INTRODUCTION
Research goals and objectives
tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
This research aims to analyze the extent to which education and family background influence wage correlation Key factors include employment characteristics such as age, marital status, ethnicity, education level, work experience, and urban or regional location Additionally, family background elements like parental education and the number of siblings will be examined to provide a comprehensive understanding of their impact on wage disparities.
This thesis develops a model to analyze the relationship between wages, employment, and family background characteristics The findings indicate that education has a direct influence on wages and enhances the impact of family background on individual outcomes Based on these results, the study offers policy recommendations.
Research questions o • • • • • • o o o o o o o o 0 3
This research aims to address key questions regarding the impact of employment and family background characteristics on wages Specifically, it investigates how various employment traits influence wage levels and examines the role of family background in shaping wage outcomes.
The thesis seeks to address two critical questions: first, whether the level of educational attainment significantly influences wage levels; and second, how family background impacts an employee's earnings.
LITERATURE REVIEW
Previous empirical studies 0 • • • • • • • • • • • • • • • • 6
Worker skills significantly influence wage levels, as supported by numerous empirical studies Buchinsky and Leslie (2009) developed a dynamic model to analyze individuals' educational investments, enabling the exploration of various forecasting strategies for future wage distributions Their research utilized data from male individuals aged 14 to 65, who were either employed or in school, sourced from the March Current Population Surveys in the U.S from 1964 to 2004 The study aims to integrate a realistic forecasting model into the analysis of educational decisions, focusing on the uncertainty of aggregate parameters, such as future wage distributions, while maintaining certainty over individual parameters like the agent's utility function This research introduces a significant innovation by incorporating parameter uncertainty into the decision-making process.
In our framework, a risk-averse individual considers not only the uncertainty of future wage outcomes based on education and experience but also the variability in the underlying distributions, including the uncertainty in the parameters of the estimated conditional wage distributions.
The VAR-Gibbs model is selected for regression analysis due to its effectiveness in addressing two key issues Firstly, it examines how varying levels of risk aversion influence educational decisions, revealing that higher risk aversion correlates with reduced educational investment, as education is perceived as a riskier investment compared to experience during the study period Secondly, the research highlights the significance of financial resources in higher education decisions by assessing how changes in initial wealth affect educational choices.
Distinct forecasting methods reveal significant variations in school attendance rates, average education levels, and the duration required to obtain a college degree The study highlights that greater initial wealth correlates with a faster accumulation of education.
Hamilton et al (2000) explored the impact of worker and employer heterogeneity on output distribution, focusing on the relationship between profits and wages within firms Unlike Sattinger (1993), their study considers a labor market where workers and firms are vertically differentiated, with a common level of general human capital but varying individual skills This approach allows for the inclusion of unique worker characteristics, highlighting that equally educated individuals may not be equally suitable for firms Additionally, the research assumes a diverse workforce suited for different types of jobs, alongside firms with varying job requirements, leading to the development of a non-hierarchical assignment model that complements existing hierarchical models.
Firms that struggle to identify the skill types of individual workers may find that those receiving less training often earn higher net wages This occurs because firms do not differentiate between workers based on their skill types, leading to better-matched workers receiving higher wages despite having similar levels of general human capital and productivity As the number of firms increases, equilibrium wages rise due to competition for better-matched workers In a scenario where the number of firms becomes very large, wages approach the competitive level of general human capital, while profits diminish to zero The competitive labor market model represents the limit of the spatial job assignment model Furthermore, an increase in the common level of general human capital boosts gross productivity and reduces training costs for each worker, resulting in higher net wages This trend is supported by numerous empirical studies, although profits may decline as firms lose some monopsony power, which offsets productivity gains.
When firms are fully aware of the quality of individual job matches prior to hiring, they can tailor their offers based on the skill types of workers This allows employers to focus on the total wage and training costs, while employees consider their net wages after training expenses The responsibility for training costs becomes a part of the bargaining process, making it less significant who pays these costs Consequently, workers who receive more training tend to earn higher net wages, not due to increased productivity, but because their training costs at alternative firms are lower This dynamic gives poorly matched workers a stronger bargaining position compared to well-matched workers, as they have better outside alternatives.
Wage and schooling are also interesting issues for economists besides the U.S
Meghir (2005) assessed the impact of educational reform on final educational attainment and earnings using data from the Swedish Level of Living Survey, national education registers, and tax records spanning from 1985 to 1996 The study employed various methodologies and divided the sample into two cohorts, with the first cohort consisting of 10,309 individuals born in 1948, including 5,235 men and 5,074 women.
The study analyzed 9,007 individuals born in 1953, comprising 4,525 men and 4,482 women, with earnings data collected from 1985 to 1996 A logit model was employed to assess the impact of educational attainment, measured through two binary outcomes related to the new compulsory education level Overall, educational attainment increased by 0.298 years, with significant effects observed Individuals with lower ability primarily moved to the new compulsory level, while those with higher ability not only reached this level but also exceeded it The reform's overall impact on earnings was modest at 1.42 percent, significant only at the 10 percent level, yet it masked considerable variation among different groups Notably, individuals with unskilled fathers experienced a significant earnings increase of 3.4 percent due to the reform.
Some older studies have attempted to investigate impact of schooling on wage
Harmon and Walker (1995) analyzed the impact of education on wages and incomes using data from the U.K Family Expenditure Survey (FES) Their study focused on a sample of 34,336 employed males aged 18-64, drawn from nine consecutive annual FES cross sections between 1978 and 1986 To address the endogenous issues, they employed an instrumental variable approach, leveraging exogenous changes in the educational distribution resulting from the increases in the minimum school-leaving age in the United Kingdom.
- - - - - - - of the working-age individuals in our data) to provide instruments for schooling By choosing this way, this research has a little bit different to previous studies such as
A study by Grist and Krueger (1991) utilized natural variations in data influenced by external factors affecting schooling decisions Their results corroborate the findings of Ashenfelter and Krueger (1994), indicating that an additional year of schooling can lead to over a 15% increase in wages Additionally, the research revealed that residents in larger cities tend to earn higher wages compared to those in other regions.
Social scientists across various disciplines have extensively studied the link between family background and adult economic and social status Numerous studies have investigated how family factors, including parental income, education, and the number of siblings, influence wages.
In 2007, researchers introduced a novel method to assess the brother correlation in earnings across two distinct time periods They utilized two separate cohorts of men from the National Longitudinal Surveys (NLS) to conduct their analysis.
The NLS data sets reveal significant changes in sibling correlations for young men born between 1957 and 1965 compared to those born between 1944 and 1952 Notably, the correlation in annual earnings has increased from 0.26 to 0.45, indicating a statistically significant rise This trend suggests that nearly half of the variance in earnings and wages for recent cohorts can be attributed to family and community influences, highlighting a decline in intergenerational mobility Interestingly, there has been no change in the correlation of years of schooling between these cohorts, implying that the stronger association between family influences and earnings does not stem from educational attainment.
The returns to education have significantly increased between the two cohorts; however, this factor accounts for only a minor part of the overall rise in the correlation of earnings among brothers.
OVERVIEW OF VIETNAM LABOUR MARKET
Employment by age and gender 0 0 0 0 0 0 0 ° 0 0
The employment-to-population ratio is a crucial indicator of economic activity in Vietnam, reflecting the proportion of employed individuals within relevant population groups This ratio has shown a steady upward trend, increasing from 62.73% in 2006 to 69.18% in 2008 Additionally, the youth population in Vietnam is approximately evenly divided by sex, highlighting the demographic structure of the workforce.
In 2008, the labor force participation rate for women was 50.33%, slightly surpassing the men's rate of 49.67% Despite an overall increase in ratios, the quantity of women participating in the labor force decreased, indicating a concerning trend Women maintained a larger proportion in the labor force compared to men, which contrasts with the findings from the Population and Housing surveys conducted between 1997 and 2008.
Surveys indicate a decline in labor force participation across all age and sex groups, with women representing a smaller percentage than men Last year's employment-to-population ratio was 53.3%, with men at 52% and women at 48% The VHLSS data shows a decrease in observations from 39,071 in 2006 to 35,154 in 2008, despite a slight increase in the labor force, which contributed to a higher workforce ratio The VHLSS primarily focuses on rural areas, where female employment often constitutes a larger share, as many women engage in agricultural activities and serve as the primary earners for their families, often working as unpaid family members or self-employed individuals, significantly impacting the sample size.
The Population and Housing survey encompasses the entire population of a country, providing comprehensive results rather than relying on a sample Notably, the most significant increase in the overall labor force is observed among the youth demographic.
In 2006 and 2008, young people aged 15 to 24 represented 27.99% and 26.45% of the total workforce, respectively A significant factor contributing to this decline is the tendency of youth to extend their education, resulting in decreased workforce participation Notably, male participation rates are higher than those of females, indicating that men tend to enter the workforce earlier Between 2006 and 2008, the youth labor force decreased by over 400, or 1.54% Despite this decline, the gap in labor force participation rates between men and women is smaller compared to many other countries Overall, labor force participation rates for both genders fell from 1997 to 2007, with the most significant decrease observed among young women aged 15 to 24.
Vietnam's employment-to-population ratio is relatively high compared to regional standards, although it falls short of East Asia's figures As of 2009, women represented 48% of the workforce, while men accounted for 52%, reflecting a consistent trend over the past 30 years (Population and Housing survey 2009).
Compared to the labor forces in East Asia, Southeast Asia, and the Pacific, Vietnam's labor force has seen a gradual decline; however, the changes in Vietnam are occurring at a faster pace While East Asia has experienced the largest drop in male workforce participation, Vietnam has shown a contrasting trend Similarly, Southeast Asia and the Pacific have witnessed a significant reduction in female labor force participation Notably, while the percentage of men in the workforce is comparable to other countries in the region, Vietnamese women represent a larger proportion For instance, women's participation in the workforce in Vietnam stands at 79.5%, significantly higher than the approximately 50% seen in the Philippines, Indonesia, and Korea.
Figure 3.2: Workforce classified by age-bands (person) i 2,500
According to the VHLSS 2008 data, labor force participation among both women and men shows a gradual decline starting from ages 15-19, with notable differences emerging in participation rates The gap in participation between genders peaks at ages 20-24 before gradually decreasing By ages 40-44, both male and female participation rates decline further, with the smallest gender difference observed in this age group Interestingly, women continue to engage in work beyond age 55, often contributing more than men, particularly in rural areas where they serve as primary breadwinners in agricultural activities and as unpaid family workers The VHLSS focuses on the living standards of rural residents, highlighting the significant role of women in the labor force across all age groups, especially in agriculture, where they do not typically retire from their roles.
Table 3.2: Labour force divided by sex, regions(%) rããã ããã2aa6ããã ããã2ao8ããã1
' ' i Regions ã"ATC~exe~ ããMale -Fe~a1e _ ãA:ifsexes -M~1e ã:pe~a!eã -~
-N~rth E-asi -T5:sã4 ãã::20Ấ:20::5s -ã5a-.-42- -isã.-~faãã -ã-sajj- -49j~ 1ã1
-N~-rth-westã - -ãã-s-:7"3 -49:"35 -ããs(Y_-6ã5- -6.-"1"6 -49-:8T -sãa:T9
-N~-rth-ãce~1~a:rc0a-5T -To:2r -49:ã55 -ã5o-.-4ã5- -9.-6"6"- -49-."73-o;~- -ã-ã-sãa:-27 solithã - -ããs:7ã4 -ããsãa:ãiãsãã -49-.-82- -ããs-.-82 -ããsãa:ãs6ão;~- -49:ã44
Central Coast ããcefii~a:rt~rgili~n:a~ - -ã6:73 -4-9:-78 -5a::20:é20:: -6-:94 ãã-ã-s-T.-s?ão;~- -48:-43
- -ããã -ããã -ããã-ãã -ããã -~ -4
Mekong-~i"~e~-cieita:ãã - ãã2o-Đsã - 49:2"3" ããs-o:??" _ ããia-."74ã - 49-.-:290/~ - ãsãa:-7To/~ - ãu;ban _ -23".-74 4&-:6ã5ã ããsT3ãsãã - ã23j~6 - -48-:6T ãs-1:"39
Rli~arãã - -7"6:26ã - 49.-?ãaãã- ããs-o:3ãaãã - 76-:14ã - ããsaã:aK ã49-:94ã -j i
Source: VHLSS 2006-2008 tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
Table 3.2 indicates that the North West region has the lowest labor force participation, while the Mekong River Delta shows a contrasting trend Interestingly, the South East and Red River Delta regions have lower participation rates despite their development Although major cities like Hanoi, Ho Chi Minh City, and Haiphong have populations estimated at 11 to 12 million, urban areas in Vietnam represent less than 24% of the total population aged 15 and over.
Between 2006 and 2008, the majority of the population remained concentrated in rural areas However, this trend has been gradually shifting as urban areas expand both in population and geographic size, leading to an increasing number of individuals relocating from the countryside to cities Consequently, since 1997, the proportion of the urban population within the labor force has shown a consistent upward trend.
We also find that women take part in workforce higher both in rural and urban areas.
Employment with education and training 0 0 0
Vietnamese youth exhibit high literacy rates, exceeding 90%, with consistent levels maintained over the past 25 years despite reductions in state education subsidies following the Doi Moi reforms A significant portion of the workforce is literate, with 86.3% of surveyed individuals in 2006 demonstrating reading and writing skills The literacy rate has shown a steady increase, rising from 93.2% in 2008 to 93.5% in subsequent years.
The 2009 Population and Housing survey indicates that the implementation of compulsory primary education and policies aimed at reducing illiteracy have effectively increased the literacy rate Notably, the workforce's literacy rate saw a marginal rise from 94.7% in 2006 to 94.8% in 2008, as detailed in Table 3.3.
Table 3.3 highlights significant disparities in educational attainment across different regions The Mekong River Delta, known for its leadership in rice exports, has the highest number of individuals without any certificates and those who have only completed primary education In contrast, the Red River Delta ranks first in the number of high school graduates, encompassing both lower and upper secondary education.
Table 3.3: Labour force classified by education (%)
Region No Primary Lower Upper College University Master certificate secondary school and higher
The South East region boasts the highest university graduation rate, while the South Central Coast excels in master's degree attainment In urban areas, individuals with upper secondary school diplomas represent the largest demographic, whereas in rural regions, primary school graduates dominate This disparity indicates that a significant portion of the population lacks education beyond upper secondary, which negatively impacts the quality of the workforce in rural areas, where 76.26% of the population resides.
The total percentage of individuals with a college degree or higher is nearly 6%, with 873 people achieving a university-level education, representing only 4.4% These figures indicate a concerning trend for our workforce, which, while abundant in quantity, suffers from a shortage of highly skilled labor, resulting in low productivity across various industries.
According to Figure 3.3, a significant portion of the working-age population consists of unskilled workers, with over 95% of the labor force lacking vocational training, excluding those with college degrees This situation presents challenges for the youth in finding employment The Vietnam Youth Development Strategy 2010 emphasizes the need to create more job opportunities Millions of young people in urban areas are actively seeking work, with many spending approximately 75% of their time employed throughout the year The increase in the employment ratio from 2006 to 2008 can be attributed to the global crisis, which negatively impacted household earnings, leading many students to leave school and enter the workforce without proper training.
A recent survey conducted by VCCI and ILO, which included insights from FDI companies and young enterprise associations, highlights key observations regarding the current landscape of human resources and personnel management.
This year, a management club survey revealed that up to 50% of enterprises frequently retrain new recruits due to a mismatch between employment skills and company requirements The survey included companies from various sectors, such as printing, aquatic processing, textiles, electronics, and tourism There is a significant shortage of highly skilled workers, particularly in manufacturing, which accounts for 67% of the deficit Consequently, a large portion of jobs are low-skilled and manual, with manual labor comprising 67.31% of total employment in 2008, according to the VHLSS 2008 report.
Between 2006 and 2008, the percentage of employed workers receiving vocational training significantly declined, from 4.26% in 2006 to just 1.81% in 2008 This drop may be attributed to the global crisis, which prompted many students to leave school and enter the labor market Consequently, an influx of untrained workers has resulted in a workforce primarily suited for manual jobs, such as laborers and agricultural workers.
Manual labor constitutes a significant portion of total occupations; however, these jobs typically have short durations, often lasting only a week or a month Additionally, many of these positions are seasonal and offer low wages.
Status-in-employment data categorize employed individuals into four groups: wage or salaried workers, self-employed individuals (including employers who hire others), and unpaid family workers, also known as "contributing family workers." The latter group works without pay in family businesses or farms, supporting the business owner in generating profits Typically, these unpaid workers are spouses, children, or extended family members such as grandparents, cousins, and aunts or uncles.
According to VHLSS data, employment status can be categorized into four groups: wage and salaried workers, employers, own-account workers, and unpaid family workers In 2006, the number of employed individuals identified in these categories was notably low.
In recent years, the percentage of employers has decreased slightly from 0.43% to 0.34% (Appendix A5) Wages and salaries workers have consistently represented the largest segment of total employment, ranging from 44.72% to 45.33% Unpaid family workers have also seen a minor increase, moving from 44.11% to 44.38% Conversely, the proportion of self-employed workers has declined from 10.83% to 9.85% The combined figures for self-employed and unpaid family workers, categorized as vulnerable employment, decreased from 54.94% to 54.23% in 2008 Although the rates of self-employment and unpaid family work have gradually diminished, they still account for a significant portion of the workforce Notably, over half of all employed women are engaged as unpaid workers in family businesses, in contrast to their male counterparts.
Table 3.4: Distributions of status in employment by sex
waies an-cr -~ -3-s-~s4 ; _ 6-ci-6+ -39:Trj_ _ 6_o~-s9- salaries workers (%)
According to the VHLSS data from 2006-2008, Table 3.4 reveals a striking statistic: over half of all employed women in Vietnam did not receive any earnings for their work (for more details, refer to Appendix table AS).
In Vietnam, women experience greater vulnerability than men, as 78% of their total employment is categorized as own-account and unpaid work.
family worker, compared with a still very high 75 per cent among men (ILO, 2007)
Unemployment
Age-bands Number of worker Percent
According to the VHLSS 2008 report, there are 4,269 individuals aged 15 to under 60 who are unemployed, representing 21.29% of the total labor force This significant ratio is influenced by the inclusion of rural residents, where the unemployment rate is the highest Additionally, the report highlights that the age group of 15-29 faces considerable challenges in employment.
In 2009, the unemployment rate reached 3.05% of the total labor force, with 1,504,888 individuals not employed, marking the highest ratio from 1997 to 2009 Among the unemployed, those aged 15-19 represented the largest group, accounting for nearly 75% of the total Out of 4,269 unemployed individuals, 75 were unable to find work, reflecting a rate of approximately 1.76%.
The report highlights that individuals aged 15-29 represent nearly 50% of the unemployment rate It also reveals significant disparities in technical training attainment between rural and urban areas, indicating that city dwellers without specific training are more likely to be unemployed compared to their rural counterparts.
Individuals with technical training have greater job prospects in urban areas The following figure illustrates the employment rates compared to the labor force from 1997 to 2009.
Figure 3.4: Workforce vs employed workers (thousand of person) c 30,000
The MOLISA labor and employment surveys provide valuable insights into the current state of the job market, highlighting trends and challenges faced by graduates These surveys are essential for understanding employment dynamics and can be accessed for detailed analysis.
Unemployment is not a significant issue for the Vietnamese economy; rather, the main challenges lie in the types of jobs available and the skills of the workforce.
THE MODEL
Before 1992, empirical studies on intergenerational income correlations faced biases due to OLS estimation methods Solon's 1992 research marked a significant advancement in understanding the link between family background and employment earnings through OLS Previous studies often relied on sibling or father-son correlations to assess the impact of family background on economic status, but they were flawed by their inability to distinguish between permanent and transitory variations, neglecting life-cycle stages, and using overly homogeneous samples According to Solon, Bielby and Hauser (1977) highlighted that measuring parental income based on sons' recollections introduced severe errors-in-variables bias Although Bielby and Hauser attempted a minor adjustment for response error, it was based on the unrealistic assumption of zero correlation between response errors at different times.
Behrman and Taubman (1985) analyzed a sample of white male twins born between 1917 and 1927, encountering downward bias due to two significant exceptions The primary issue was the absence of direct measures of long-term status, as many studies focused solely on single-year earnings or income.
Samples in this study were not selected randomly While some research from this period employs Instrumental Variable (IV) estimation, Solon (1992) and Card (1998) demonstrated that using family background variables as instruments can introduce additional bias to Ordinary Least Squares (OLS) results, even if these variables do not independently affect earnings.
Family background variables may be correlated with unobserved factors, impacting various outcomes.
- - - - - - ability If one uses the family background variables solely as instruments, the bias is worse than OLS
John Bound et al (1993) emphasize the importance of carefully selecting instruments, as a weak correlation between instruments and wages can lead to significant bias in Instrument Variable (IV) estimates, even in large samples (Harm, 1995) To mitigate estimation bias, Card recommends controlling for family background variables, noting that Ordinary Least Squares (OLS) estimates of education returns are less biased when these variables are included.
In this study, I aim to analyze the impact of education and family background on employment wages using Ordinary Least Squares (OLS) regression A key strength of my research is the utilization of cross-sectional data from two different years, which helps mitigate bias associated with single-year estimates The samples are drawn from the Vietnam Household Living Standards Survey (VHLSS), ensuring a sufficiently large and randomly selected dataset Additionally, I incorporate parental education and the number of siblings as control variables to account for background influences The model is outlined below.
A A A f3JXlf'eniS _ educatim+ ~sibilings+ f3sDcan _irdex+8 4 urlxTn+u; with ui is residual The dependent variable is natural logarithm of nominal hourly wage Betts
(200 1) used the log of nominal wages as the dependent variable in calculating determinants of wage
1 Age is one of characteristic belonging to demographic factors of a person
As individuals age, their wages tend to increase, with a notable rise of approximately 13.38% for each additional year of employment (Harmon, 1995; Betts, 2001).
Ethnicity is a crucial variable in studying employment behavior, as it falls under demographic characteristics In my research, ethnicity is treated as a dummy variable to examine whether Kinh workers receive higher wages compared to other workers.
3 Married: this is an important variable belongs to demographic characteristics (Dang et al 2005) A married worker will be got higher salary than others ( Betts, 200 1)
4 Educational attainment: employment's education directly affects skills of labour Empirical studies show that employment is skilled or higher education attainment will get higher wage (Hamilton et al 2000; Betts,
5 South east: residences live in center or big city often tendency to be paid higher wage (Dang et al, 20005; Harmon, 1995)
6 Parents' education: education of mother is important directly effect on children's knowledge as well as job selection decision at adulthood It results in positive relation to wage (Betts, 2001)
7 Siblings: number of siblings will negative affect employees' wages (Betts,
Research indicates that individuals with multiple siblings tend to seek employment and start working earlier than those with fewer siblings (Dang et al., 2005).
8 Urban: allocation residence also impacts differently on wage Betts, 2001 found that workers who live in central areas are paid higher wage
The Duncan index is a crucial variable representing family background, specifically the socioeconomic status of the job held by the head of the household It is scored according to the Duncan index from 1961 and the social occupation index from 1992 This variable is anticipated to have a positive correlation with wage increases (Betts, 2001).
Determinant Variable Definition Expected I sign
Age Age The age of the employee (over 15 + to under 60)
Ethnicity Ethnicity ã The ethnicity of employee
rvia~Ie