contribute up to 83% to the total gender wage gap

 

Chapter III
Characteristics of gender wage employment in Vietnam

I. The social economic circumstance and gender issue in the labor market

The social role and status of Vietnamese women have been improved remarkably throughout the country’s development process. In the past, Vietnam has a long history of federal regime. Women were in extremely low status. Since 1945, under socialism, the women’s status has improved. Women have equal rights with men in all respects, such as in access to education, health services and employment and also being protected by the legal constitution. Thus women attained a high level of literacy and have been shoulder to shoulder with man in all fronts (Hung & Anh. 2000).

Renovation toward a market-oriented economy since 1986 has presumed a number of improvements in the functioning of the labor market toward a more active economic environment. Simultaneously, many women worked in SOEs previously being sacked and now changed to the informal sectors doing small business. Job situation becomes tenser as industrialization and modernization are accelerated, as this process will require higher skill levels and qualifications thus the situation may sharpened for female labourers.

II. Wage employment and its comparisons across various fields
1. General picture of wage employment in Vietnam

Vietnam’s current population is recorded at 78.5million as of the time of the 2000 World Development Indicator. The population in working ages is approximately 46 million people. The rate of women participation in Vietnam is 79.5 percent, only slightly lower than that of men at 83.4 percent. Meanwhile, the rate of women labour participation in Indonesia, the Philippines and South Korea are only at levels close to 50% (Bales. 2000).

Among the total social labor force, self-employed in farm and non-farm account for 80%, wage employment only account for about 20%. The wage employment differs between male and female. Only about 15% of women have wage jobs compared to about 25% of men.

 

More women work in state sector than men. In contrast, the proportion of women workers in the domestic private sector (34%) is much lower than men (47%). In foreign invested sector, the rates between the two sexes are nearly equal.

2. Gender wage comparison across various fields

The total hourly wage of female is around 15% lower than male. About the components of total wage, the higher wage of male compared to female comes from both higher basic wage by 16% and higher allowances by 13%.

- Wage comparison across economic sectors: Female is paid less than males in all three sectors. Hourly wage in SOEs is lowest and foreign invested sector is highest to both sexes. The gap of female wage among sectors is lower than male.

- Wage comparison across positions: Employees in high positions get high wage. Average wage of women is lower than that of men in all positions. The gap is become narrower in higher positions.

- Wage comparison across education levels: People with no education get the lowest total wage in both sexes. In general, especially for male, hourly total wage increase with higher education levels.

- Wage comparison across age groups: At early age less than 25 years old, the started wage of female is slightly lower than male. From 25 to 30, women have to do child bearing and rearing, thus they have discontinuous time at work, in this period their wage increase at low level. Female wage level started to increase from around 30 years old, this time their children become elder.

- Wage comparison across kinds of contract terms: More stable contract brings higher wage for female. But for male, the highest wage level is for seasonal or specific task less than one year.

- Wage comparison across industries: Female wage is lower than male in all industries. The lowest gap is in tourist, restaurant and hotels and the highest gap is in mining, quarrying, electric and water.

- Wage comparison across regions: The average wage level of the North and the Middle are nearly equal, but in the South is much higher. Female wage is lower than male in all regions.

 

Chapter IV
Econometric Analysis of gender Wage gap in Vietnam

I. Description of variables, optimum regression and results
1. Description of variables

The dependent variable is the natural log of hourly wage.

The definition of individual wage is included the basic wage (salary) and other remuneration at the time of survey. Basic wage is calculated by the multiply of the wage level and its coefficient. Other remuneration include bonus, lunch payment, overtime payment, allowances and others.

The set of independent variables are:

- Education: This is the previous education of the worker. This category is classified into 5 levels: no education, primary education, lower – secondary education, upper – secondary education, technical vocational training, and university or higher. No education is treated as the benchmark. Four dummy variables are used for the remaining levels. It is expected that higher levels of education will have bigger returns.

- General experience: General experience is potential years of working experience of the worker. It is defined as (current age – age when respondent stopped studying and began working). As stated by the human capital theory, workers with higher general experience are likely to receive higher wage rates. Furthermore, human capital theory also suggests that wages should generally not be constant after leaving school but should follow a parabolic shape, peaking somewhere in midlife so that we need to add experience in quadratic to the regression function.

- Specific experience: It is experience obtained in the current main occupation. This category is measured in months. Specific experience squared is also included in the regression function for the same reason as the general experience squared.

- Started year of working: This is the age when one first entering the enterprise. This variable is expected to have different impact depends on each sex in each sector.

- Gender: It is the sex of the worker. Because wages may be different between male and female workers as well as between sectors, this variable is taken into consideration to examine whether there are gender wage discriminations between the two sectors. A dummy variable is defined for gender that takes the value one if the worker is male and otherwise.

In addition, some variables reflecting job characteristics such as industry, position, contract term, monthly working hours, sectors are also taken into the regression function. The inclusion of these variables is based on the argument that works are different and hence are compensations. The region variable is used to reflect the wage differential between the North, the Middle and the South.

2. Explanation for the optimum regression result

The equation (2) introduced in the first chapter is used to decompose the wage differentials between male and female. The optimum regression functions with and without selectivity correction have been achieved by utilizing a “top – down procedure” to drop out insignificant variables (Gujaratti, 1995, chapter 8). Having these optimum regression functions at hand, we have no worry about insignificant coefficients as they have been dropped out at 10 per cent level of significance.

3. Regression result

Table IV.1 to IV.4 report the Mincerian wage equation of pooled sample and different sector (SOE, domestic private and foreign invested) regressions.

Other things constant, education is significant to both sexes. Returns to formal education of female is higher than male. Especially in SOEs one more level of education increases female worker’s wage by up to 25%. Workers with more experiences (both specific and potential ones) are likely to get higher wage.

The age of starting the current work has negative coefficient. It is highly shown in female wage equations, the older they start apply for a job the lower wage they are likely to receive. The same case also happened to male workers in SOEs and foreign invested sectors however in domestic private sectors, it is vice versa.

The suitability of trained professional and current work is significant to both men and women in pooled sample regression. Employees with more suitable current work to trained occupation can get higher wage than others. However, examine in to difference sectors the suitability is more important to female labourers and male labourers working in SOEs and foreign invested sectors. In domestic private sectors, this variable is not significant in wage determinant.

The position of employees strongly affects the total wage, higher position results in higher wage. Wage gap among positions of male is higher than female. The range of variation among positions in foreign invested sector is lower than the SOEs and domestic private sector.

The coefficient of monthly working hours in SOEs is positive. However, the coefficients of domestic private and foreign invested enterprises are positive, the higher monthly working hours the workers do the less hourly wage they can receive.

Contract term has different affect in each sector. In SOEs and domestic private enterprises workers with more stable contract terms can get higher wage but in foreign invested enterprises, this variable are not significant in wage determination.

The wage variation among industries is different in each sector. Wages also differ among regions. Labourer in the South can get significantly higher wage than the North and the Middle.

II. Econometric decomposition of the gender wage gap

Oaxaca and Newmark models with selectivity bias and without selectivity bias were initially calculated. However, as denoted in Mincerian regression the lack of detailed worker’s information make selectivity bias correction insignificant. Decomposition result without selectivity bias is more meaningful than with selectivity bias. For that reason, I mainly use Oaxaca and Newmark decomposition results without selectivity bias to explain.

1. Decomposition of gender wage gap in SOEs

1.1. Oaxaca method

Male employees have richer characteristic than female. These gender characteristic differences make the total characteristic gap accounts for around 35% of the total wage gap. These characteristics included more potential and specific experiences, higher formal education, higher positions, and larger share of men working in the region with relatively high wage than women.

Return gap is positive at 0.248 and accounts for up to 165% the total gap. In SOE, on average return to characteristic of men is appreciated higher than woman. The large different in returns to characteristic can be explained mainly by different in returns to individual characteristics. These include higher return to: specific experience, the rate of increase in return to experience, started working year (each year started working earlier brings more benefit to men than to women), and return to high position. It is interesting to find that even female are more likely to get jobs which is suitable to trained occupation but return to the suitability of male is higher than female.

Third, negative environment gap is –0.151, which offset characteristic and returns gap by large, –101%. In SOEs, economic environment at work to women is more favorable than men after controlling for all other variables.

Discrimination is combination of return gap and environment gap after controlling for the same characteristic, account for nearly 64% of the total gap or at 0.097 in absolute value. There is still exist wage discrimination against women in SOE sector, for employees with equal characteristic men can get higher hourly total wage than women.

1.2. Newmark method

The total gender wage gap in Newmark method is decomposed in to four components: characteristic gap, return gap come from male advantage, return gap come from female disadvantage and environment gap.

Characteristic gap is positive and equal to 43% of the total gap in favor of male employees.

Return gap is the sum of male advantage and female disadvantage compared to the value of pooled sample. Return to male advantage is positive at 0.075, accounting for 50%, and return to characteristic of female disadvantage is 0.161, accounting for 107%, the total return gap is 0.236 accounting for 157%. It can bee seen that return to characteristic of male is overestimated and return to characteristic of female is underestimated at relatively high level compared to the non-discriminatory structure.

Total environment gap is the same to Oaxaca decomposition. However, the components of environment gap comprise the male’s favor and female out of favor. The negative value of environment in favor of male at -0.042 or equals -28% and the negative value of female out of favor at -0.110 equals –73% mean that environment is good for female but it is not good for male.

Discrimination against female (compared to nondiscrimination wage structure of pooled sample) is 0.051 in absolute value or 34% by relative. The result shows that there is wage discrimination against female employees in SOEs.

On the other hand, the sum of male advantage in return to characteristics and favor environment is 0.033 (equals 0.075 plus –0.042) or 22%. Opposite to female, male employees get higher pay when compared to nondiscriminatory wage structure from the pooled sample.

The level of wage discrimination against female in the Oaxaca decomposition is higher than the Newmark. This is because in the Oaxaca decomposition male wage structure is used as a base system to compare. But the fact shown in Newmark decomposition is that male wage structure is not nondiscriminatory one.

2. Decomposition of gender wage gap in domestic private enterprises
2.1. Oaxaca method

The characteristics gap between male and female account for -69% of the total wage gap. The characteristic gap in domestic private sector is higher than SOEs. This gap is the contribution of some major variables as: more potential and specific experience, a little higher education level, higher positions, higher rate of men working for the South.

Return gap is negative by large –0.191, which accounts for -184%. Returns to characteristic of men in domestic private sector are appreciated lower than women. More specifically returns to potential experience, education of female is higher than male. But others variables as specific experience, started year of working, the suitability of current and trained occupation and contract term, monthly working hours men has lower return to characteristic than women causing negative return gap. The negative value offset all positive one. On average, return to characteristics of male is lower than female. This result is opposite to SOEs.

Meanwhile the men are underestimated compared to women in returns to characteristics. The high positive environment gap means that in domestic private enterprises, economic environment at work to men is more favorable than to women.

Discrimination is equal to 31% or 0.032 in log wage difference. The result is lower than SOEs sector. Taking male wage structure as benchmark, it can be said that women in domestic private sector get more favorable treatment than SOEs.

2.2. Newmark method

Characteristic gap is positive and equal to 0.108, account for nearly 103% of the total gap in favor of male. Both Oaxaca and Newmark decomposition show that male workers in domestic private enterprises significantly have richer characteristics than female.

Total return gap is –0.227 in value account for -218% of the total wage gap. Return gap to male advantage is negative, account for -57%. Return gap to female disadvantage is positive account for up to -161%. Opposite to SOEs, return to worker’s characteristic of male is lower than female.

In contrast to return to characteristics, working environment is favor more to male workers than their counterpart. Components of environment gap to both sexes are positive at 0.055 and 0.168 means that economic environment not only in favor of male workers but also being out of favor to female workers.

Discrimination against female employees compared to the pooled sample equals to zero. There is no wage discrimination against female workers in domestic private enterprises in comparison to the pooled sample.

The sum of male advantage in return to characteristics and favor environment (compared to nondiscriminatory wage structure) is -0.004 (equals -0.059 plus 0.055) or -4%. Other things being equal, male employees get a little lower paid than the wage average level.

3. Decomposition of gender wage gap in foreign invested enterprises
3.1. Oaxaca method

The characteristic gap between male and female is 0.066 accounting for 36% of the total gap. The gap is nearly equals to SOEs, but lower than domestic private sector.

Return gap is negative and account for large share, -0.480 or -265%. Men are underestimated in their returns to characteristic in comparison to women, and at quite high level.

Third, environment gap offset all characteristic gap, account for 328%. Economic environment at work to men is more favorable than to women.

In this case, discrimination is equal to 63% or 0.115 in value. The rate of discrimination in foreign invested enterprises is equal to SOEs in relative value but higher in absolute value.

3.2. Newmark method

Characteristic gap is positive and equals to 20% of the total gap. Compared to the previous two sectors, female employees has less low characteristic than male.

Total return gap is –0.451 or accounting for -249%. Return to characteristics of male advantage is negative at –0.299 or –165% and return to characteristics of female disadvantage is –0.152 or –84%. This case is the same to domestic private enterprises. Compared to non-discriminatory wage structure, return to characteristic of male is underestimated; and return to characteristic of female is overestimated.

Different from return to characteristics, foreign invested working environment is the very advantage of male and disadvantage of female compared to the other two sectors. The positive environment gap of 0.356 and 0.240 for male advantage and female disadvantage respectively offset all return gap.

Discrimination against female employees is equal 0.088 or 48%. Wage discrimination against female employees in foreign invested enterprises is highest compared to SOEs and foreign invested sectors.

The sum of male advantage in return to characteristics and favor environment compared to the pooled sample wage structure of both sexes is 0.057 or 31%. This can be explained that in domestic enterprises thanks to the advantageous working environment, male employees get higher paid than female and also higher paid than the other two sectors.

Table IV.1:     Summary Result of Oaxaca and Newmark Decompositions
(Optimum regression)

 

	State Owned Enterprises		Domestic Private Enterprises		Foreign Invested Enterprises
	Gap value	Percent	Gap value	Percent	Gap value	Percent
I. Oaxaca (Male wage structure as base) 1. Characteristic gap 2. Return gap 3. environment gap     Discrimination = return gap + environment gap (Against female compared to male wage structure)	0.053 0.248 -0.151   0.097	35% 165% -101%   64%	0.072 -0.191 0.223   0.032	69% -184% 215%   31%	0.066 -0.480 0.595   0.115	36% -265% 328%   63%
II. newmark (Weighted wage structure) 1. Characteristic gap 2. Male advantage      Return gap     Environment gap 3. Female disadvantage     Return gap     Environment gap   Discrimination = return gap + environment gap  (against female compared to the pooled sample)	0.065 0.033 0.075 -0.042 0.051 0.161 -0.110   0.051	43% 22% 50% -28% 34% 107% -73%   34%	0.108 -0.004 -0.059 0.055 0.000 -0.168 0.168   0.000	103% -4% -57% 53% 0% -161% 161%   0%	0.037 0.067 -0.299 0.356 0.088 -0.152 0.240   0.088	20% 31% -165% 196% 48% -84% 132%   48%
Total	0.150	100%	0.104	100%	0.181	100%

Source: Author’s estimate from the survey data.

Discrimination rate is smallest in domestic private sector, however this sector also has low average wage to both men and women. Foreign invested sector has highest discrimination rate and also going with highest wage level.  In SOEs, the discrimination rate is also high; only a little lower than the highest one, but the average wage is relatively low.  This phenomenon can be explained by the current situation of domestic private enterprises in the economy.

- The private sector mainly includes small and medium enterprises. Thus the situation of capital poor domestic private enterprises have been much more likely to focus on labor-intensive activities, taking advantage of Vietnam’s greatest comparative advantage than state or foreign invested firms. As a result of a low labor’s productivity the worker’s wage is low. On the other hand, domestic private sector operates under high competition pressure in the market. There is incentive for the employers to reward worker’s wage according to productivity related. The gender wage gap and level of discrimination are low.

- Vietnamese state and foreign invested sectors are both primarily focused on capital-intensive projects and frequently aimed at domestic markets at present. According to national industrial strategy, the state favored some kind of enterprises such as: transportation, mining, energy supply, computer science, information technology, and heavy industries. (Still man, 2000). In short run, the way of resource allocation in some state favored industries is away from developing country’s comparative advantage, which rest on the abundance of unskilled labor. In addition, women are usually employed in a limited number of industrial sectors. Capital intensive and heavy industries are more likely to employ male workers, whereas labor intensive, light industries are more likely to employ female workers. The labor market are segmented by gender. More over, the wage fund in SOEs is still regulated by the state, is a small pie of cake. The criteria for work assessment are vague and not yet related to productivity. To some extent, to obtain higher wage, employees try to impose other non-productivity related values. These are discrimination attitude, which include gender discrimination. As a result the average wage in SOEs is low but the gender wage gap and the discrimination rate is high.

- Foreign invested sector is a sector with active economic operation and most market orientation. The emergence of enterprises in this sector is derived from the demand side to exploit the nation’s comparative advantage or take advantage of the government’s priorities. This sector is capital intensive and the use of capital is efficient. More over, employees in foreign invested enterprises usually work with great intensity, the wage system is productivity related. The capital intensive and most market-oriented operation of foreign invested sector bring higher wage to all employees in comparison with the other two sectors but its own characteristics also create highest gender wage gap in this sector.